Une analyse comparative des budgets des principales communes urbaines de Belgique Zijn de Brusselse gemeenten spilziek? Een vergelijkende analyse van de gemeentebegrotingen van de grootste stedelijke clusters van België Are the municipalities of Brussels guilty of wasteful spending? A comparative analysis of the budgets of Belgium’s main urban municipalities

The municipalities of Brussels are often accused of being a source of financial chaos and of requiring urgent reforms. The present article attempts to objectivise the question of relative wasteful spending by the municipalities of Brussels based on a comparison of the finances of the 19 municipalities of Brussels with those of the municipalities of the four other major urban entities in Belgium: Antwerp, Charleroi, Ghent and Liege.Our methodology does not allow us to affirm that the services of the municipalities of Brussels could not be provided more efficiently. It does, however, allow us to affirm that in the case of excessive spending by the municipalities of Brussels, it would not be more than that of the other entities studied, especially if we consider the fact that the other municipalities are able to hand over part of their competences to the provinces or to a higher number of intermunicipal companies.We were not able to identify a local sector of competence with significant shortcomings, apart from the health and hygiene sector which deserves a more in-depth analysis. On the contrary, the management constraints which the municipalities of Brussels are subjected to are what clearly emerged.


Introduction
In recent years and months, Brussels institutions have been accused of being poorly managed, a source of financial waste and in need of urgent reforms. Among these institutions, the 19 municipalities of the Brussels Region are the main targets.
It is a praiseworthy goal to try to improve the management of the state, but unfortunately the points of view described above are rarely supported, except by some significant -yet often rough or even misleading -indicators. Reforms are not conceivable on this basis.
The present article 1 attempts to objectivise the question of relative wasteful spending by the municipalities of Brussels based on a comparison of the finances of the 19 municipalities of Brussels with those of the municipalities of urban entities linked to the four other major cities in Belgium: Antwerp, Charleroi, Ghent and Liege.
In our comparison, we have paid special attention to the choice of units of measurement, as they guarantee the quality of the comparison. Thus, for example, if we observe that the municipalities of Brussels spend 22% more on secondary education than the average per inhabitant, we may be inclined to consider it wasteful spending. If we take a closer look, we see, however, that the expenditure per student is 27% lower than the average. In this case, what may have appeared to be wasteful spending seems to be related to a problem of equity.
It is in this spirit that the main categories of municipal revenue and expenditure are analysed, each time seeking to explain the differences observed.
This exercise must be understood as a first step in the direction of a more in-depth analysis of efficiency that would require not only the quantity but also the quality of public goods and services to be measured. The available statistical data and the length of the study (six months) do not allow such a close examination of each area of competence of the municipalities. 2 The purpose of our approach is therefore to highlight the sectors in which the municipalities of Brussels seem to fall short, in order to allow the public authorities to establish priorities as regards future studies and reforms.

Methodological precisions
For our comparison, we defined geographical sectors comparable to the entity comprising the 19 municipalities of Brussels. We therefore defined 'urban entities' made up of the municipalities of Antwerp, Charleroi, Ghent and Liege, widened to include their neighbouring municipalities with an 'urbanised' character according to the typology used by Dexia. 3 This selection resulted in a sample of 67 municipalities distributed as follows: entity of Brussels (19 municipalities/1,048,491 inhabitants), entity of Antwerp (19 municipalities/815,966 inhabitants), entity of Ghent (5 municipalities/300,539 inhabitants), entity of Charleroi (10 municipalities/372,850 inhabitants) and entity of Liege (14 municipalities/498,650 inhabitants).
In order to make a useful comparison of the budgetary data of the municipalities which belong to the different entities, it is necessary to consider three important institutional differences between them. Firstly, the fire services and rubbish collection and waste treatment services are provided by the municipalities in Flanders and in Wallonia, but are regionalised in Brussels. Secondly, in the Flemish and Walloon Regions, the provinces are responsible for many other services which are similar to those devolved to the municipalities, whilst in Brussels this institutional level no longer exists. Thirdly, there are municipal management methods which lead to a rather disparate institutional landscape from one urban entity to the next, for example as regards the use of government corporations, non-profit organisations or other intermunicipal companies for the management of certain municipal activities.
We were able to neutralise the impact of the regionalisation of fire, rubbish collection and waste treatment services in the budget for our comparison. For the matters shared between municipalities and Walloon and Flemish provinces, a neutralisation was not possible, but figures related to the staff employed by the two levels of authority enable an estimation of the extent of the tasks ensured by the provinces in Flanders and Wallonia, while only the municipalities are responsible in Brussels. The Flemish provincial staff represent 8.6% of the total Flemish municipal staff. In Wallonia, the percentage is 23.82%. The same approach allows an estimation of the extent of externalised management methods: in Brussels, the number of people employed by intermunicipal companies corresponds to 11.03% of the municipal staff, in Flanders, 11.24% and in Wallonia, 52.49%. Brussels Studies the e-journal for academic research on Brussels 2 2 See Hindriks, Albrecht and De Vos (2006) for an example of a comparison integrating qualitative elements. It does, however, analyse only a limited number of municipal competences and restricts the comparison to Flemish municipalities between themselves and Walloon municipalities between themselves, due to a lack of sufficient comparability of data between the regions.
This institutional comparison indicates that the institutional differences whose financial impact could not be neutralised lead us to overestimate the relative expenditure of the municipalities of Brussels. In other words, if the comparative analysis of the municipal finances of the five entities were to lead to the observation of equivalent situations, this would mean that the municipalities of Brussels are more economical than their Walloon and Flemish counterparts.
Let us end this methodological preamble by specifying that the comparative study is limited to ordinary revenue and expenditure (i.e. for operations), as special revenue and expenditure (i.e. related to investments) are too unpredictable for conclusions to be drawn based on the observation of a few fiscal years only (from 2002 to 2006).
The results are presented according to a functional classification based on municipal activities covered by revenue or expenditure (education, housing, police, health, etc.), and only the headings representing a significant share of revenue or expenditure are discussed here. 4 The sum of these headings therefore covers 79% of revenue and 78% of expenditure of the Brussels entity. Let us note that the general conclusions do not change if the smaller revenue and expenditure items are included.

Comparison of the municipal finances of five Belgian urban entities
Before beginning the analysis of the comparison results, it is important to understand the significance of the relative position of Brussels according to the units of measurement used. When the unit is the number of euros per inhabitant, it is generally the cost (for an expenditure) or the actual yield (for revenue) which is targeted. It allows us to see, for example, whether the expenses covered by inhabitants is higher or lower than the average. When the unit of measurement is different, the efficiency of revenue or expenditure is tested instead, in terms of an economical use of public monies.
If we take the example of CPAS (public social welfare centres) expenditure, Brussels is situated far above the average in terms of euros per inhabitant, but is far below the average in terms of euros per beneficiary of social integration revenue. This means that the socioeconomic situation of Brussels is such that CPAS expenses are much greater for the inhabitants of Brussels than for the inhabitants of the other entities, but that these expenses are not due to excessive expenditure because -for each beneficiary of social integration revenue -the municipalities of Brussels use their means more sparingly than their Walloon and Flemish counterparts.
More frugal expenditure therefore does not necessarily imply a better budgetary situation.

Comparison of municipal revenue
In this article, we shall limit the discussion of results to the four main sources of municipal revenue, covering in total 79% of the means of the municipalities of Brussels: taxes and fees, grants, primary education subsidies and secondary education subsidies. The other types of revenue, which are not discussed in this article, are, for example, public corporation or intermunicipal company dividends, the regional financing of municipal hospitals or the renting of municipal properties.
Taxes and fees (tax surcharge on property tax and on personal income tax, municipal taxes, parking fees, etc.) constitute on average 43% of the means of the municipalities of Brussels. They rely much more on their own means than the other entities, where the share of tax revenues in the total revenue varies between 27% and 34%.
The Brussels entity benefits financially from the economic activity which takes place in its territory, as this type of revenue brings in 26% more per inhabitant than the average of the urban entities studied. The reason is, on the one hand, that the density of the built-up area is higher in the Brussels Region than it is elsewhere and that, on the other hand, in order to benefit from this asset, the municipalities apply a local piggyback tax on the regional property tax which is, on average, higher than it is in the Flemish or Walloon urban municipalities. Municipal taxes, which concern inhabitants as well as economic activity, also benefit from a relatively good yield per inhabitant in the Brussels entity, which is home to a particularly high number of companies.
In general, the advantageous tax position of the municipalities of Brussels hides the particularly unfavourable position of the local piggyback tax on personal income tax in terms of revenue per inhabitant, at 88.5% of the average. The tax rates as well as the incomes of inhabitants are lower than the average. The relatively low incomes with respect to the average are the result of the urban flight of members of the middle and upper classes to the outskirts of Brussels, which has been going on for many years. The lower than average tax rates reflect the will to limit this flight, faced with the competition from municipalities in the outskirts of Brussels where the tax rates are more advantageous for taxpayers. 5 For the Brussels entity, 95% of the grant revenues are made up of the General Regional Grant for the Municipalities (formerly, the municipality fund) and represent 19% of the means of the municipalities of Brussels, compared with 34% to 38% in the other urban entities. This grant is financed by the regional authority.
For this item, the position of the Brussels entity is unfavourable, as the municipalities of Brussels benefit from 39% less means in euros per inhabitant than the average of the five urban entities. Let us moderate this statement by specifying that as certain municipal competences were taken over by the Brussels Region, it is logical that it allocates less financial means to its municipalities. But the difference is above all due to the absence of solidarity between the municipalities of the Brussels Region and those of its surrounding area, which belong to other regions. This factor accounts for more than half of the difference with respect to the average. The four other enti- ties fully benefit from this solidarity. Let us mention that when the municipality fund was regionalised in 1970, the share of the municipalities of Brussels of the total grants dropped from 20.5% to 9.36%.
Municipal primary education revenue (community subsidies and contributions from parents) accounts for 10% of the revenue of the municipalities of Brussels on average. They may appear to be in a particularly comfortable position, as they are 22% above the average of the five urban entities per inhabitant. The difference is especially pronounced with respect to the Flemish entities due to a significantly greater proportion of students in the public French-language school system than in the catholic school system, which is more developed in Flanders.
The image changes greatly when the revenue per student is taken into account. The position of Brussels is 8% lower than the average. The Flemish entities are well above the average, reflecting the greater financial means of the Flemish Community.
Secondary education revenue (also made up of subsidies and contributions from parents) accounts for 7% of the revenue of the municipalities of Brussels on average, and follows a pattern which is similar to that of primary education, although in a more pronounced way. In terms of revenue per inhabitant, the municipalities of Brussels are situated 22% above the average, due to the fact that there are two to five times more students in the municipal system. But in terms of revenue per student, they are situated 27% below the average. The positive discrimination mechanism implemented for the financing of schools therefore does not appear to play a positive role for Brussels at secondary level, contrary to primary level.
In total, the ordinary revenue per inhabitant for the municipalities of Brussels (neutralising the effect of the regionalisation of fire and waste treatment services) amounts to 104.1% of the average. The positions of the other entities are: Antwerp, 104.5%, Ghent, 112.9%, Liege, 95.5% and Charleroi, 74.0%. Brussels remains below the figures observed in Antwerp and in Ghent -despite its role as political and economic capital which stimulates part of its tax revenues -due to the fact that the absence of solidarity from its surrounding area via the General Regional Grant for the Municipalities constitutes a serious handicap. This is added to the drop in the average taxable income per inhabitant as a result of the urban flight of middle and upper classes.
Let us recall that the municipalities of Brussels have competences which the municipalities in the other regions share with the provinces and large intermunicipal companies (in Wallonia). As it was not possible to neutralise this factor due to insufficient data, we know that the revenue and relative expenditure of the municipalities of Brussels are overestimated, but we do not know to what extent.
From this first part of the comparative analysis, we may affirm that, in relative terms, the municipalities of Brussels are not overfinanced.
Brussels Studies the e-journal for academic research on Brussels

Comparison of municipal expenditure
Seven items make up 78% of the municipal expenditure of the municipalities of Brussels: primary education, general services, justice and police, social aid and security, adult education and culture, health and hygiene and, finally, secondary education.
Primary education expenditure represents 15% of the total. Brussels spends relatively more per inhabitant (128% of the average) due to a higher proportion of students in the public French-language school system compared with the Flemish municipalities, where most students are enrolled in the catholic school system. But in terms of expenditure per student, Brussels is situated close to the average (104% of the average).
The municipalities of Brussels spend much less per student than the two Flemish urban entities, but considerably more than the Walloon urban entities. This could indicate that the specificities of the Brussels school population with, for example, a less good knowledge of the French language, lead to higher costs which are only partially compensated for by the positive discrimination mechanism introduced in the subsidies given to schools by the French Community. A more in-depth analysis would be necessary in order to determine whether this hypothesis is borne out or whether economies are possible for this item.
In terms of general services -which account for 15% of expenditure -the Brussels entity does not differ particularly from the other entities, at 94% of the average expenditure per inhabitant.
Justice and police expenditure is made up mostly of the municipal grant for police areas and represents 14% of total municipal expenditure. Police expenditure per inhabitant is higher in Brussels (128% of the average of the five entities), among others due to additional costs created by the economic, political, social and cultural activity which takes place in its territory.
Today, the police areas are financed jointly by the federal state and the municipalities. A royal decree 6 determines the mechanism for the distribution of the total federal grant between the police areas. Furthermore, the budget for the police areas may never present a deficit. The municipalities therefore fill the gap between the federal grant and the total needs of the area.
On average, the municipalities of non-Brussels urban entities devote 9.42% of their budget to the financing of police areas. In Brussels, this figure reaches 12.05%. This difference is due to the current federal financing mechanism for police areas, which is unfavourable to the areas in Brussels because it does not consider the criteria which have an obvious influence on the police workload in a capital, although it is based on 14 objective criteria (population, average taxable income, share of the elderly population, etc.). 7 These missing criteria include road traffic, the number of demonstrations, night and weekend activity, specific crimes in the territory of the capital, etc. This imbalance requires a bigger financial effort on behalf of the associated municipalities of Brussels, as they must compensate for the higher deficits in their police area.
Social aid and security expenditure (on average, 11% of the budget), is above all earmarked for the CPAS grant. In this area, the municipalities of Brussels spend 23% more per inhabitant, but 20% less per beneficiary of social integration revenue. The financial burden is therefore significant for the population, but the expenditure per social beneficiary is quite limited.
Apart from CPAS grants -i.e. sheltered workshops, at-home meals, association subsidies, etc. -Brussels is characterised by lower expenditure per inhabitant than in all of the other entities (€12/inhabitant versus €30/inhabitant on average). A certain austerity is most likely imposaed by the considerable amount devoted to the CPAS grants.
Although the municipalities of Brussels were forced to limit social aid and social security expenditure, it is partly because, as a whole, they had to face a 104% increase in the number of beneficiaries of social integration revenue between 1999 and 2008. In comparison, this number decreased by 32% in the Antwerp entity during the same period.
This first -greatly exogenous -factor in the growth of CPAS expenditure, and therefore of the municipal grants which they receive, is amplified by the method of financing of CPASs. They benefit from federal intervention equal to a percentage of the monthly amount of the allocated social integration revenue. 8 The basic grant is 50% of this amount. But the intervention is raised to 60% for CPASs with 500 or more eligible beneficiaries, and 65% for CPASs with more than 1000 eligible beneficiaries. This method of financing is surprising, as it is not so much the absolute number of beneficiaries (or eligible beneficiaries) which counts for CPASs and the municipalities which finance them, but the expenses which this represents per inhabitant. This rule is favourable to the vast and highly populated municipalities such as Antwerp. We therefore observe that almost half of the municipalities of Brussels have a number of social integration revenue beneficiaries per inhabitant which is higher than the rate observed in the city of Antwerp although they only benefit from a 50% or 60% intervention, whereas Antwerp benefits from a 65% intervention.
As regards adult education and arts, worship and libraries (on average 8% of the budget), the expenditure per inhabitant in the Brussels entity is 13% below average. It appears that Brussels is forced to be economical in these less 'vital' areas which Brussels Studies the e-journal for academic research on Brussels 7 7 One of these criteria -called the 'degree of facilities' -is supposed to reflect the presence of administrative, cultural, sports and commercial functions. The Belgian municipality which obtains the highest rating for this criterion is Neuchâteau (DE MAESSCHALCK et al., 2000, annexe 6). Note that the municipalities of Brussels do not obtain the highest score because the international, federal, regional and community administrations are not taken into account in the 'degree of facilities' criterion. Consequently, municipalities such as Ixelles or Saint-Josse are listed as non-urban municipalities. ( VAN HECKE, 1998, p.73) have the characteristics of a 'superior good', i.e. whose consumption increases more than proportionally as revenue increases.
The expenditure item entitled health and hygiene includes hospitals, school medicine and ONE (Childbirth and Childhood Office) consultations, as well as cemeteries and the environment. It accounts for 8% of municipal expenditure in Brussels. In this category, the Brussels entity spends 251% per inhabitant with respect to the average of the five entities.
The biggest item in this section concerns public hospitals. The density of public hospitals financed by the municipalities is much greater in Brussels than elsewhere (11 public hospitals in 6 municipalities). This explains the extreme position of Brussels. In addition, more than a quarter of the patients in these hospitals are not inhabitants of Brussels and therefore participate only partially in the financing of these infrastructures, with the rest being covered by the inhabitants of Brussels.
Apart from hospital expenditure, health and hygiene expenditures remain clearly above the average. The special situation of the municipalities of Brussels may also provide an explanation: the birth rate is relatively high, school medicine concerns a greater number of students, and the standard of living and sociocultural origin of inhabitants leads to higher consumption (of ONE consultations, for example).
Secondary education accounts for 7% of the budget of the municipalities of Brussels and costs 20% more than the average of the five urban entities per inhabitant.
Although at first it would appear that the municipalities of Brussels are not very economical in this area, an analysis of expenditure per student changes this image radically. Municipal expenditure per student is 30% below the average of the five entities.
Expenditure per inhabitant is high due to the large proportion of students in the municipal public system (between two and five times more than in the other entities) and a greater proportion of students who commute to school and who are not inhabitants of Brussels (24% of the student population compared with an average of less than 10% in the other entities, primary and secondary levels combined). Faced with these constraints, the expenditure per student had to be reduced.
Before ending this section on municipal expenditure, let us have a look at expenditure in the area of rubbish collection and waste treatment, as well as fire services and emergency medical assistance. These services were regionalised in the Brussels Region but are still provided by the municipalities in the other urban entities. It is interesting to compare the regional expenditure in Brussels with the municipal expenditure in the other entities.
The rubbish collection and waste treatment services cost the Brussels Region the equivalent of 6% of the municipal budgets and are 11% higher than the average in terms of euros per inhabitant, yet are below the expenditure per inhabitant in Ghent.
Given that these services benefit inhabitants as well as other users of the city (waste from offices, restaurants, hospitals, shops, etc.), it is useful to make a comparison which takes into account the presence of 350,000 commuters per day in the Brussels territory. In terms of expenditure per 'inhabitant + commuter', Brussels is about average, at 99%.

Brussels Studies the e-journal for academic research on Brussels
The same exercise may be carried out for the fire and emergency medical assistance services, whose regional cost amounts to 4% of the total budget of the municipalities of Brussels. In terms of expenditure per inhabitant, Brussels obtains a good score, at 10% below the average of the five urban entities. Since fire services concern dwellings as well as non-residential buildings, a comparison which takes commuters into account is also be useful. Brussels achieves an even better result in this case, at 20% below the average.
Scale economies resulting from the regionalisation of these services could explain the favourable position of Brussels and may constitute a source of inspiration for reforms in other sectors.
By neutralising the effect of the regionalisation of fire and waste treatment services, the municipalities of Brussels spend in total 106.3% of the average per inhabitant. The values for the other urban entities are: Antwerp, 99.4%, Ghent, 108.2%, Liege, 99% and Charleroi, 78.6%.
If the constraints described above are taken into consideration, due to the fact that the population of the Brussels entity increases by more than a third during the day, and the fact that the city's competences are not handed over to a province or to many intermunicipal companies, we may deduce from this second part of the comparative analysis that the relative expenditure per inhabitant of the municipalities of Brussels is certainly not excessive -on the contrary.

Main conclusions
Although the objective of the comparative exercise was to identify the deficiencies in the management of the municipalities of Brussels and to suggest reforms, the conclusion of our analysis unintentionally resembles a plea for a better consideration of the needs of the municipalities of the Brussels-Capital Region. We were not able to identify a local sector of competence with significant shortcomings, apart from the health and hygiene sector which certainly deserves a more in-depth analysis. On the contrary, the management constraints which the municipalities of Brussels are subjected to are what clearly emerged.
Our methodology does not allow us to affirm that the services of the municipalities of Brussels could not be provided more efficiently and for a lower cost. This would require audit-related work to be carried out. It does, however, allow us to affirm that in the case of excessive spending by the municipalities of Brussels, it would not be more excessive than that of the other entities studied, especially if we consider the fact that the other municipalities are able to hand over part of their competences to the provinces or to a higher number of intermunicipal companies.
The financial handicaps which the Brussels entity is faced with at municipal level are quite real. The effect of this situation is, on the one hand, lower expenditure in certain areas such as household and company grants and social assistance other than that financed by the CPAS grant, as well as in the areas of the arts, adult education, libraries, sport, etc. The awkward financial position of the components of the Brussels entity, on the other hand, results in management measures aimed at limiting expenditure and at improving economic efficiency. This translates in limited opera-Brussels Studies the e-journal for academic research on Brussels tional expenditure or to achieve the lowest cost per student in secondary education. Certain choices of the Brussels Region also contribute to reducing expenses for the municipalities. For example, the Region took over certain debts of the municipalities in order to relieve their burden in terms of interest charges.
One cannot deny that efforts have been made to compensate for the financial handicaps in Brussels by controlling expenditure, but there is less room for improvement today: a decrease in the services offered or an increase in taxes would only worsen the phenomenon of urban flight of the middle and upper classes, and would trigger a vicious circle, thus jeopardising the current balance. Furthermore, the institutional reforms which are the topic of much discussion lately are not likely to be able to compensate effectively for the identified handicaps, in particular due to the fact that they are for the most part exogenous.
Reforms would no doubt allow a reduction of possible inefficiencies -as long as they can be identified. The regionalisation of certain competences, for instance, could benefit from scale economies. Their main advantage would be a more equal distribution of expenses between the 19 municipalities. But we do not see how the reforms which are covered in the media or in political declarations would compensate for the lack of solidarity on behalf of the surrounding area, reduce the number of school or hospital commuters, lessen the capital's appeal among low wage earners and in particular among asylum seekers, or improve the financing mechanisms of police areas and CPASs.
Until today, the municipalities of Brussels -like the other urban entities apart from Charleroi -have been able to maintain a positive accumulated budget balance overall. But the perspectives are bleak. The annual aggregate balance of the 19 municipalities has been negative since 2004, and sooner or later the accumulated capital reserve will be depleted.