Printer Friendly

Chapter 3: improving fiscal federalism.

The current system of fiscal federalism is creating imbalances between the federal and the sub-federal governments (vertical imbalance), and between sub-federal governments (horizontal imbalance). Without reform, the vertical imbalance will widen as the fiscal burden from the ageing of the population fails mainly on the federal level. Reform should therefore strengthen the fiscal capacity of the federal government by improving its revenue sources and by shifting some spending obligations to sub--federal governments. The imbalance between regions arises because of the lack of coherence between taxation and spending. Shared revenues from the personal income tax are allocated to the region of residence, while the region of the workplace does not benefit, which particularly affects Brussels' revenue level. This imbalance could be eliminated by allocating more of the shared personal income tax to the region of the workplace. Furthermore, the system of equalisation grants should be re-designed to provide incentives to the recipient regions to develop their own revenue base. The performance of the fiscal system could further be improved by raising the efficiency of spending in areas of national interest which have been assigned to sub-federal governments or where there are overlapping responsibilities, such as in employment, R&D, training, education, energy and environmental policies.

**********

For about 15 years, Belgium has achieved an impressive degree of fiscal consolidation with a significant reduction of the debt-to-GDP ratio. Fiscal consolidation has been achieved against the background of the transition from a unitary state to a federal state, which in practice was mostly implemented in 1988-89 before being constitutionally established in 1993 (see Annex 3.A1). In line with annual agreements between all levels of government, the federal government significantly reduced its budget deficit (aided by falling interest payments) while sub-federal governments roughly balanced their budgets. On the other hand, the sharing of tax revenues is such that it has contributed to the nonfederal governments ending up with the fastest growing tax revenues, which means that these levels of government as a whole have been faced with limited budgetary pressure to rein in spending.

Overall, it appears that the devolution of responsibilities to regions and communities has helped to better tailor public goods and services to the preferences of the people; it also introduced some benchmarking between jurisdictions, which increased the efficiency of the public sector. However, devolving spending and revenues to regions and language communities appears to have been, to some extent, at the cost of the federal budget. Furthermore, there is a risk of excessive institutional complexity, fragmentation of policies and diseconomies of small scale in the provision of public services. A recent cross-country econometric study of 21 OECD countries, including Belgium, over the period 1970 to 2000, finds that generally fiscal decentralisation has improved public sector efficiency, although in Belgium the improvement was rather small despite significant decentralisation during the 1990s (Adam et al., 2008).

Reforming fiscal federalism will also have to consider political economy issues, as attempts to rationalise were frequently blocked by some parties fearing they may lose too much or not gain enough. Perceptions of net gains/losses can actually be misleading. For example, the income gap between the two main regions has continued to widen since 1965, while Brussels-Capital has achieved the highest income per capita (Table 3.1). However, a closer look reveals that income gaps not only exist between the northern and southern regions but also within each regions/communities. In international comparison, income gaps in Belgium are not particularly large and Belgium still belongs to the group of highly egalitarian countries in the OECD (OECD, 2008a); an important reason for this relatively narrow income distribution are the implicit transfers of the fiscal system to individuals, independent of their regional location.

Overall, fiscal federalism in Belgium has to cope with two main challenges. The first is to ensure fair burden-sharing of public spending between the different levels of government, including the upcoming ageing costs. The second is to increase efficiency of spending at all levels of government by reducing economic disincentives in the transfer system and by improving policy co-ordination and co-operation in public service provision. The following sections address these challenges.

Despite decentralisation, the bulk of the deficit reduction has been achieved by the federal budget

A main driving force for fiscal consolidation in the 1990s was the convergence plan to fulfil the Maastricht criteria of EMU. As shown in Table 3.2, both the federal government and the sub-federal entities have contributed to the improvement in the general fiscal deficit. However, the federal government has achieved a substantially bigger primary surplus than sub-federal government entities. The primary surplus was improved by reducing the ratio of primary spending to GDP, while this ratio has increased for regional entities. Only part of this evolution relates to devolution of government expenditures to regions. This was the case in the early 1990's, when more spending responsibilities were allocated to the regions than shared tax revenue, which led (during a brief transitional period) to a so-called "natural deficit" of the regions (Stienlet, 2000; Gerard, 2002). More recently, it is the more dynamic revenue growth of the regions that explains the bulk of the increase in primary spending. The bigger effort of the federal government in controlling expenditures related to a decentralisation process that did only require federal entities to balance their budget while granting them more dynamic revenues than the revenues of the federal budget.

The burden-sharing with respect to fiscal consolidation assigned to regions has been limited since regions only committed to reach a balanced budget from a relatively limited initial deficit. In the Co-operation Agreement (Accord de Cooperation) of 1996, regions and communities promised to achieve balanced budgets. This commitment did not require a major effort since the fiscal deficit of regions as a whole was already roughly balanced, although larger efforts were required by some of the sub-federal levels governments as compared with others. Also, tax revenues raised by the regions, especially Flanders, have been dynamic, which limited the necessary efforts in controlling expenditure. Other measures to foster the role of sub-federal governments in fiscal consolidation had only a limited impact on expenditure growth. The federal government can impose restrictions on sub-federal loan financing if regional debt financing interferes with macroeconomic goals or if there is a risk of structural slippage in the budgetary position of a non-federal government. The High Council of Finance (HCF) was strengthened with the objective to ensure budgetary discipline at all levels of government, but only as an advisory and coordinating body.

Conversely, the devolution process has left the federal government with the bulk of consolidation efforts. Following the special Finance Act (1989), the existing debt burden, and the related high debt service, has remained almost entirely with the federal government. The social security system also remained at the federal level, implying that the federal government is confronted with the responsibility of financing the expected increase in ageing related costs. The increase has such amplitude that it necessitates a considerable amount of prefunding as part of the overall strategy to finance the budgetary consequence of population ageing (see below). However, as described in Chapter 2, prefunding has repeatedly been delayed and will, according to the Stability programme, only start after 2015 and will require a strengthening of the federal government's fiscal position. The federal government was allowed to compensate for the regionalisation of some taxes and fees (such as the road fund tax and the registration fee on the transfer of real property) and the increase in contributions to the EU by cutting the amount of shared taxes to the regions (the so-called negative term). Although this arrangement was broadly revenue-neutral for Wallonia and the Brussels-capital region, Flanders benefited as its revenues from shared income tax increased more than the negative term, i.e. the amount by which shared taxes were reduced. As a result, and to ensure horizontal neutrality for Brussels and Wallonia, the federal government could not reduce the transfers (shared taxes) to the regions as much as would have been necessary to cover its fiscal costs from decentralisation (van der Stichele and Verdonck, 2002). Maintaining horizontal neutrality was de facto achieved by renouncing to vertical neutrality, with a higher burden on the federal government.

Spending pressure from demographic change will mainly affect the federal government budget

Due to population ageing, there will be significant pressure on future public spending. Public pensions are most affected, but public health care spending is also expected to increase, while education spending could decline. As a result, the federal level of government (social security and the federal budget) will bear more than 90% of the fiscal costs stemming from ageing, with the share financed by Regions and Communities will be comparatively small (Chapter 2).

The government decided to pre-fund an important part of the ageing cost by--as stipulated in the December 2006 update of the Stability Programme--generating budgetary surpluses from 2007 onwards, reaching about 1 1/2 per cent of GDP between 2013 and 2018, and then gradually reduce the surpluses to zero until 2030. However, as mentioned in Chapter 2, this strategy needs to be revisited in light of the slippages in recent years and the increased deficits arising from the economic crisis. This raises two issues. First, as the contribution of pre-funding to the cost of ageing will be smaller than expected, more emphasis needs to be put on structural reform of the pension system (Chapter 2). Second, to achieve a fair burden sharing of age-related costs, fiscal relations between the different levels of government need to change in favour of the federal budget. This could be done by reforming tax sharing arrangements or by re-allocating spending responsibilities. An obvious candidate for shifting spending from the federal to the regional level is the payment of civil servant pensions of regional and community administrations, which is currently a federal government responsibility. This would relieve the federal budget and reduce the incentives of the regions and communities to allow their civil servants to retire early. Such a change in spending responsibilities could also induce regions and communities to pre-fund part of these pension liabilities, thereby increasing their contribution to the pre-funding objectives.

Sub-federal governments spend almost 40% of total government expenditure

In 2006, Belgian sub-federal governments (Regions, Communities and local authorities) spent around 38% of total government expenditure, an increase of about 5 percentage points since the mid-1990s (Table 3.3). The basic principle in Belgium for assigning spending power is that devolved responsibilities, which can be organised on a territorial basis, are assigned to regions, while those related to the use of language are assigned to communities. Both the regions and the communities can also -within their specific areas of responsibility--cooperate internationally, which allows them to conclude treaties and agreements of cooperation with other countries and the EU. While responsibilities of Communities are defined by the Constitution (Articles 127-30), those of the regions are listed in Special Laws, which are adopted by both houses of parliament with a two-thirds majority.

The devolution of powers to regions has significantly reduced the power of the federal state; its main remaining responsibilities are defence and police, parts of economic policy, public debt financing, social security, state-owned enterprises, and parts of judiciary, as well as foreign relations in those areas (see Box 3.1). There has been continued demand from Flemish groups to decentralise parts of social security, notably unemployment insurance and health insurance. However, there are good economic reasons for keeping social insurances centralised such as risk sharing and the capacity for a homogenous use of sanctions. At the same time, decentralisation of some parts of social policy, such as labour market activation policies can--if properly designed--be efficient, as has been shown in the Netherlands (OECD, 2008b).
Box 3.1. Responsibilities of regions and communities

Communities are mainly responsible for education (except
determining the beginning and end of compulsory schooling, minimum
conditions governing granting of diplomas and the pension plan),
parts of public health (hospitals and prevention), social
assistance and cultural affairs, including radio and television
broadcasting. Regions are responsible in particular for economic
policy, employment policy, public transport (excluding the
state-owned railway company), zoning, environment, housing, water
policy (except distribution), agriculture and parts of energy
policy. Energy policy responsibilities are split between the
federal and regional governments (see Chapter 5).

Regions are principally responsible for designing and implementing
policies for energy efficiency, renewable, non-nuclear energy R&D
and market regulation for the distribution and supply of
electricity and gas through distribution networks. The federal
government is responsible for issues such as electricity and gas
tariffs, market regulation for large infrastructure for storage,
transport and transmission of energy, the nuclear fuel cycle and
R&D in both nuclear fusion and fission. In addition, the
municipalities have a legal monopoly on electricity distribution.
Nearly all municipalities have transferred the distribution of
electricity to inter-municipal companies, called "intercommunales",
which partially finance the local municipal governments (IEA,
2006). Regions also supervise local authorities (provinces and
municipalities). There are 589 municipalities, which are grouped
into five provinces in each of the Flemish and the Walloon region
and into the Brussels-Capital-Region, which includes 19
municipalities.


Despite increased taxing power of regions, their revenues continue to be dominated by transfers from the federal government ...

According to the Belgian constitution, regions can levy their own taxes on any activities provided that there is no federal tax on this activity, although in some cases the autonomy to set the tax base is restricted. Regions do set bases and rates for the following taxes and fees: the estate, inheritance and gift taxes; registration fees on real estate transfers; mortgage registration fees; tax on gambling and betting; taxes on the opening of drinking establishments; tax on automatic amusement devices; radio and television licence fees; road fund tax on automobiles; vehicle registration fees and Eurovignette. The regions have also the right to add to the federal personal income tax (but not on the corporate income tax) by levying a percentage of the amount which the taxpayer has to pay to the federal government, but this possibility is not used. Regions can also grant personal income tax reductions (currently an opportunity only used by the Flemish region). (1) The tax surcharges or reductions by the regions do not affect the transfers (shared taxes) from the federal government but affect only their own revenues. Municipalities levy a local property tax on homeowners, which is calculated on the imputed rental value attributed by the federal government to the property (Chapter 4). The tax paid varies according to the commune and generally lies between 20 and 50% of the deemed rental value. Regionalised tax policies can lead to tax competition. Some tax competition exists between Belgian regions, but its degree appears to be limited.

After the 2001 devolution of taxing power, regional taxes have gained in importance, although most of the tax revenues continue to accrue at the federal level. A good part of federal government tax revenues is, however, automatically transferred as "shared taxes" to the regions and communities. Since 2002, there has been some transfer of tax autonomy to sub-federal levels of government, but the degree of tax autonomy remains limited. The regions and communities combined have full taxing power over about a fifth of their total revenues (increasing to one third for the regions alone) and the main sources include tax on real estate, registration tax on cars, inheritance tax, and some environmental taxes. The regions also have some powers with respect to personal income taxes. For example, in the Flemish region, a tax rebate on earned income will reduce tax revenues by EUR 730 million in 2009 compared with total revenues of EUR 20.4 billion in 2007.

Two main taxes are shared: the personal income tax (usually referred to below as the "income tax") for the regions and the VAT for the communities. In 2007, about 45% of personal income tax revenues (PIT) and more than half of VAT revenues were transferred to the regions and communities. The share of transferred PIT increased during the 1990s, before declining after regions have received more own taxes (due to the so-called negative term as mentioned above). More recently, it has started to increase again. The share of transferred VAT has declined from 65% in the early 1990s to 50% since 2000 (Figure 3.1). Shared taxes continue to be the main revenue source of regions and communities (Table 3.4).

[FIGURE 3.1 OMITTED]

... which are not aligned neither with actual revenue collection nor spending needs

The tax sharing arrangement in Belgium differs, however, from that in other OECD countries where sub-federal governments receive a fixed share of actual tax revenues. In Belgium, the so-called "shared taxes" which are distributed to regions and communities are not directly dependent on actual government tax receipts. They are therefore, according to OECD principles, not treated as shared taxes but as transfers. The total annual amount of the "shared income tax" is derived from a lump sum amount, which has been defined in 1989 and is indexed mainly to the consumer price inflation and national real GDP.

The allocation of personal income tax between regions and communities is also independent on any measure of spending needs. The overall amount of the shared income tax to the regions is based on each regional income tax yield (according to the principle of "fair return"). This allocation favours Flanders, which has a relatively large personal income tax base. The part of the VAT, which is transferred to communities, is partly allocated according to the share of the population below the age of 18 and partly to the share of pupils. Hence, some link between the transfer received and communities' education spending needs is maintained. It has, however, been decided by law to gradually and partially change the distribution criteria away from needs towards taxing power (similar to the shared income tax), so that this indirect link to education spending will become weaker in the future. This will further disconnect shared-taxes to spending needs.

For Brussels, a problem arises from the fact that the shared revenue from the personal income tax is fully allocated to the region of the residence of the taxpayer while the region of the workplace does not benefit. As more than half of the Brussels workforce is living in the neighbouring Flanders and Wallonia, this arrangement reduces revenues of the Brussels-Capital region and increases revenues of Flanders and (to a lesser extent) Wallonia. The Brussels-Capital region receives a transfer from the federal government which is meant to cover the additional costs which arise from carrying out its function as an international city (BELIRIS programme) and it also receives equalisation transfers as its per capita revenue from personal income tax is below the national average (see below). However these transfers do not fully compensate Brussels for the additional costs and for the spill-over of its public services to the citizens who are living in the other regions; as a result, Brussels has a relatively weak revenue base despite the relatively high share of GDP which is produced in this region (see Box 3.2).

Table 3.5 shows that per capita expenditure and revenue are much larger in Flanders than in Wallonia and in the Brussels-Capital region. However, this is also due to the fact that the Flemish Community has merged with the Flemish Region while the budgets of Wallonia and the Brussels-Capital Region are limited to regional matters as the French Community provides the community services to the French-speaking citizens of these regions. If one allocates (approximately) the budget of the French Community to these two regions, the per capita spending and revenue in Flanders and Wallonia are quite similar, while per capita spending in the Brussels-Capital Region is about 25% higher (numbers in brackets). Considering that urban agglomerations tend to have higher spending per capita (due to higher unit costs and more spillovers of service benefits to neighbouring regions) this difference in per capita spending is rather small.

The problems of Brussels are in many ways similar to those in agglomerations in other OECD countries and in others unique to the region. It has been shown that the additional costs of the Brussels-Capital Region and the spillover benefits of its public services to citizens of the other regions, including through public transport, are substantial (van der Stichele, 2003). There are various ways to address this problem although the measures differ with respect to practicability, accuracy and distributional effects:

* First, enlarging the Brussels-Capital region by including the local communities where the majority of the commuters live, so that the region receives a larger part of the taxing power of its workers. Such amalgamation, although it happened in the past, is politically difficult today and may not be an effective measure today due to the distance that many commuters travel every day.

* Second, changing the allocation of the shared income tax so that the region where the firm of the worker is located receives a proper share ("workplace principle"). However, the tax allocation according to the location of the firm can pose problems if a firm has workplaces in different regions while wages are paid by a centralised wage accounting unit. This problem does not arise if the allocation between regions is based on regional employment or regional GDP shares, but then revenue from the progressivity of the income tax does not accrue to the region where the higher wage earners work, but is distributed between the regions according to the regional shares of employment or GDP.

* Third, granting a higher weight to the population of Brussels when calculating federal transfers. For example, in Germany, the city-states (Hamburg, Bremen and Berlin) get a special treatment in the horizontal equalisation system by assuming that their spending needs per capita are higher than in the other states; as a result, a correction factor ensure that city-states either receive higher transfers from the other states or pay lower transfers (Hamburg). Austria also provides higher weights to the populations of larger cities before calculating interregional transfers and Australia gives a higher weight to the population of the National Capital Territory.

* Fourth, compensating the costs of spillover effects by matching grants from the federal government. The federal government could, for example, increase the BELIRIS programme and reduce its transfers to the other regions accordingly. There are many examples in OECD countries where agglomerations receive special matching grants from provincial or federal governments (e.g. Canada and the United States).
Box 3.2. Fiscal problems of the Brussels-Capital Region

The Brussels-Capital Region (BCR) has some specific characteristics
as compared with the other two regions. Both spending and revenue
per capita are higher than in Flanders and Wallonia, which
internationally is also observed in other agglomerations with high
population density. The BCR is divided into 19 municipalities,
which are responsible for providing communal services. In addition,
there is also a special administrative district that largely
coincides with the BCR and which is responsible for
cross-municipality areas, such as disaster control, gun legislation
and visa delivery. Brussels is a bilingual city with the French and
Flemish Communities being responsible for education and culture
(and some other personal services) in the city. Infrastructure
projects, which are linked to the status of Brussels as a major
international city and urban agglomeration, are financed by a
special programme of the federal government (BELIRIS programme,
amounting to EUR 124 million in 2007). As personal income tax
receipts have fallen below the national average, the BCR also
receives from the federal government equalisation transfers (ISN).
Furthermore, the BCR is (partly) compensated for the revenue losses
due to the exemption of federal government buildings from the real
estate tax (so-called main-morte), but not for revenue losses due
to tax-exempt government buildings, which belong to the other
regions and the communities.

The main revenue sources of the BCR are regional taxes and fees
which amount to about half of total revenues. Taxes and fees on
holding and buying real estate, on inheritance, on gifts and on
road vehicles are the most important own revenues. The introduction
of road charge tolls for trucks (following the German example) is
also discussed but no agreement has been reached so far between the
three regions. Other main revenue sources of the BCR are the shared
personal income tax and the equalisation transfer ISN.

On the spending side, the largest share (about a quarter) is spent
on equipment and public transport (mostly for the "Societe des
Transports Intercommunaux de Bruxelles"/ "Maatschapij voor het
Intercommunaal Vervoer te Brussel"), followed by housing, transfers
to municipalities and administration.

While the fiscal position of the BCR has improved in recent years,
also helped by the strengthening of the economy, revenues remain
vulnerable to cyclical downturns including the housing sector.
Furthermore, revenues are generally considered as insufficient to
finance the regional and municipal spending needs of the city (de
Callatay, 2007). This is partly due to the fact that many people
who work in Brussels (more than half of Brussels workforce) are
living in Flanders and (to a lesser extent) in Wallonia. As the
revenues from the personal income tax are allocated according to
the residence of individuals and not to their workplace, Brussels"
share of personal income tax revenues is relatively low. As a
result, the revenue of Brussels from this source is below average.
Thus, the Brussels-Capital region, despite being the richest region
in terms of GDP per capita, has below-average personal income tax
revenue, qualifying for receiving equalisation grants from the
federal government. Furthermore, the many international civil
servants living in Brussels are exempt from personal income tax. In
2003, it was estimated that the additional cost of the BCR as the
capital of Belgium and international city amounted to around EUR
490 million while the compensatory transfers to cover these costs
have been much less (de Callatay, 2007 and van der Stichele, 2003).
The fact that Brussels must provide public infrastructure to many
people who do not contribute to its tax base, leads to a risk of
underinvestment. The 1 million people living in Brussels have to
finance infrastructure and public services for themselves and 1/2
million daily commuters. (van der Stichele, 2003).

Fiscal problems in the BCR may, however, also arise from
inefficiencies in spending. There seems to be room for improving
spending efficiency in many of the institutions, which provide
public services, such as job placements, hospitals, public housing
and economic development and also in the administration of the BCR
(de Callatay, 2007). Furthermore, responsibilities between the BCR
and the 19 municipalities in Brussels should be clarified, as there
are sometimes overlapping activities and problems of co-ordination.
Efficiency could, perhaps, also be improved by amalgamating
municipalities in Brussels or, more radically, by abolishing the
municipal level and the special administrative district altogether
and transferring municipal responsibilities and revenues to the
BCR.


Equalisation transfers penalise economic efforts

Regions with personal income tax yield per capita that are below the national average receive an equalisation transfer from the federal government, the National Solidarity Measure (NSM); its size is proportional to the percentage point gap in per capita personal income tax yield. Originally, only the Walloon Region received this transfer but over the past ten years, the Brussels-Capital region has also been a beneficiary.

The fiscal equalisation system in Belgium leads, as does any redistributive system, to a trade-off between equity and efficiency objectives. The reduction of the revenue gap of poorer regions by the additional transfer enables them to improve their infrastructure and provide better public services to their population, thus narrowing differences in living standards across the country and meeting equity objectives. At the same time, however, the fiscal equalisation system may provide disincentives to the poorer regions to develop their growth potential as an increase in their tax base reduces the transfers from the federal government, thus levying an implicit tax on economic efforts of the region.

Cattoir and Verdonck (1999) and Gerard (2002) found that this implicit tax is above 100% for Wallonia and Brussels, i.e. if Wallonia or Brussels increase their income tax base, their overall revenue declines as the increase in the transferred income tax from the federal is outweighed by the reduction in the equalisation transfer. Conversely, if economic performance (as defined by the contribution to the federal income tax yield) weakens in Wallonia and Brussels, the decline in the transfer from the shared income tax is more than offset by the increase in the equalisation transfer (Table 3.6). In a recent study, Algoed and Heremans (2008) confirm the large adverse effects of the transfer system on economic incentives for catching-up. They find that an autonomous increase in GDP in Wallonia reduces revenues of the Walloon government, as the loss of equalisation transfers is larger than the gain in other revenues. Brussels can, according to their calculations, expect a positive effect on its revenue from an increase in its GDP, but this effect is very small. On the other hand, as the communities and municipalities would see an increase in their share of the personal income taxes, there would be a mitigating effect when considering the geographical areas.

There are two reasons for the economic disincentives of the transfer system. First, the amount of equalisation for each percentage point of the tax gap is relatively large. Second, the federal government distributes only part of the additional income tax yield to the regions. If a region, which receives equalisation transfers, improves its tax base, the federal government has a double gain as it receives a higher tax yield and can also reduce equalisation transfers to this region. As for the region, the withdrawal of the equalisation grant outweighs the increase in transferred income tax, implying that the region has no overall revenue gain, and may experience a loss. By contrast, an autonomous increase in the income tax yield in Flanders increases revenues in Flanders, which does not receive an equalisation transfer and therefore is not affected by a transfer cut; higher tax revenues in Flanders also benefit Wallonia and Brussels because the increase in the tax gap raises equalisation transfers. To reduce the disincentive effects for Wallonia and Brussels to develop their tax base, the equalisation transfers should be made less progressive, for example by reducing the amount of equalisation for each percentage point of the tax gap.

How big are inter-regional transfers?

In Belgium, the political debate about the appropriate degree of decentralisation is influenced by a discussion about the size of transfers between the regions. As there are no direct transfers between regions (as for example in Germany) interregional transfers arise implicitly through the tax and social security system. Earlier studies found relatively continuous net transfers from the Flemish Region to the Walloon Region and sometimes also to the Brussels-Capital Region. These mainly reflected differences in the demographic structure and economic performance leading to differences in the ability to pay taxes and in the need of unemployment and other social benefits. However, the estimates of interregional transfers differ significantly between various studies. De Boeck and van Gompel (1998) estimate that the implicit transfer of Flanders to Wallonia amounted to EUR 4.6 billion in 1996 (2.2% of national GDP). This is double the estimate in the study by Cattoir and Docquier (1999), but slightly smaller than the estimate by Abafim (the Flemish authority for Finance and the Budget).

The National Bank of Belgium recently examined the size of interregional transfers (Dury et al., 2008), using net tax payments i.e. taxes including social security contributions paid to the central government minus social benefits received from the central government. If the per capita net tax payment of a region is above the national average, the region is defined as a contributor to interregional transfers, and if the per capita net tax payment is below the national average, the region is defined as a recipient. The study finds that the Flemish Region transferred through the tax and social security system EUR 5.8 billion in 2005 (1.9% of national GDP) to the Walloon Region and the Brussels-Capital region transferred 212 million (0.1% of national GDP). The main reasons for the implicit transfers to the Walloon Region are its lower personal income, which leads to a lower base for income tax and social security contributions, and its higher unemployment, which leads to higher social benefits.

The study also projects how inter-regional transfers could develop until 2030 under different scenarios regarding labour market developments in the three regions. In the scenario, which assumes persistence of current divergences of employment rates, the net contribution of the Flemish Region is more than halved while the contribution of the Brussels-Capital Region increases and the transfer to the Walloon region declines. The main reason for the change in inter-regional transfers is the relatively faster ageing of the population in Flanders, leading to net transfers to Flanders in the areas of pensions and health care. In a second scenario, which assumes convergence of employment rates by 2030, the Flemish Region would switch from a net contributor to a recipient of interregional transfers by 2030. The Walloon Region would remain a beneficiary of interregional transfers but these would be less than half of the level in 2005 while the Brussels-Capital region would be the only contributor (Table 3.7).

An alternative and broader approach to analysing interregional transfers would be to consider that the federal government returns a good part of its tax collection back to the regions through shared taxes and through other transfers (including equalisation transfers to the Walloon and the Brussels-Capital regions) so that the net tax payment of the regions to the federal government and thus the overall amount of interregional transfers would be smaller. (2)

Another problem of estimating inter-regional transfers is how to treat tax payments of commuters who live in one region and receive their income (and their tax base) in another region. In the above-mentioned study, tax payments of commuters are fully allocated according to the "residence principle" (as generally used in income taxation). (3) However, if one instead applied the "source principle" (as is done when measuring the contribution of commuters to regional GDP), the inter-regional transfers would be different; the Brussels-Capital Region would be a much larger contributor, Flanders would be a smaller contributor, and Wallonia would be a larger net beneficiary.

Improving coherence of policies of sub-federal governments in areas of national interest

Given the complex division of responsibilities between the federal and sub-federal governments, it is a major challenge to achieve policy goals that have been fixed at the national level. In a number of areas, such as employment, economic growth, energy and environment, there are clear national responsibilities and commitments although policies of sub-federal governments have major influences. It is therefore of key importance that actors at all levels of government co-operate closely so that national goals can be achieved. In this respect, dialogue between different institutional levels is pursued through a number of sector specific bodies, such as the Belgian Forum of Regulators. By contrast, insufficient co-ordination and co-operation reduces effectiveness and efficiency of policies for the whole country.

The management of the unemployed is part of labour market policy and therefore mostly a regional and partially a community responsibility. Regions are responsible for employment policy while Communities are responsible for training. Due to the merging of the Flemish community with the Flemish region these policies are integrated in Flanders. Employment policy and training have also been integrated between the Walloon region and the French-speaking Community, while these policies are not integrated in the Brussels-Capital region. However, the payment of unemployment benefits is part of social security, which falls under federal jurisdiction. This inconsistency between responsibility and funding suggests that there may be not enough incentives for the lower level of government to pursue activation measures for the unemployed with the necessary vigour. Indeed, it has been argued that the employment agency in the Walloon Region has in the past been relatively lax with the unemployed who refuse job offers (Gerard, 2002). More recently, measures have been taken in all regions to improve activation policies and also to increase labour mobility between the regions by improving exchange of information between regional employment offices. Currently, the resident principle in the revenue sharing system gives regional governments stronger incentives to facilitate job offers to unemployed from its own region as compared with those living and commuting from other regions. Labour mobility could also be improved by better training and by systematically informing people about job offers in the various regions. Furthermore, limiting the duration of unemployment benefits would shift some of the costs for the long-term unemployed from the federal government to the municipalities. Transferring budgetary and implementation responsibilities of social assistance to the municipalities and allowing them to use budget savings in this area for other purposes should also be considered. Such a scheme has been implemented in the Netherlands and is providing strong incentives to monitor and activate benefit recipients (OECD, 2008b).

In the area of R&D policies, responsibilities are divided between regions, communities and the federal government so that altogether five governments are in charge of R&D policies (governments of the three regions, the French Community and the federal government). Regions are responsible for all research, which is carried out in the areas under their competence, such as economic policy, energy, agriculture and environment. This includes most of applied research, including the promotion of business R&D. As the Flemish region has been merged with the Flemish Community, it is also responsible for the research of Flemish universities and higher education colleges, while the French-speaking Community is responsible for research in its universities and higher education colleges. The federal government is responsible for international research in the area of aerospace, for the legal framework of property rights (patents), for research that requires coordination between different regions and communities, such as joint projects of universities in different regions (poles d'attraction interuniversitaire) and for research in the federal scientific institutions. While it can be argued, that the regionalisation of R&D has the advantage of adjusting policies better to the local needs, there is a clear risk of fragmentation with projects and research centres failing to meet the critical mass. According to the European Innovation Scoreboard (EIS) Belgium belongs to the group of countries which are classified as "innovation followers". This also suggests that there is room for improving R&D policies. By contrast, four other smaller European countries (Sweden, Finland, Switzerland and Denmark) are classified by EIS as "innovation leaders" (together with Japan and Germany) (European Innovation Scoreboard, 2007). In order to improve R&D policies, it would be desirable to evaluate R&D policies in Belgium at all levels of government and to consider if efficiency could be improved by giving more responsibility to the federal government. At the minimum, all ministries involved with R&D policies need to cooperate closely and also exchange information about the effectiveness of policies.

Human capital formation is of key importance for economic growth and, given the ageing of the population, this source of growth will become even more important in the future. The decentralisation of education to the three linguistic communities (Flemish, French and German) has led to a degree of heterogeneity of educational institutions in Belgium which is probably larger than in any other OECD country (Joumard and Kongsrud, 2003). Such heterogeneity is not necessarily a problem, and could even be beneficial, if it reflects different preferences of citizens or leads to quality-enhancing competition. Problems arise, however, if regionalisation of education policies leads to major differences in the quality of education in the various parts of the country. The last Survey analysed in more detail issues concerning tertiary education and found that in Belgium tertiary education has been expanded over the past decades, setting the country on a course towards having relatively high human capital formation, but it also found that there has been no matching increase in funding (OECD, 2007). Spending per student is significantly lower in the French community than in the Flemish community so that without additional means, the quality of tertiary education for French speakers could suffer, restraining growth in Wallonia and in Belgium as a whole.

In primary and secondary education, a study by the OECD (Gonand et al., 2007) found that in both the Flemish and the French communities, efficiency of spending is below OECD best practices. While the sub-indicator "user choice" is in both communities close to best practices, and the Flemish Community (but not the French Community) also receives a high ranking with respect to "decentralisation", the other sub-indicators are close to OECD average with the exception of "benchmarking" which appears to be very poor in both communities and receives the lowest ranking in the Flemish Community (Figure 3.2).

In the area of environmental and energy policies, the regional and the federal governments are taking measures to tackle climate change and achieve their part of the Kyoto commitment through such measures as rigorous building performance measures and green certificate trading schemes. In energy policy, a notable recent positive development is the decision to create a centralised oil stockholding agency (IEA, 2006). There are various bodies to co-ordinate energy and environmental policies between the regions and the federal government. The Cellule (CONCERE/ENOVER or Concertation Etat-Regions pour l'Energie/Energieouerleg) is an advisory body for discussions between the federal government and the regional governments on all energy matters, which have been transferred to the regions. The National Climate Commission is a forum for policy cooperation, which will propose a draft National Climate Plan to the Extended Interministerial Conference for the Environment. Another forum for policy co-ordination is the federal Interdepartmental Commission for Sustainable Development. Furthermore, the various regulators of the electricity and gas markets have established a regular consultation process. Nonetheless, there is room for improving policies in these areas (see IEA, 2006), in particular by:

* Further harmonizing regional energy markets rules and regulations. While significant progress has been made in liberalising the electricity market by allowing transparent and non-discriminatory access to the grid, this market is not fully integrated as market rules and regulations are not fully harmonised across the regions, which creates higher costs for market participants and particularly new entrants. The current multi-layer and multi-regulator structure should be reviewed. At the very least, co-ordination and cooperation among the regulators should be further strengthened and regional and federal governments should work together to fully harmonise rules and regulations (including public service obligations and licence requirements) (see also Chapter 5).

* Creating an integrated green certificate trading scheme by making all the green and combined heat and power (CHP) certificates transferable between all regions; currently, most certificates issued in different regions cannot be traded between regions, except between Wallonia and the Brussels-Capital region.

[FIGURE 3.2 OMITTED]

The political economy of reforming the fiscal system

The implementation of reform needs to take political economy considerations into account. It has been argued that specifying responsibilities of the regions by Special Laws rather than by changes in the Constitution provides more flexibility to adjust to new developments. Nonetheless, regions and language communities have de facto the power to prevent changes in fiscal federalism, which they perceive as unfair. A reform of the current division of competences between the federal State, the Communities and the Regions requires either a revision of the Constitution or a so-called "special law". The political renegotiation process in Belgium is therefore complicated and has also been critisised as favouring sub-federal governments at the expense of the federal government (Choudhry and Perrin, 2007).

Although in Belgium there seems to be agreement that the fiscal system should be reformed, it is difficult to find a consensus in which direction to go. The parties involved have so far not been able to agree on a new reform, originally planned after the June 2007 federal elections. However, without reform, the federal budget will probably not be able to carry the whole burden of consolidation and of age-related costs and--if the resistance to reform persist--could be forced to shift the burden to citizens through higher taxes, which are already among the highest in the OECD (see Chapter 4). Furthermore, persisting vertical horizontal imbalances could in the end also lead to lower growth and lower living standards in Belgium as a whole. Finding a consensus therefore requires that all parties have to be convinced that they gain from the reform in the longer run even if some of the measures may be costly for them in the short-term. Reinforcing the longer-term perspective will be easier if the reform package is broad enough and is not perceived as being purely redistributive, but rather as being essential to increase efficiency of the public sector and of living standards in the whole country. The main policy recommendations are presented in Box 3.3.
Box 3.3. Main policy recommendations for improving fiscal
federalism

Strengthen the fiscal position of the federal government

* Improve the revenue base of the federal government by reducing
the amount of shared income taxes, which is transferred to the
regions. Redesign the shared income tax in a way that these
transfers are not growing faster than actual tax revenues.

* Allocate pension expenditure for civil servants of sub-federal
administrations, which is currently a federal responsibility, to
the lower government levels. This would reduce incentives of
sub-federal governments for allowing early retirement of their
civil servants, as they would have to bear the full costs.

* Encourage sub-federal governments to better use their potential
to create own revenues (such as personal income taxes or property
taxation and user fees).

Redesign the allocation of shared personal income tax between
regions

* Take measures to internalize fiscal externalities between
regions, notably in the Brussels-Capital region. This could be
done, for example, by allocating (partially) the shared income tax
to the region where the firm of the worker is located ("workplace
principle").

Redesign the system of equalisation transfers

* Reduce the disincentives for regions which receive equalisation
transfers to develop their own tax base. This could be achieved by
reducing the amount of equalisation for each percentage point of
the tax gap so that the system becomes less progressive.

Improve coherence of policies and spending efficiency of
sub-federal governments in areas of national interest

* In the area of employment policy, activation policy should be
improved to best practice. People should be systematically informed
about job offers in all regions, for example by creating a joint
website for vacancies in Belgium. The duration of unemployment
benefits should be limited to shift some of the costs for the
long-term unemployed from the federal government to the
municipalities. Furthermore, consideration should be given to
transfer budgetary and implementation responsibilities of social
assistance to the municipalities and allowing them to use budgetary
savings in this area for other purposes, providing the
municipalities with strong incentives to monitor and activate
benefit recipients.

* In the area of R&D policies, evaluate R&D policies at all levels
of government and also explore where efficiency could be improved
by giving more responsibility to the federal government. At the
minimum, all ministries involved with R&D policies need to
cooperate closely and also systematically exchange information
about the effectiveness of policies.

* In the area of education, ensure that a high quality is offered
in the whole country. This requires an adequate allocation of
transfers from the federal government, increasing spending
efficiency and efforts at the sub-federal level to create own
revenues, such as from tuition fees in tertiary education, combined
with income contingent loans.

* In the area of environmental policy, create an integrated green
certificate-trading scheme by making all the green and combined
heat and power (CHP) certificates transferable between all regions.

* In the area of energy policy, review the current multi-layer and
multi-regulator structure and, at the very least, strengthen
co-ordination and co-operation among the regulators and fully
harmonise rules and regulations (see also Chapter 5).


ANNEX 3.A1

Main historical steps of Belgium's federalism

Belgian federalism is an evolving process as is illustrated below (van der Stichele and Verdonck, 2002). The Lambermont Agreement of 2001 was the most important recent reform, and a consensus for a new reform has not been reached yet despite the setting up of a group of experts (2007).

* First state reform 1971: The Constitution recognises the existence of three regions (Flanders, Wallonia and Brussels) and three cultural communities (French, Dutch and German).

* Second state reform 1980: The regions and the communities were fully established with their own responsibilities for matters relating to the person (communities) and the territory (regions).

* Third state reform (special financing act, 1989): Belgium became de facto a federal state composed of regions and communities, which received expanded responsibilities, including the financial sovereignty over the use of transfers from the federal government.

* Fourth state reform 1993:

* Saint Michel act, 1993: Revision of financing mechanism concerning regions and communities.

* Saint Quentin decree, 1993: transfer of some tasks from the French-speaking community to the Walloon region and the Brussels region (Commission communautaire francaise or "Cocof").

* Saint Eloi act 2000: Revision of the allocation of the VAT transfer between communities.

* Fifth state reform with Lambermont (or Saint Polycarpe) agreement, January 2001: reform of community financing; extension of taxing power of the regions.

* Lombard act, April 2001: Financial assistance for community commissions and communes in Brussels to ensure the functioning of the institutions.

* Saint Boniface agreement, June 2001: Solidarity measure among French-speakers as the Walloon region and Cocof provided financial assistance to the French-speaking community. Furthermore, it was determined how the French-speaking community should use the future funds obtained after the Lambermont agreement.

Bibliography

Adam, A., M.D. Delis and P. Kammas (2008), "Fiscal Consolidation and Public Sector Efficiency: Evidence from OECD Countries", CESIFO Working Paper, No. 2364, August.

Algoed. K and D. Heremans (2008), The Incentive Effects of the Belgian Financial Arrangements foer the Regions, Spoor A2: Toekomstscenario's voor de vlaamse begrotting en fiscaliteit, Faculteit Economie en Bedrijfswetenschappen, February.

De Boeck. E. and J. van Gompel (1998), "Financiele stromen tussen de Belgische gewesten opnieuw bekeken", in C. Vanderveeren and J. Vuchelen (eds.), Een Vlaaamse fiscaliteit binnen een economische en monetairee unie.

Callatay, E. (2007), "Finances publiques et reformes institutionnelles : le role central de la Region de Bruxelles-Capitale", in Regards economiques, IRES-UCL, No. 52.

Capron, H. (2000), "The Sources of Belgian Prosperity", in: H. Capron and W. Meeusen (eds.), The National Innovation System in Belgium, Heidelberg-New York.

Cattoir, Ph. and F. Docquier (1999), "Securite sociale et solidarite interregionale", Chapter 8, in F. Doequier (editor), La Solidarite entre les Regions, Bilan et Perspectives, Brussels : de Boeck Universite, pp. 51-98.

Cattoir, Ph. and M. Verdonck (1999), "La perequation financiere, analyse de quatre pays", in F. Docquier (editor), La Solidarite entre les Regions : Bilan et Perspectives, de Boeck Universite, Brussels.

Choudhry. S. and B. Perrin (2007), "The Legal Architecture of Intergovernmental Transfers: A Comparative Examination", in R. Boadway and A. Sha, Intergovernmental Fiscal Transfers, the World Bank.

Dury. D., E. Eugene, G. Langenus, K. van Cauter, L. van Meensel (2008), "Interregional Transfers and Solidarity Mechanisms via the Government Budget", National Bank of Belgium, Economic Review, September.

EIS (European Innovation Scoreboard) (2007), Comparative Analysis of Innovation Performance, Maastricht Economic Research on Innovation and Technology and Joint Research Centre of the European Commission.

Gerard, M. (2002), "Fiscal Federalism in Belgium", Paper presented at the Conference on Fiscal Imbalance, Quebec City, 13-14 September 2001.

Gonand. F., I. Joumard and R. Price (2007), "Public Spending Efficiency: Institutional Indicators in Primary and Secondary Education", OECD Economics Department Working Paper, No. 543, OECD, Paris.

IEA (2006), "Energy Policies of IEA countries", Belgium 2005 Review.

Joumard. I. and P.M. Kongsrud (2003), "Fiscal Relations Across Government Levels", OECD Economics Department Working Paper, No. 375, OECD, Paris.

OECD (2007), OECD Economic Survey Belgium, Paris.

OECD (2008a), Growing Unequal, Paris.

OECD (2008b), OECD Economic Survey Netherlands, Paris.

Van der Stichele, G. and M. Verdonck (2002), "The Lambermont Agreement: Why and How?", Paper presented at the Conference on Fiscal Imbalance, Quebec City, 13-14 September 2001.

Van der Stichele, G. (2003), "Les modes alternatifs de financement de Bruxelles", update of the study of 1999 by Cattoir, Ph., J.-P. Lambert, M. Taymans, H. Tulkens, M. Tulkens, G. van der Stichele and M. Verdonck, Saint Louis University, Brussels.

Stienlet, G. (2000), "Institutional Reforms and Belgian Fiscal Policy in the 1990s", in Eds. R.R. Strauch and J.V. Hagen, Institutions, Politics and Fiscal Policy, Kluwer Academic Publishers Boston.

Notes

(1.) The fiscal autonomy margins for additional taxes and tax reductions were set at 3.25% until end 2003 and at 6.75% up from 2004 of the income tax yield in the region. The federal government passes these changes on by changing the withholding taxes.

(2.) In 2007, the transfers received from the federal budget amounted for the Flemish Community (and Region) to EUR 16.6 billion (per capita EUR 2 714), for the French Community to EUR 7.35 billion (per capita EUR 1 725), for the Walloon Region, EUR 3.32 billion (per capita EUR 965) and for the Brussels Capital Region EUR 745 million (per capita EUR 723). If one attributes (approximately) the transfers received by the French Community to the Walloon Region and the Brussels Capital Region by using their shares in population (around 77% for the Walloon Region and 23% for the Brussels Capital Region) in order to make the numbers comparable with those in Flanders, the Walloon Region received about EUR 9 billion (per capita EUR 2 610) and the Brussels-Capital Region around EUR 2.5 billion (per capita EUR 2 376).

(3.) The redistributive effects which are derived from applying the residence principle show, strictly speaking, the inter-personal redistribution of income between the people who are living in the different regions while, when applying the source principle, the effects show the redistribution of income between the regions where income is created independent of the places where people are living.
Table 3.1. GDP per capita in the regions of Belgium
Mean of EU15 countries = 100

        Walloon    Flemish    Brussels-Capital
         Region     Region         Region        Belgium

1955      103         92            155            103
1965       93         93            163            100
1975       90        106            161            107
1985       87        106            167            106
1995       83        111            230            113
2000       77        105            215            106
2005       78        107            213            108
2006       77        107            208            107

Source: Capron, H. (2000), "The sources of Belgian prosperity",
in: H. Capron and W. Meeusen (eds.), The National Innovation
System in Belgium, unti11985; National Bank of Belgium for more
recent data, which, for EU15, include German new Lander.

Table 3.2. Fiscal developments of the federal government
and of Regions and Communities

Percentage of GDP
                           1990-94   1995-99   2000-04   2005-07
Federal government (1)
  Revenue (2)                 17.9      17.6      17.7      15.6
  Primary expenditure         13.8      12.8      12.5      13.0
  Primary balance              4.1       4.7       5.1       2.6
  Interest payments           10.0       7.2       5.4       3.8
  Fiscal balance              -5.8      -2.5      -0.2      -1.2
Regions and Communities
  Revenue                     11.8      13.2      13.9      14.3
    of which:
      transferred taxes        8.6       9.3       8.9       8.5

  Primary expenditure         12.4      13.0      13.5      13.9
  Primary balance              0.6       0.2       0.5       0.4
  Interest payments            0.2       0.3       0.2       0.1
  Fiscal balance              -0.8      -0.1       0.2       0.2

(1.) Excluding social security.

(2.) Excluding transferred taxes.

Source: National Bank of Belgium.

Table 3.3. Spending and revenues by levels of government

                         Share in general government spending (1)

                       Federal government    Intermediate level
                      1995 (3)   2006 (4)   1995 (3)   2006 (4)

Federal countries
Austria (5)               46.3       45.7       13.1       15.1
Belgium                   31.7       23.2       20.5       23.6
Canada                    35.7       29.6       40.4       45.1
Germany                   29.6       19.1       18.7       21.9
Switzerland               17.4       14.8       31.0       33.7
United States (6)         58.3       56.3
Unitary countries
Czech Republic            64.9       60.3
Denmark                   38.2       31.8
Finland                   36.8       29.9
France                    39.3       35.0
Greece                    63.0       53.7
Hungary                   51.4       49.4
Iceland                   55.2       50.0
Ireland                   58.0       69.7
Italy                     49.7       33.5
Japan (5)                 35.1       33.8
Korea                     43.2       40.4
Luxembourg                44.6       45.9
Netherlands               32.4       29.5
New Zealand               89.5       89.3
Norway                    63.9       68.8
Poland (5)                53.8       35.4
Portugal                  62.5       54.1
Slovak Republic           61.4       48.4
Spain                     34.4       22.4       21.6       35.8
Sweden                    52.3       43.9
United Kingdom            74.2       71.6

                         Share in general government spending (1)

                        Local government       Social security
                      1995 (3)   2006 (4)   1995 (3)   2006 (4)

Federal countries
Austria (5)               15.3       12.2       25.3       26.9
Belgium                   12.3       14.1       35.4       39.1
Canada                    18.6       19.3        5.3        6.1
Germany                   14.2       15.2       37.5       44.0
Switzerland               23.3       20.9       28.3       30.6
United States (6)         42.8       44.9
Unitary countries
Czech Republic            22.8       27.5       12.3       12.3
Denmark                   53.8       63.3        8.0        5.0
Finland                   30.5       39.2       32.8       30.9
France                    17.4       20.2       43.2       44.7
Greece                     4.9        6.3       32.1       39.9
Hungary                   24.9       24.7       23.7       26.0
Iceland                   27.6       31.7       17.2       18.3
Ireland                   30.6       19.7       11.4       10.6
Italy                     24.0       31.5       26.3       35.0
Japan (5)                 38.5       31.9       26.5       34.4
Korea                     45.2       45.5       11.5       14.2
Luxembourg                13.6       11.5       41.8       42.6
Netherlands               33.7       33.5       33.9       37.0
New Zealand               10.5       10.7        0.0        0.0
Norway                    36.1       31.2        0.0        0.0
Poland (5)                19.0       30.7       27.2       33.9
Portugal                  11.1       12.9       26.4       33.0
Slovak Republic           13.2       17.5       25.5       34.1
Spain                     11.3       13.4       32.6       28.5
Sweden                    37.7       44.8        9.9       11.3
United Kingdom            25.8       28.4        0.0        0.0

                         Share in general government revenues (2)

                       Federal government    Intermediate level
                      1995 (3)   2006 (4)   1995 (3)   2006 (4)

Federal countries
Austria (5)               42.2       44.2       14.4       15.3
Belgium                   57.3       55.9        5.4        9.1
Canada                    40.4       39.2       42.1       42.9
Germany                   27.5       28.2       24.0       23.7
Switzerland               30.0       33.0       27.6       28.3
United States (6)         63.7       63.6
Unitary countries
Czech Republic            72.8       67.8
Denmark                   64.7       62.3
Finland                   46.2       48.4
France                    41.2       38.6
Greece                    69.7       66.4
Hungary                   59.6       58.2
Iceland                   77.1       74.5
Ireland                   79.8       81.5
Italy                     61.1       53.2
Japan (5)                 29.4       32.2
Korea                     67.5       59.6
Luxembourg                64.9       67.2
Netherlands               55.6       58.7
New Zealand               90.9       91.4
Norway                    80.0       86.3
Poland (5)                48.4       50.5
Portugal                  63.7       61.0
Slovak Republic           61.1       53.0
Spain                     52.2       37.7        7.4       23.5
Sweden                    56.1       54.3
United Kingdom            91.8       91.1

                         Share in general government revenues (2)

                        Local government       Social security
                      1995 (3)   2006 (4)   1995 (3)   2006 (4)

Federal countries
Austria (5)               16.0       12.8       27.4       27.6
Belgium                    8.1        7.3       30.0       28.2
Canada                    11.8       10.4        5.7        7.4
Germany                   12.8       12.2       38.9       37.5
Switzerland               22.2       20.0       23.5       20.2
United States (6)         37.6       37.7
Unitary countries
Czech Republic            19.9       20.1       10.7       14.0
Denmark                   33.4       35.8        1.9        1.9
Finland                   26.7       26.5       27.6       26.0
France                    15.0       16.2       45.1       46.2
Greece                     3.9        3.9       28.7       32.0
Hungary                   15.1       17.1       28.1       31.6
Iceland                   23.4       25.7        0.0        0.2
Ireland                   13.7       13.0       10.7       10.8
Italy                     12.9       20.3       28.0       28.1
Japan (5)                 37.0       32.9       33.6       34.9
Korea                     22.0       22.0       15.8       22.5
Luxembourg                 9.7        7.4       26.3       26.4
Netherlands               13.5       11.9       32.5       30.6
New Zealand                9.1        8.6        0.0        0.0
Norway                    20.0       13.7        0.0        0.0
Poland (5)                18.6       18.9       33.0       30.9
Portugal                  10.6       11.5       27.9       29.8
Slovak Republic            6.5       13.6       32.6       35.0
Spain                     10.6       10.9       31.4       29.7
Sweden                    33.6       34.7       10.6       11.0
United Kingdom             9.1       10.7        0.0        0.0

(1.) Excluding the transfers paid to other levels of government.

(2.) Excluding the transfers received from other levels of
government and including tax sharing arrangements. In Belgium, the
so called shared tax is not included but treated as a transfer.

(3.) Or earliest year available: 1996 for Japan, Netherlands and
Norway; 1997 for the Czech Republic; 1998 for Iceland; and 2000 for
Greece, Korea and Hungary.

(4.) Or latest year available: 2005 for New Zealand.

(5.) Unconsolidated data (only in 1995 for Poland).

(6.) For the United States, no breakdown between state and local
governments is available.

Source: OECD National Accounts database, Statistics Canada, US
Bureau of Economic Analysis.

Table 3.4. Revenue sources of regions and communities

                          As a percentage of their total revenue

                            1990     1995    2000     2007

Total revenue              100.0    100.0   100.0    100.0
  Of which:
    Own taxes                6.7      7.6     8.8     16.6
    Transferred taxes       74.7     71.6    69.2     59.3
    Revenues from
      other government
      transfers              9.1      8.5     8.7      9.5
    Other revenue            9.5     12.3    13.3     14.7

                                 As a percentage of GDP

                            1990     1995    2000     2007

Total revenue               11.1     12.6    13.2     14.3
  Of which:
    Own taxes                0.7      1.0     1.2      2.4
    Transferred taxes        8.3      9.0     9.1      8.5
    Revenues from
      other government
      transfers              1.0      1.1     1.1      1.4
    Other revenue            1.1      1.5     1.8      2.1

Source: National Bank of Belgium, OECD calculations.

Table 3.5. Expenditure and revenue of regions and communities (1)
2007, in euro per capita

                     Flemish Region             Walloon
                      and Community              Region

Expenditure                   3 623    1 781 (3 615) (2)
Revenue                       3 755    1 848 (3 677) (2)
  Own revenue                   705                  643
  Transfers
    received                  2 714      965 (2 612) (2)
  Other revenue                 337                  240
Population                6 117 440            3 435 879

                   Brussels-Capital              French
                             Region           Community

Expenditure      2 670 (4  496) (2)               1 921
Revenue           2 672 (4 491) (2)               1 915
  Own revenue                 1 240                   0
  Transfers
    received        723 (2 362) (2)              17 250
  Other revenue       709 (889) (2)                 189
Population                1 031.215           4 260 851

(1.) The numbers in bracket (approximately) allocate spending
and revenues of the French Community to these two regions by
using population shares.

(2.) Including expenditure and revenue of the French Community.

Source: National Bank of Belgium, OECD calculations of the
numbers in brackets.

Table 3.6. Budgetary effects of autonomous increases in the
personal income tax yield in Belgian regions
Change in total revenues
                                                          Federal
                       Wallonia   Flanders   Brussels   government
Increase in PIT
  yield in Wallonia          -          -          +          +++
Increase in PIT
  yield in Flanders          +          +          +           ++
Increase in PIT
  yield in Brussels          +          -          -          +++

Source: Gerard, M. (2002), "Fiscal Federalism in Belgium", Paper
presented at the Conference on Fiscal Imbalance, based on Cattoir,
Ph. and M. Verdonck (1999), "La perequation financiere, analyse de
quatre pays", in F. Docquier (editor), La solidarite entre les
Regions : Bilan et Perspectives, de Boeck Universite.

Table 3.7. Estimate of implicit interregional transfers
In per cent of national GDP

            2005                              2030
                   Scenario A (1)   Scenario B (2)

Flanders    -1.9             -0.8              0.2
Wallonia     2.0              1.6              0.9
Brussels    -0.1             -0.7             -1.1

(1.) Scenario A: Current differences in employment rates persist.

(2.) Scenario B: Employment rates converge in 2030.

Source: Dury. D., B. Eugene, G. Langenus, K. van Cauter, L. van
Meensel (2008), "Interregional transfers and solidarity mechanisms
via the government budget", National Bank of Belgium, Economic
Review, September 2008.
COPYRIGHT 2009 OECD Publications and Information Centre
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Publication:OECD Economic Surveys - Belgium
Geographic Code:4EUBL
Date:Jul 1, 2009
Words:10758
Previous Article:Chapter 2: securing fiscal sustainability during the economic downturn and beyond.
Next Article:Chapter 4: how to reform the tax system to enhance economic growth.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters