Chapter 1. A bird’s eye view of fiscal decentralisation1

This chapter offers a succinct overview of fiscal decentralisation in OECD countries and identifies common trends. To that end, it seeks to answer a few crucial questions. How does decentralisation evolve overall? Which countries have undertaken intergovernmental fiscal reforms, and which were the most common and important? What was the impact of the 2008 crisis on sub-central deficits and debt? How did sub-central power and responsibilities evolve in the aftermath of the crisis? Is there a “new normal” in intergovernmental fiscal relations and sub-national public finance, and what does it look like? To answer these questions, the chapter reviews the evolution of the main fiscal indicators, such as spending and revenue decentralisation, tax autonomy, the tax and spending composition of sub-central governments, the size and structure of the intergovernmental grant system, and deficit and debt developments at the sub-national level. Finally, the chapter looks beyond purely financial decentralisation indicators.



This chapter seeks to provide a condensed, nuanced account of fiscal decentralisation trends in OECD countries. To that end, it fully exploits the wealth of data collected by the OECD Fiscal Network.2 It assesses the level and evolution of current government spending, investment, and tax revenues. It also looks at intergovernmental grants together with deficits and debt at the sub-central government (SCG) level. While decentralisation ratios can yield a first impression of the role of SCGs, they sometimes also convey an oversimplified, or even distorted, picture of the true extent of sub-central autonomy. To paint a more fine-grained picture of how intergovernmental fiscal relations really work, the chapter also considers an additional set of institutional indicators that measure tax autonomy, spending power and the conditionality of intergovernmental grants. Finally, the chapter briefly introduces a set of non-financial decentralisation indicators that gauge the power and influence of SCGs in specific policy areas.

Sub-central spending and revenue shares

The most common measures of fiscal decentralisation are the consolidated shares of SCGs in general government spending and revenue (Box 1.1). The shares vary widely from country to country, but have evolved relatively little over the last two decades or so – even after the onset of the crisis in 2008. OECD-wide in 2014, SCGs accounted for around 33% of consolidated government expenditure on average, with values ranging from 7% in Greece to 69% in Canada. The sub-central share in total own revenue averaged 19%, with Greece and Canada again lying at opposite ends of the spectrum. Spending is clearly more decentralised than revenues, with intergovernmental grants accounting for a considerable proportion of sub-central spending (Figure 1.1). Although showing decentralisation as a proportion of GDP rather than of general government spending and revenue does not change the overall picture, it does move a few highly decentralised countries with a large public sector up the decentralisation scale. In the Nordic countries, for example, the municipal sector accounts for up to 20% of GDP. Constitutional provisions account only in part for differences in sub-central autonomy, as some federal countries are more fiscally centralised than a number of unitary ones.

Box 1.1. The consolidation of spending and tax revenue data

Spending and revenue data from various databases must be “consolidated”. Such consolidation is necessary in order to prevent certain categories from being accounted for twice or even three times, i.e. at the central level and at one or two tiers of sub-national government. Intergovernmental grants and tax sharing arrangements are the main reason why data on spending and revenue across governments must be consolidated.

Spending consolidation

Spending data from the OECD National Accounts must be consolidated in order to net out intergovernmental grants. Intergovernmental grants – spent at an upper tier of government and then spent again at a lower tier – make the sum of central, state or regional and local spending larger than effective overall government expenditure. In order to avoid double counting, the OECD Fiscal Decentralisation Database consolidates expenditure at a certain tier of government by calculating total spending minus that tier’s intergovernmental transfer expenditure. Consolidated total general government expenditure is defined as global total expenditure by general government plus total intergovernmental property expenditure. Property expenditure is added because it consists of payments by one tier of government for a service provided by another tier, which should not be consolidated. The same procedure applies to the calculation of sub-central spending shares of individual government functions – education, healthcare, etc.

Tax revenue consolidation

Double imputation is usually less of a problem on the tax side, because tax revenues can be unambiguously allocated to a single government most of the time. This also holds true when taxes are shared across different tiers of government. Taxes collected by one tier of government are assigned to the tier on behalf of which it collects the taxes. A number of OECD countries – all federal countries plus Spain – distinguish between state and local government tax revenues, and in some figures a distinction is being made between the two levels of government. Social security contributions are merged with central government tax revenues, for simplicity and to allow for comparisons across countries of which some do not have institutionally separate social security systems.

Debt consolidation

In a few countries, the debt of one tier of government may be held by another tier and, thereby, count as an asset. In order to prevent some public debt from being counted twice, the government debt held by another tier of government should be netted out. The OECD National Accounts Statistics Database provides consolidated debt data for most OECD countries.

Source: OECD Fiscal Decentralisation Database,

Figure 1.1. The fiscal power of sub-central governments varies widely from country to country
Decentralisation ratios, 2014 or latest available year

Note: Sub-national expenditures include intergovernmental grants, while sub-national revenues do not. Latest available data for Korea are from 2012 and for Mexico from 2013. Australia, Chile, Japan, New Zealand and Turkey are not included.

Source: OECD Fiscal Decentralisation Database,

The OECD area has grown more decentralised over the last two decades at least, although reforms that have profoundly changed the institutional set-up of fiscal federalism are rare and confined to a few countries only (Figure 1.2). Italy and Spain, for example, embarked on a secular decentralisation process, with Spain, once one of the most centralised OECD members, becoming one of the most decentralised in the space of 20 years. Several Eastern European countries, such as Estonia, Slovakia, Slovenia and Poland, also devolved much more financial power to sub-central jurisdictions, often linked to the EU enlargement process and devolution to regional authorities. A handful of countries defied the trend and re-centralised a number of core spending functions. A case in point is Norway which, in the mid-2000s, moved both fiscal and regulatory power in hospital care from the counties back to central government. Finally, a few countries took revenue and spending decentralisation in opposite directions. Denmark in particular saw vertical fiscal imbalance – the difference between own revenue and own spending – increase by 14 percentage points, in particular after the 2007 reforms.

Figure 1.2. Many countries devolved more fiscal power, but a few re-centralised
Decentralisation ratios, change 1995-2014

Note: Data without Ireland. Sub-national expenditures include intergovernmental grants, while sub-national revenues do not. Data for Greece cover the years 2006-14; for Iceland 1998-2014; for Mexico 2003-13; for Poland 2005-14.

Source: OECD Fiscal Decentralisation Database,

While both revenue and spending have become increasingly decentralised over the past 20 years, the pace of spending decentralisation has clearly outpaced revenue decentralisation, resulting in wider vertical fiscal imbalances and larger intergovernmental transfer systems (Figure 1.3).3 Although the immediate aftermath of the crisis saw a considerable increase in devolved spending, as stimulus programmes targeted sub-central expenditure, the trend reverted in subsequent years. Sub-central revenues evolved more quietly both during and after the crisis, which suggests that they were less cyclical and/or less affected by policy changes. Tax decentralisation varies more sharply between countries than spending decentralisation which shows a lower coefficient of variation. Overall, decentralisation arrangements are edging towards an intermediate level, with a few highly decentralised countries re-centralising and several strongly centralised countries devolving more power to lower government levels, as shown by the slightly downward-sloping lines representing the annual coefficient of variation.

Figure 1.3. Decentralisation has increased and converged slightly across countries

Note: When there were no data for 2013, 2012 data were used. Sub-national expenditures include intergovernmental grants, while sub‐national revenues do not.

Source: OECD Revenue Statistics Database,

Simple decentralisation ratios say little about a SCG’s power, autonomy or discretion over fiscal resources, and the way intergovernmental fiscal und budgetary institutions function. On the revenue side, national limits on tax bases, rates and relief reduce the extent to which SCGs can sway tax revenues. On the expenditure side, SCGs may be strongly constrained by central or federal government regulation, so losing discretionary power over various policy areas. In some countries, the transfer of financial responsibility for education, healthcare and/or social welfare was hardly more than a change in accounting procedures, while essential regulatory power over service delivery remained with or was even tightened at the central government level. Very few fiscal measures adopted to tackle sub-central crisis symptoms changed the institutional set-up of fiscal relations. Taken at face value, sub-national revenue and spending shares might be misleading in some instances. They nevertheless provide a starting point for further examination of intergovernmental fiscal relations and sub-national public finance (see Chapter 6 on spending power).

Policy areas under sub-central responsibility

The various policy areas on which governments spend money are decentralised to different degrees (Figure 1.4). Housing and community services – a typical sub-central service – are the most devolved, with SCGs accounting for more than 70% of spending, followed by the environment, and education, then leisure, culture and religion. Healthcare appears less centralised, although the average hides sharp differences between countries. The composition of sub-central responsibility varies with the degree of overall decentralisation in a country (Figure 1.5). In highly centralised countries, the bulk of sub-central spending is confined to general economic and public services – chiefly in housing, local development, and primary and secondary education. In more decentralised countries, spending structure and policy responsibility are different, with healthcare and social welfare accounting for a larger share of sub-central spending. Education remains a core local government responsibility.

Figure 1.4. Some policy areas are more decentralised than others

Note: For Germany, unconsolidated expenditure was used. 2011 data were used for Austria, Korea, Israel and the United States.

Source: OECD National Accounts,

Figure 1.5. The sub-central spending composition changes with the degree of decentralisation
2012 or latest available year

Note: For Korea, Israel, Austria and the United States, 2011 data were used. The four categories are formed so as to contain a similar number of countries.

Source: OECD National Accounts Database,

A number of countries have made sizeable changes to the assignment of functions since 2000. They have decentralised education and SCG spending now makes up more than 60% of all education expenditure. Health care responsibilities, by contrast, have been re‐assigned to central government, although there have also been reforms that go in the opposite direction. The most salient shift took place in economic affairs – mostly transport –, where SCGs have gained considerably more influence. Again it should be stressed that spending and changes in the spending shares of certain functions do not necessarily imply a change in effective responsibility. Several countries have reassigned government functions to sub-national jurisdictions, while introducing an array of additional regulations that have kept effective SCG spending under tight supervision. Assessing true spending power requires a more refined set of indicators, as shown in Chapter 6, “The Spending Power of Sub-central Governments”.

Investment across government levels

Investment, or capital spending, is a major expenditure category at the sub-national level. SCGs account for almost two-thirds of capital spending, although the percentage varies widely across countries – from more than 90% in Belgium to barely 10% in Chile (Figure 1.6). The share of sub-national spending is greater in federal than in unitary countries although, again, that high share tends to overstate sub-national responsibility for investment since central government grants often provide the resources for sub-national government capital spending. Given the importance of investment for long-term productivity growth, measures that foster SCGs’ capital spending have become the focus of economic and fiscal policy (OECD, 2013 and OECD, 2015). In 2014 the OECD endorsed a Recommendation on effective public investment across levels of government.

Figure 1.6. The sub-central government level accounts for most public investment
Ratio of sub-central to general government investment, 2013

Source: OECD (2015), Sub-national governments in the OECD: Key data (

Public capital expenditure as a share of all government expenditure has declined over the last decade – particularly in the aftermath of the economic and fiscal crisis which saw countries use public investment as a key adjustment variable during consolidation episodes. However, SCGs’ share of that expenditure has risen slightly due to factors that include: declines in central government capital spending, transfers of responsibility for investment to lower tiers of government, and increases in the cost of SCG investment relative to central government investment projects. The sharp decline in overall investment when stimulus packages came to an end around 2010 affected sub-national jurisdictions, albeit in different ways – the share of sub-national spending rose further in federal countries, but tended to fall in unitary countries. While measurement issues make precise cross-country comparisons difficult, empirical evidence suggests that decentralisation could boost investment (Blöchliger et al., 2012; Kappeler et al., 2013).

Sub-central revenue composition

The sub-central revenue mix: Taxes, grants and user fees

Sub-central government revenues consist essentially of three resources: taxes (whether their own or those shared with other tiers of government), intergovernmental grants, and user fees, i.e. income from the provision of public services. In a few countries, property income (revenues from assets such as royalties and rents) are also important. The sub-central revenue mix is thus quite different from that of central government which depends mainly on taxes.

Although sub-central revenue mixes vary widely across OECD countries, they have changed little over time apart from in a few Nordic countries (Figure 1.7). OECD-wide, 42% of SCG revenue is covered by own and shared taxes, 44% by intergovernmental grants, and 14% by user fees. From country to country, however, the relative shares of the revenue sources differ considerably. While taxation accounts for almost 90% of SGC revenue in Iceland, it yields only 10% in the Netherlands, where the reliance on fees and intergovernmental grants is correspondingly higher. Federal countries allocate a slightly higher tax share to SCGs than unitary countries on average, and SCGs with a lower tax burden tend to charge more user fees. The make-up of SCG revenue has remained largely stable, with a slight increase in sub-central transfer dependency, more reliance on user fees, and less taxes.

Figure 1.7. The sub-central government revenue mix varies widely, and transfers now play a greater role

Note: 2006 instead of 2005 for Greece. Australia, Chile, Japan, Korea, New Zealand and Turkey are not included because one or more of the relevant data points are not available.

Source: OECD Fiscal Decentralisation Database,

The sub-central tax mix: More than just property taxes

The mix of sub-central taxes, i.e. those assigned to sub-central governments, depends largely on how decentralised a country is. The tax that SCGs in strongly centralised countries administer most widely is the property tax and, more specifically, the recurrent tax on immovable property (or real estate tax) which is almost exclusively sub-central. The tax mix evolves with decentralisation, mostly because immovable property taxes are difficult to raise as spending responsibilities increase. Thus, the greater expenditure decentralisation is, the more SCGs rely on income and, to a lesser extent, consumption taxes. The sub-central tax mix has changed slightly over the last 20 years, with a decline in property taxes from 33% to 31% of SCG tax revenue and a rise in the share of consumption and income taxes – especially personal income tax (Figure 1.8).

Figure 1.8. The sub-central tax mix is more varied at higher levels of decentralisation
Sub-central tax composition by degree of tax decentralisation, 2013

Note: 2012 data for Australia, the Netherlands, Mexico and Poland.

Source: OECD Revenue Statistics Database,

Various factors can explain the relatively small share of revenue from property taxation especially in the more decentralised countries. To begin with, the tax base, i.e. property and cadastral values, has not been updated for years – even decades – and a variety of measures such as tax caps, abatements and exemptions are gnawing at local property tax revenue. Moreover, the dwindling significance of manufacturing with large physical plants may also explain weak property tax revenues. Higher property taxes could bolster SCGs’ public finances with relatively few negative economic side effects (see Chapter 3, “Reforming the Tax on Immovable Property”). Indeed the share of revenue from property taxation has been on the rise since 2009. So, too, though, has revenue from personal income tax, which has become more widespread at the sub-central level, with several countries having assigned a share of tax revenue to sub-national jurisdictions. Finally, changes in the share of sub-central consumption taxes are mostly a consequence of more tax-sharing arrangements.

Tax autonomy

The term “tax autonomy” captures various aspects of the freedom that SCGs have over their own taxes, such as the right to introduce or abolish taxes, set tax rates, define the tax base, and grant allowances or relief to individuals and firms. In many countries, taxes are not assigned to one specific government level but shared between central and sub-central governments. Although such tax-sharing arrangements prevent single SCGs from controlling tax rates and bases, SCGs may collectively negotiate sharing formulae with central government.

In an attempt to measure the wealth of explicit and implicit institutional tax arrangements, the OECD Fiscal Network drew up a taxonomy of degrees of SCG taxing power. They are rated from highest to lowest in five main categories (plus an auxiliary level) and a number of sub-categories (Table 1.1):

  • in “a”, SCGs enjoy full power over the rates and bases of their own taxes;

  • SCGs in category “b” may set tax rates only – essentially surcharges and “piggy-back” taxes;

  • “c” denotes power over the tax base only – i.e. the right to grant businesses and households tax credits and relief;

  • “d” refers to different types of tax-sharing arrangements;

  • “e” means no taxing power at all;

  • category “f” denotes non-allocable taxes.

Table 1.1. Degrees of sub-central taxing powers
Categories and indicators


The recipient SCG can set the tax base, tax rates and any reliefs without needing to consult a higher level government.


The recipient SCG can set the tax base, tax rates and any reliefs after consulting a higher level government.


The recipient SCG can set the tax rate, and a higher level government does not set upper or lower limits on the rate chosen.


The recipient SCG can set the tax rate, and a higher level government does set upper and/or lower limits on the rate chosen.


The recipient SCG can set some tax reliefs (tax allowances and/or tax credits) but not tax rates.


There is a tax-sharing arrangement in which the SCGs determine the revenue split.


There is a tax-sharing arrangement in which the revenue split can be changed only with the consent of SCGs.


There is a tax-sharing arrangement in which the revenue split can be changed unilaterally by a higher level government, but less frequently than once a year.


There is a tax-sharing arrangement in which the revenue split is determined annually by a higher level government.


Other cases in which the central government sets the rate and base of the SCG tax.


None of the above categories a, b, c, d or e applies.

Source: OECD Fiscal Decentralisation Database,

In order to better capture complex institutional details the OECD Fiscal Network further divided the five categories into sub-categories: two for categories “a” and “b”, and three for “c”. It gave special attention to tax-sharing arrangements, where the four “d” sub‐categories denote the various institutional arrangements for determining jurisdictions’ and governments’ shares of total tax takes. Altogether 13 categories were established to capture the various tax autonomy arrangements in OECD countries. And, since category “f”, or “non‐allocable taxes”, was hardly used, they seem to reflect the tax-power picture accurately.

Although tax autonomy varies widely from country, most SCGs have some discretionary power over their own taxes (Figure 1.9). On average, the share of tax revenue over which SCGs – i.e. regional and local government – have full or partial discretion (Categories A, B and C) amounts to more than 70%. However, the degree of tax autonomy varies between the two tiers of sub-central government: while regions often enjoy the high tax powers of Category A, local governments are allowed to levy supplements on selected regional or central taxes only (Category B). From 1995 to 2011, tax autonomy increased, particularly in Category A, at the expense of tax-sharing systems. During that period, the overall share of total tax revenue over which SCGs have full or partial policy discretion rose from 9% to more than 12% at regional level and from 7% to 8% at the lower-tier local level.

Figure 1.9. Sub-central taxing powers vary, but are generally on the increase

Note: The letters “a” to “f” refer to the tax autonomy categories described in Table 1.1. The figure for the state level in panel B contains federal countries only. Tax autonomy of local governments in the United States varies across the states and is not assessed.

Source: OECD Fiscal Decentralisation Database and OECD Revenue Statistics,

Intergovernmental transfers

Intergovernmental transfers – or grants – plug the gap between sub-central spending and sub-central own revenue, often referred to as the “vertical fiscal imbalance”. OECD-wide, grants have grown in size – from 11% to 12% as a share of general government spending and from 6% to 7% as a share of GDP between 2000 and 2010. Apart from offsetting the expenditure-revenue imbalance, transfers serve to equalise interregional income and fiscal disparities and steer SCGs towards spending for certain purposes.

In order to reflect the variety of grant systems and the incentives they trigger, the OECD Fiscal Network established a taxonomy of intergovernmental grants in 2003. The main dividing line runs between earmarked and non-earmarked grants. As for the categories “matching” and “non-matching”, they denote how grants are connected to sub-central spending (Figure 1.10). A third distinction assesses the conditionality of intergovernmental grants. Overall, in 2010 upper-level governments earmarked around half of all transfers, while the other half was disbursed with no conditions attached. Shares of each category vary widely across countries but have changed little over time, with a slight trend from earmarked towards non-earmarked grants.

Figure 1.10. The composition of intergovernmental grants
Level 2000 and evolution 2000-10

Note: The first line in each cell shows averages of upper-tier SCGs (regions or states) for federal OECD countries and the second line local government averages for all OECD countries.

Source: OECD Fiscal Decentralisation Database,

The size and make-up of grant systems are useful indicators for gauging fiscal relations, such as effective sub-central autonomy, degrees of overlap in public service delivery, and spending pressure at central and sub-central levels. Yet the above classification might not fully reflect true SCG autonomy since the grant system design has evolved in recent years.

  • First, grants and regulation are replacing rather than complementing each other. Several countries have reformed their grant system and abolished earmarked grants, while tightening their regulatory frameworks. Denmark, for example, abolished conditional school grants for paying teachers and refurbishing school buildings and introduced a general purpose grant. At the same time, it increased central regulatory control over the municipal education authorities (Blom-Hansen, 2013). The same shift from tightly earmarked towards more general grants was observed in the Netherlands, which compensated with a stronger central-over-local regulatory framework (Boerboom and Huigsloot, 2010).

  • Second, the emerging category of performance-based grants is difficult to assess using the OECD Fiscal Network’s taxonomy. The reason is that results-based grants are thought to match outcomes rather than spending, while still allowing for budget flexibility. Consequently, they fit none of the taxonomy’s categories (Boadway and Shah, 2009).

Sub-central deficits and debt

Deficits are largely back to normal

Sub-central public finances have been affected by three large troughs in the last 20 years. The first occurred around 1995, the second shortly after the turn of the millennium, and the third in the wake of the economic and financial crisis in 2008. Developments at the SCG level largely followed general government patterns, although sub-central fiscal policy is much less volatile. Generally speaking, with the exception of a few sub-central governments in the OECD area that grapple with severe fiscal problems, deficit and debt sustainability is a central rather than a sub-central policy issue.

The last economic and financial crisis shook both central and sub-central budget positions. Yet sub-central government deficits in virtually all OECD countries have now largely recovered despite a small trough in 2013 (Figure 1.11). OECD-wide, general government deficits grew from around 1% of GDP in 2007 to 8% in 2009, before sliding back to 4% in 2014. SCG balance sheets were close to balance in 2007 before slumping to a deficit of around 2%, then recovering to less than 1% of GDP in 2014.The 2008 crisis hit most SCGs with deficits of a size that those at local level had not experienced since the beginning of the 1980s and higher-tier jurisdiction since the 1990s. SCGs that had larger deficits before the crisis also experienced greater deficit increases during the crisis. Those in federal countries were more exposed than in unitary ones, where they were shielded by intergovernmental transfers that compensated for lower own revenues (Foremny and von Hagen, 2012). The way in which deficits and debt evolve thus seems to be related to the institutional environment in which fiscal policy decisions are taken (see Chapter 2).

Figure 1.11. Sub-central government deficits in 2014 were nearly back to earlier levels

Source: OECD National Accounts Database,

Sub-central spending is falling, while revenues remain stable

Spending and revenues evolved differently over the cycle, especially after the onset of the 2008 crisis. After being inconspicuously flat in the decade prior to the 2008 crisis, sub-central spending as a share of GDP rose considerably after 2007, reflecting both outright GDP declines in many countries and the effect of stimulus programmes (Figure 1.12). When consolidation started in 2010 it fell back again. Sub-central tax revenues slowly increased during the decade to 2008 and have remained almost stable since then, suggesting that SCG own-source taxes – often property taxes – are less responsive to the cycle than central governments’ (see Chapter 3, “Reforming the tax on immovable property”). Intergovernmental transfers – a significant revenue source for SCGs in many countries – behaved unobtrusively (and are not, therefore, shown, in Figure 1.12). Overall, the return to lower deficits has so far been achieved mainly through SCG spending cuts, while own revenues and intergovernmental transfers have remained broadly the same. Again, the overall picture hides sharp differences between countries – as shown by the deciles – and even between SCGs in the same country.

Figure 1.12. Spending has fallen since consolidation started in 2010, while tax revenues have remained stable

Source: OECD National Accounts Database,

Debt has still not entirely stabilised

After a relatively long period of stable or declining debt-to-GDP ratios, sub-central debt jumped to new record levels, following the surge in crisis-related deficits post-2008. It is now stabilising, but – as in earlier crisis periods – at higher levels than before the downturn, and has recently risen again (Figure 1.13). The liabilities of state and regional governments amounted to around 18% of GDP on average in 2014, with the highest-tier SCGs in some countries – particularly Canada, Germany and Spain – showing considerably higher debt ratios. As for local-level jurisdictions, their liabilities stood at around 8% of GDP in 2014, with the debt ratios of some – especially in Japan – as high as 35% of GDP. Debt and the need to manage it well is likely to be the “new normal” for many sub-central governments in coming years.

Figure 1.13. After a rapid rise during the crisis, sub-central debt is stabilising

Note: Local-level jurisdictions do not include those in Australia, Chile, Mexico, New Zealand and the United States.

Source: OECD National Accounts Database,

Debt creates externalities across all tiers of government, since sustainability is determined by their joint action. Moreover, national averages may conceal the unsustainable positions of individual jurisdictions, and the financial difficulties of a few may quickly raise concerns about the fiscal health of all. SCGs’ difficulties can also create financial problems for central governments, particularly where markets factor an expected bail-out to the cost of sovereign debt (Vammalle and Hulbert, 2013). SCGs experience growing difficulty in securing access to credit as a consequence of increased borrowing costs, a contraction in the volume of credit provided by financial intermediaries in high‐debt countries), and a reduction in foreign investment inflows (OECD, 2015). Keeping sub-central debt at prudent levels and managing it efficiently has become a core fiscal policy objective for all government levels (see Chapter 5 which addresses the monitoring of sub-central debt).

Beyond fiscal decentralisation

While this chapter approaches decentralisation from a fiscal and financial point of view, intergovernmental frameworks may also be seen from other angles. Decentralisation is a concept where the multiple facets of sub-central autonomy share a common denominator: the notion of dividing and sharing power, resources and authority between different tiers of government. Essentially what is transferred is either fiscal – e.g. the power to tax and spend – or non-fiscal, such as constitutional set-up or regulatory powers. Power may also be extended to autonomous agencies or directly to the public and private providers of public services, such as schools, hospitals and transport companies. The use of other indicators in addition to financial flows can thus usefully complement assessments of true decentralisation and power sharing.

There are a number of indicators to measure sub-national institutional and regulatory architecture. In 2014 the OECD Fiscal Network developed an indicator of constitutional decentralisation to measure how fiscal constitutions are related to economic and fiscal outcomes (see Chapter 2, “Fiscal Constitutions”). Drawing on the Programme of International Student Assessment (PISA), the OECD Directorate for Education and Skills assesses decentralisation in primary and secondary education by measuring the extent to which educational functions are assigned to lower government levels and schools (OECD, 2014). Linked to education performance, the education decentralisation indicator delivers more pertinent results than the traditional fiscal decentralisation indicators (Fredriksen, 2013). Finally, the Regional Authority Index measures the authority of regional governments in a large number of countries through ten dimensions (Hooghe, Marks and Schakel, 2010). In an empirical analysis linking decentralisation to levels of debt and how they have evolved in the regions of six OECD countries, Ahrend, Curto-Grau and Vammalle (2013) found that the Regional Authority Index yielded fuller understanding of regional debt levels than traditional decentralisation variables. To sum up, then, institutional indicators are a useful complement to policy analysis, often supplying information not contained in fiscal data.

This chapter has sought to show that fiscal decentralisation and sub-central fiscal autonomy is a multidimensional concept that cannot be easily summarized in a single indicator. Indicators include the sub-central revenue or spending share, the size and composition of intergovernmental grants, the composition of sub-central spending and revenues, levels of sub-central deficits and debt. Institutional indicators – like tax autonomy, spending power and intergovernmental grant conditionality – are also useful. Taken together, all these different indicators paint a full picture of the fiscal and institutional architecture of a country, the functioning of the different government levels and how they work together, and the policy issues that may be at stake.


Ahrend, R., M. Curto-Grau and C. Vammalle (2013), “Passing the Buck? Central and Sub-national Governments in Times of Fiscal Stress”, OECD Regional Development Working Papers, No. 2013/05, OECD Publishing,

Aldasoro, I. and X. Seiferling, (2014), “Vertical Fiscal Imbalances and the Accumulation of Government Debt”, IMF Working Paper, No. 14/209, 20 November.

Blom-Hansen, J. (2013), “The Fiscal Federalism View of Grants: Some Reflections from Political Science”, in J. Kim, J. Lotz and N. Mau (eds.),General Grants versus Earmarked Grants: Theory and Practice, Korea Institute of Public Finance and the Danish Ministry of Interior and Health.

Boadway, R. and A. Shah (2009), Fiscal Federalism: Principles and Practice of Multiorder Governance, Cambridge University Press.

Boerboom, H. and P. Huigsloot (2010), “Earmarked General Grants and General Earmarked Grants in the Netherlands”, in Kim, J., J. Lotz and N. Mau (eds.), General Grants versus Earmarked Grants: Theory and Practice, Korea Institute of Public Finance and the Danish Ministry of Interior and Health.

Blöchliger, H., B. Egert and K. Bonesmo Fredriksen (2013), “Fiscal Federalism and its Impact on Economic Activity, Public Investment and the Performance of Educational Systems”, OECD Economics Department Working Papers, No. 1051, OECD Publishing, Paris,

Foremny, D. and J. von Hagen (2012), “Fiscal Federalism in Times of Crisis”, CEPR Discussion Paper, No. DP9154.

Fredriksen, K. (2013), “Decentralisation and Economic Growth – Part 3: Decentralisation, Infrastructure Investment and Educational Performance”, OECD Working Papers on Fiscal Federalism, No. 16, OECD Publishing.

Hooghe, L., G. Marks and A. Schakel (2010), The Rise of Regional Authority: A Comparative Study of 42 Democracies, Routledge, London.

Kappeler, A. et al. (2013), “Does Fiscal Decentralization Foster Regional Investment in Productive Infrastructure?”, European Journal of Political Economy, No. 31.

OECD (2015), OECD Economic Outlook, Volume 2015, Issue 1, OECD Publishing, Paris,

OECD (2013), Investing Together: Working Effectively across Levels of Government, OECD Publishing, Paris,

Vammalle, C. and C. Hulbert (2013), “Sub-national Finances and Fiscal Consolidation: Walking on Thin Ice”, OECD Regional Development Working Papers, No. 2013/02, OECD Publishing,


← 1. The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms of international law.

← 2. The OECD Fiscal Network is short for the OECD Network on Fiscal Relations across Levels of Government, created in 2004. For more on the OECD Fiscal Network, go to

← 3. There is no consensus regarding the accurate measurement of vertical fiscal imbalance. In general it is equated with transfer dependency, i.e. transfers received by sub-national governments as a share of their total revenue or expenditure. However, this measure neglects sub-national borrowing as a formof financial expenditure. Many sub-central governments – subject to fiscal rules restraining them – indeed borrow or sell assets to cover current or capital spending. Sub-national government borrowing can be important in accounting for the dynamics of vertical fiscal imbalances (Aldasoro and Seiferling, 2014).