2. Putting public finances on a sustainable path.
47. Population ageing will impose a budget cost of 3.4% of GDP by 2030 according to the official projections. This cost reflects higher outlays for pensions and healthcare net of reductions for some other outlays (e.g. unemployment benefits) (See Table 1.7). In 2030, the share of public pension outlays and healthcare expenditure in GDP will increase by respectively 2.8 and 2.4 percentage points from their levels of respectively 9.2 and 6.9% of GDP in 2003. These increases are relatively modest by international standards.
48. Some specific features of the first pillar pension scheme (1) account for this moderate growth in public pension outlays. Pension benefits are related to average career earnings, but the annual earnings taken into account are subject to a relatively low ceiling. Between 1982 and 1998, this ceiling was only adjusted in line with movements in the CPI, implying that a larger fraction of earnings will be affected by the ceiling. As from 1999, the ceiling is indexed to the evolution of conventional wages. (2) Past earnings are converted into present values using the CPI and a work history of 45 years is required to qualify for a full pension, otherwise the pension benefit is adjusted proportionally to the numbers of years worked (or equivalent) divided by 45. The result is a gross replacement rate--defined as the pension benefit relative to the gross salary earned in the year prior to retirement--for a newly retired single man who has always earned the average wage in the scheme of 36.5% in 2002 (Table 2.1). This rate is projected to remain roughly constant until 2030. With the encouragement of collective sectoral pensions since 2003, second pillar pensions are projected to grow in importance, raising total pension replacement rates over time. In addition, pension replacement rates look more generous when related to net income, reflecting the high burden of taxation on labour relative to pensions. Apart from some discretionary welfare adjustments for older and low pensions, pension benefits are linked to the "health index" (3). But there are no automatic real increases provided by the law. As a consequence, the purchasing power of a pensioner falls behind that of an active person as he/she gets older. The ageing projections include an assumption that there will be political pressure to discreetly raise the real value of pensions (See also Annex 2 A1 for key assumptions underlying the projection).
49. It needs to be emphasised that the projection entails a significant slowing in the real growth rate of public healthcare expenditure from 4.5% per year during the period 2003-07 to 2.8% on average between 2009 and 2030, which explains the moderate increase in the share of healthcare expenditure in GDP over the projection period. The estimate of long-term trend growth of public healthcare expenditure is obtained using a methodology developed by the Federal Planning Bureau for this purpose. In this method, health care expenditure is broken up into a demographic component and a historical component. The demographic component takes into account the evolving age-profile of the population (including population growth) as well as the statistical relationship between per capita consumption of health care services and age, based on an expenditure profile by age and sex for 1997. Given the empirical findings that healthcare spending on an elderly person (age 70) is 3.5 times higher than spending on a young person (aged 30) and healthcare spending on a frail elderly person (aged 90) 12 times higher, a larger proportion of elderly and frail elderly in the population will drive up expenditure on health care services. The demographic component has been estimated to increase health care expenditure by 0.8% per year on average between 2009 and 2030. The non-demographic component is estimated empirically by a linear expenditure model with a constant marginal propensity to consume and a variable elasticity of per capita healthcare expenditure to per capita GDP, which is greater than 1 in the short run but converges to 1 in the long run. It was felt that this model fitted the observations better than a model which assumes a constant elasticity (4). According to this estimation procedure, non-demographic factors will drive real health care expenditure up by 2.1% annually on average. Adding the two components together yields a long-term average growth rate of 2.8% between 2009 and 2030. However, healthcare expenditure growth is estimated at 5.1% annually for the period 2003-07 owing to measures taken by the federal government in 2003, more specifically an upward revision to 4.5% of the aggregate cap on health expenditure growth until 2007 and the compulsory inclusion as of January 2006 of health insurance against small risks for independent workers into the public package, resulting in an extended coverage for independent workers and a one-off increase in public healthcare expenditure. Combining the two sub-periods and assuming a real growth rate of 3.5% in 2008 yields a long-term average growth rate of 3.2% for 2003-30.
50. Such projections are inherently uncertain. Predictions about the long-term growth rate of health care expenditure are in any case difficult to make because technology, the main non-demographic factor driving expenditure growth, is difficult to model and therefore not explicitly modelled here. Technological progress in the medical sector takes the form of new and/or better treatments, for which there is a growing demand as the population ages. Supply constraints may then produce steeper price increases than in the past. However, it is important to achieve the projected slowdown in healthcare expenditure, which requires the continuous implementation of structural reforms that help contain spending (see Chapter 3), because estimates of the budgetary impact of ageing are very sensitive to the hypotheses about the average long-term growth rate of healthcare expenditure. The High Finance Council (HFC) presents an alternative scenario in which the long-term growth rate of healthcare expenditure after 2008 is 1 percentage point higher than in the baseline scenario (i.e. 3.8% instead of 2.8%). This leads to a 2 percentage point of GDP increase in the estimated cost of ageing relative to the baseline scenario.
51. Finally, it is worth noting that the cost of ageing is determined completely by changes in age-related spending programmes as the share of tax revenues in GDP is assumed to be unaffected by ageing. More specifically, the model uses a Cobb-Douglas production function, which is characterised by constant shares of labour and capital income in GDP. This option is a classical one for long term projections.
52. In view of its high public debt (100% of GDP at the end of 2003), Belgium finds itself in a position whereby it can finance the expected cost of ageing through a budgetary policy aimed at a gradual build-up of the budgetary surpluses in the medium run which would then be run down in the longer term as debt interest payments fall. This could speed up the positive dynamic of lower public debt and lower interest charges. The budgetary margin created by lower interest charges could then be used to finance the initial cost associated with population ageing. When the costs of ageing begin to increase more rapidly (after 2015), this strategy of debt reduction can be combined with a gradual reduction of the surplus in order to cover the additional cost without having to raise future taxes and/or cut spending programmes. More specifically, the HFC recommends the following long-term budgetary strategy, which would ensure that public finances remain on a sustainable path.
--The short-term objective would be to transform the current budget balance into a surplus of 0.3% of GDP in 2007 and 0.6% of GDP in 2008 according to the latest stability programme (2005-08) adopted by the Belgian government on 3 December, 2004;
--Next, the government would increase its overall budget surplus by on average 0.3% of GDP each year to reach 1.5% of GDP in 2011; half of the improvement would come from a further fall in interest rate payments and the other half from an increase in the primary surplus (HFC, 2004). This high surplus should be maintained until 2018, but declining interest charges imply that the primary surplus can be reduced gradually;
--After 2018, the government can bring down the surplus gradually to be back at zero in 2030. In this period, the primary surplus can also fall more rapidly, reflecting both lower interest charges and a relaxation of the overall budget target. In this scenario, public debt would fall below 30% of GDP by 2030.
53. In order to bring the long-term strategy with respect to ageing described above to a happy conclusion, the government must successfully meet the following challenges.
--First, it must reach a structural surplus (5) of 0.3% in 2007; this cannot be done without the implementation of consolidation measures in 2005-07. This short-term challenge will be further elaborated in the next section.
--Second, long-term growth of healthcare spending must be contained through a package of structural reforms to bring growth rates into line with the projections. This challenge will be covered more in depth in chapter three.
--Third, given that the government cannot use the created budgetary margins in the short run to finance further tax cuts on labour income, needed to increase the labour utilisation rate in line with the projections, additional budgetary margins must be created by improving the efficiency with which the government meets its objectives. This will be discussed in the second half of this chapter.
The short-term challenge: achieving a small surplus by 2007
54. In contrast to the worsening fiscal deficits observed in most countries in the euro-area, the Belgian general government budget has been in balance (or slightly above) from 2000 until 2003. A balance was achieved for the first time in 2000 after a decade of budgetary consolidation during which a high net primary surplus--defined as the difference between the general government balance and net interest payments--was steadily built up, reaching 6.8% of GDP in 2001 (Table 2.3, Figure 2.1). Since then, the net primary surplus, as well as the cyclically adjusted net primary surplus, has started to decline. The government has been able to maintain a budget balance thanks to the use of non-recurring measures and a decline in interest charges, reflecting both lower interest rates and a reduction in public debt. (6)
[FIGURE 2.1 OMITTED]
55. The net primary surplus dropped by 1 percentage point in 2002, three quarters of which could be attributed to the economic slowdown and the remaining part to a smaller positive impact of one-off factors. As a result, the structural net primary surplus--defined as the net primary surplus corrected for the impact of the business cycle and non-recurrent measures--did not change. However, it deteriorated sharply in 2003, namely by 1.1 percentage point. The strong positive impact of non-recurrent factors amounting to 1.5% of GDP led to a small budgetary surplus of 0.3% of GDP--which exceeded the corresponding objective of a budget balance included in the Stability Programme 2003-05. The most striking one-off measure was the large capital transfer (7) from Belgacom, which was only partially offset by advancing part of the operational subsidies and all investment subsidies to the national railway company (NMBS/SNCB).
56. The structural deterioration of the fiscal position can be explained by an acceleration of primary expenditure growth on the one hand, and tax reforms lowering the burden of taxation (on labour in particular) on the other hand. Corrected for the impact of one-off measures, primary expenditure grew by slightly more than 3% in real terms in 2002 and 2003, which is well above the long-term average growth rate of 2.1% (National Bank of Belgium, 2004a). Federal spending was driven up by the need to improve security (reform of the police forces), to invest more in rail infrastructure and to modernise the federal government (Copernic reform). Public healthcare spending was contained in 2002, but has been allowed to grow at an annual real rate of 4.5% between 2003 and 2007 as part of the coalition agreement of the national government and has grown rapidly since. The introduction of the more generous time credit scheme, which will gradually replace the old system of career interruption, and discretionary real increases in selective government transfers--older pensions in the scheme for private sector employees, all pensions in the scheme for independent workers and the guaranteed minimal pension, transfers to disabled people and people on social assistance--have contributed to higher growth of social security outlays. On the revenue side, the federal government has continued with the implementation of the multi-annual labour income tax reform and the reductions in social security contributions, and has also abolished the supplementary crisis tax on labour income. The impact of these measures on total revenues has been somewhat attenuated by higher indirect taxes in 2003 and higher supplements to the labour income tax imposed by the local governments. In addition, the Flemish region also introduced some tax reforms in 2002, leading to a lowering of registration taxes, the abolition of the special contribution for public broadcasting, and a one-off personal income tax rebate. Taken together, these measures caused the share of tax revenues in GDP to drop by 0.7 percentage points in 2002 and a further 0.4 percentage points in 2003.
57. Some of the same factors underlying the sharp deterioration of the structural net primary surplus in 2003 continue to exert downward pressure on the structural net primary balance in 2004, which is projected to fall to 4.4% from 4.7% of GDP in 2003. However, the structural deterioration has been contained by a slowdown of total primary expenditure growth (corrected for one-off measures) to a long-term growth rate of 2.1%, a lower impact of the personal income tax reforms and the fact that the reduction of social security contributions (0.2% of GDP) has been compensated by indirect tax hikes. According to the latest forecasts of the OECD (OECD Economic Outlook, No. 76), a small budget deficit of 0.1% of GDP re-appeared in 2004. However, according to the Excessive Deficit Procedure methodology used in the Stability Programme, (8) the Belgian government succeeded in attaining its target of budget equilibrium in 2004. The main reason for the deterioration in the budget in 2004 is that it now benefits much less from the favourable impact of non-recurrent measures. The latter, including a change in the calendar of subsidy payments to the NMBS/SNCB alleviating expenditure in 2004, real estate sales by both the federal and the Flemish government and a fiscal amnesty measure (Declaration liberatoire unique, DLU), have also been lower than expected. (9) However, this setback is offset by more buoyant tax receipts, in particular from indirect taxation, reflecting stronger than expected economic growth. Although to some extent corrective measures have been implemented, an overrun of healthcare expenditures of to 634 million (0.2% of GDP) is expected for 2004 (10).
58. The objectives laid down in the most recent Stability Programme (2005-08) aim for the maintenance of a general government budget balance for 2005 and 2006, turning into a small surplus of 0.3% of GDP in 2007. Assuming that the output gap will be closed by 2007 (so that the cyclical component of the budget is then zero) and the disappearance of non-recurring measures, the structural balance--defined as the actual balance corrected for the impact of the business cycle and nonrecurring measures--will also be about 0.3% of GDP. Moreover, assuming that net debt interest payments will remain unchanged in 2007 (Federal Planning Bureau, 2004), the corresponding net primary surplus target will be 4.4% of GDP. Meeting the budgetary targets for 2005-07 will be challenging as simulations have shown that, if the consolidation objective is to be achieved without raising taxes, real primary spending growth should be about 1.25 percentage points lower than the growth of real output (Table 2.4), given the assumptions that:
--The impact of the current one-off measures--0.7% of GDP in 2004 and 0.4% of GDP in 2005--will unwind in 2006.
--Short-term market interest rates will remain unchanged and that long-term rates will go up to 4.5%, such that the implicit interest rate on government debt continues to decline to some 4.6% by 2007 and 4.4% by 2011.
--Public debt is increased by 2.5% of GDP in 2005, following the takeover by the Belgian government of the historical debt of the NMBS/SNCB of 7.4 billion [euro] as part of the restructuring of the national railway company.
--Additional revenue cuts have already been planned, amounting to close to 1% of GDP in the period 2005-07. These include the continued reform of the personal income tax system and additional reductions in social security contributions for employers and employees, which are only partly offset by increases in indirect taxes, mainly excise taxes on tobacco and mineral oils.
--Apart from the planned measures, taxes will grow in line with GDP and the deflator of GDP and primary expenditure will grow at the same rate.
59. In a reasonable scenario with economic growth averaging around 2.4% in the period 2005-07, primary spending can only increase by 1.2% a year at constant prices, which is low by historical standards. The spending margin becomes even lower if one takes into account the allowed growth norm for total health care spending of 4.5% a year at constant prices until 2007 and the future spending increases decided already in the course of 2004 in the so-called Gembloux and Ostend packages. The burden of consolidation will fall on entity I (federal and social security) considering that this is where the deterioration was most pronounced in the first place and that the benefits of declining interest charges accrue almost exclusively to entity I. Therefore, the outlays of entity II should continue to grow at their long-term average rate of 2.1%. In view of the spontaneous growth in social security spending--2.2% on average--federal expenditure would need to shrink in real terms in order to meet the budgetary objectives in the absence of additional tax increases or one-off measures.
60. This is not the path chosen by the government in its budget for 2005, which allows for a real increase in federal government spending of 1%. Earlier commitments to budgetary increases for the departments of security and justice and finance (ICT investment), investment funding for the NMBS/SNCB and cooperation with Brussels have been honoured, but all other departments and programmes are subject to economies (freeze of the personnel envelopes and real reductions in working capital among others) and the government expects to obtain a downward revision of its contribution to the EU budget. One-off measures, more specifically the sale of real estate and the securitization of tax arrears, (12) will again improve the budget by some 0.3% of GDP (Table 2.3). The budget also benefits from new tax increases (higher taxes on packaging waste) on top of earlier measures raising excise taxes on tobacco and mineral oils, some tax-base broadening measures (a greater part of the value of company cars will be subject to employers' social security contributions), the positive impact of the European Savings directive as of July 2005, and continued efforts to combat fraud. Moreover, the better than expected economic growth in 2004 will have some positive carry-over effects for the budget. Even though the government estimates that the measures taken are sufficient to close the gap of 4 billion [euro] needed to balance the budget in 2005, the OECD thinks that it may be necessary to take further measures amounting to 0.4% of GDP in order to preserve an equilibrium.
61. The budgetary outlook for 2006 is particularly bleak, as the impact of the personal income tax reform reaches its peak, lowering the share of tax revenues in GDP by another 0.6 percentage point of GDP on top of the growth norm of the public healthcare budget by 4.5% in real terms. In the absence of any further measures, the structural net primary balance will fall to 3.5%, which is 0.9 percentage points below the target for 2007. Although the temptation to resort once again to non-recurrent measures will be strong, the federal government should take advantage of the economic upturn to focus on structural measures that permanently reduce government spending in combination with measures that reinforce the budgetary process (Box 2.2) This would reduce the risk of a pro-cyclical fiscal policy which would intensify business cycle fluctuations at a later date. Moreover, recent one-off measures, such as the Belgacom operation which generates additional pension payments for the government or the frequently used real estate sales and rent back operations, are self-reversing implying the short-term gains are offset by a stream of outlays in future years. If assumed to be actuarially neutral, self-reversing one-off measures do not modify the inter-temporal budget constraint of the government, implying that they do not help in restoring the long-term sustainability of public finances.
Box 2.2. Measures to reinforce the budgetary process In order to achieve the objectives laid down in the Stability Programmes, the Federal Government, the Regions and the Communities concluded a multi-annual cooperative agreement for 2001-05, in which each committed itself to respect a pre-determined path for the budget balance. Regions and Communities are in principle free to achieve their budget targets by adjusting outlays or taxes. (1) During the fourth quarter of each year, there is extensive monitoring by the Federal Government in cooperation with the Regions and Communities to ensure compliance with the objectives. There are no formal sanctions in the event of non-compliance, but it is felt that this arrangement has contributed towards the respect of the general government budget targets, and a new cooperative agreement for the period 2006-10 should be concluded soon. Improved budgetary mechanisms to ensure fiscal discipline are especially important at the level of the Federal Government, which is under most pressure to consolidate. It is at this level that most initiatives are being developed. The anchor principle introduces a monthly monitoring of expenditure in each Federal Department to make sure that Departments respect the historical under-utilisation rate by some 2% of the budgetary credits allocated to finance spending programmes excluding personnel expenditure. Given that wage increases are quite uniform across departments and that public sector wages are fully indexed, the timing of which can be predicted with reasonable accuracy, the Budget Department has the ability to set a global envelop for increases in the wage bill, which can then be divided between departments. This implies that the different departments have only limited autonomy with respect to personnel management. The anchor principle is effective in avoiding that budgetary windfalls are spent in the year in which they occur and that growth ceilings are complied with, but it does not stop spending pressures from piling up (especially in the face of windfalls) which will eventually cause the breakdown of spending constraints. The focus on policy priorities can be strengthened further by systematically putting into practice the existing clause that any proposition for new initiatives must be accompanied by compensation measures. The current practice is still that the departments prepare proposals for new initiatives with an estimate of their cost and then bargain for additional budgetary resources (bottom-up budgeting). Choices are made in accordance with the policy priorities of the Federal Government. The low emphasis on compensatory measures can be partly attributed to the absence of a systematic review of all spending programmes, which would indicate the areas in which savings can be made. The government should gradually move from bottom up to zero base budgeting. The Federal Government intends to move towards performance-based budgeting (PBB) by 2007, but still has some way to go. Since 1986, the publication of the annual budget is accompanied by a justification, making explicit the overall policy orientation of the Federal Government, the missions assigned to the different departments and programmes, the objectives aimed for and the means devoted towards the achievement of each objective. Nevertheless, the quality of the budget justification could be improved as it is currently mainly descriptive with few quantitative indicators to measure outcomes. This makes it difficult to evaluate ex post to what extent the objectives have been achieved. However, international experience suggests that it can be challenging to find a limited number of quantitative indicators that adequately capture the variety and complexity of the tasks performed by a department and poor performance indicators can reduce the quality of the public service when efforts are shifted into tasks whose outcomes are easy to quantify at the expense of equally important tasks that are hard to quantify. Some federal departments are currently in the process of developing quantitative indicators for their services. Finally, the Federal Government aims to strengthen the analytical ability to measure the cost effectiveness with which the objectives are achieved. In view of the existing budgetary constraints, costs are taken into consideration by the different departments, but this is not done in a standardised way. In the framework of a new accounting system, to be implemented progressively from 2005 on, a limited analytical component will be available to calculate the costs of government policies. In addition, the Federal Government wants to create a knowledge centre with a mission to elaborate more sophisticated methods to estimate the costs of public policies. (1.) Regions enjoy more tax autonomy (registration taxes, inheritance taxes, real estate taxes). However, communities have no such freedom as they have no tax autonomy in practice. (2.) All proposals for new initiatives need to be approved by the Inspection of Finance.
Enhancing the efficiency of public spending
62. Increasing the efficiency with which the government achieves its objectives is one way in which the government can make room for future reductions in taxes on labour income without undermining the sustainability of public finances, thereby enhancing incentives to work. Current and prospective measures in the healthcare sector will be discussed in the next chapter. Some reforms to raise the efficiency of public administration have already been taken with the aim of improving the quality of public service provision within a given budget, and further reforms are planned to modernise the public administration and to implement the e-government strategies of the federal government, the regions and the communities. This will lead to an innovative and modernised administration with faster delivery of services. It should also create scope for the realization of cost savings and productivity gains in the public sector over a longer period of time. For example, reductions in the share of government employment in total employment could be achieved mainly by the partial replacement of departing employees.
Reducing inefficiencies in public administration
Reducing public sector employment
63. Belgium is characterised by a high level of public employment (excluding education) relative to the total population (Figure 2.2), suggesting there may be some scope for cost reductions. Successive constitutional reforms since 1969 have resulted in the creation of three Regions and three Communities next to the federal level and the local level (which includes provinces and local communities). This process has been accompanied by an expansion of public services and employment. Traditionally low mobility and life-long employment in the public sector adds to the problem by making public spending behaviour asymmetric: public employment tended to be adjusted upwards when new demands emerged, but failed to be adjusted downwards when demand waned because services are rarely reorganised or abolished. Downsizing will become easier soon, when baby boomers start retiring in large numbers. The government should seize this opportunity to develop a long-term vision with respect to the partial replacement of retiring employees by higher qualified staff and the re-organisation of its services.
[FIGURE 2.2 OMITTED]
64. In all OECD countries, the proportion of high-skilled employment in the public sector is higher than in the private sector, especially in departments responsible for policy-making, (13) and has increased over time. Skill upgrading has exerted upward pressure on the wage bill of the public administration. In addition, pension arrangements in Belgium are more generous for statutory civil servants (statutaires) than for private sector employees. Statutory public sector employees are covered by a final-wage defined benefit scheme, with pensions based on the average wage earned during the last five years and the number of years of public sector employment, whereas private sector employees and contractual public sector employees (14) are covered by an average-wage defined benefit scheme in which earnings are subject to a ceiling. As a consequence, the replacement rate is higher in the public sector pension (15) scheme than in the private sector pension scheme. Pension arrangements in the public sector are viewed as deferred wage payments and this is reflected in lower public sector wages.
65. One key challenge in raising the efficiency with which the government reaches its objectives is to enable the government to respond with greater flexibility to changing public-service demands through an internal reallocation of resources. Barriers to labour mobility within the public sector should be reduced. Selor, the centralised recruitment office of the federal government, has been formally given a mandate to organise an internal job market, but it still needs to be put in practice. A better internal communication strategy about the existing employment possibilities in different departments could be developed by centralising all vacancies in one database to which all civil servants have access. Departments should be obliged to report all vacancies to Selor and be encouraged to make greater use of internal transfers as opposed to external recruitment. Measures to extend the internal labour market to include the other levels of government should also be considered. A well functioning and large internal labour market would also offer advantages to departments with overstaffing which can make use of it to locate new employment opportunities within the public sector for redundant employees. In addition, labour mobility and employees' adaptability can also be enhanced through improved training opportunities and stronger incentives for life-long learning.
66. There is also scope for raising productivity in the public sector by modifying payment schemes, which have traditionally been based on formal qualifications and seniority, providing little incentive for high performance. The Copernicus reforms (16) of the federal government aim to break with this tradition by basing remuneration more on the possession and development of competencies during a career in the public sector. The introduction of competency premia should therefore be coupled to a reduction in seniority premia. Competencies refer to skills, abilities to learn and personal traits, and all functions will be redefined in terms of required competences. A system of development circles is expected to be introduced in 2005. This means that every supervisor is obliged to hold a meeting with each of his employees individually during which information about the function is exchanged, performance objectives set, training needs identified, and progress towards achieving the objectives discussed (in subsequent rounds). Such a system would allow outstanding performance to be rewarded and shirking to be sanctioned, although problems with specifying performance targets (outputs and outcomes are not always easy to measure) and inability or unwillingness on behalf of supervisors to differentiate strongly between employees have been found to reduce the effectiveness of such arrangements in practice (Joumard et al., 2004). Training and continued learning are viewed as essential building blocks in the development of competencies, and participation in training programmes is the right and duty of each employee. The federal government has committed itself to raise the amount spent on training from the current 1.2% of the total wage bill to 1.9%. In addition, life-long learning could be stimulated further if promotion prospects and not only salaries took into account the accumulation of skills and the willingness to use them.
Improving senior management performance
67. Productivity gains in the public sector can only be fully realised when supported by good leadership that stimulates public sector performance. This requires a shift towards a new management approach, focused more on results instead of compliance with procedures. Senior management reforms have been implemented at the level of the federal government, and Regions and Communities, but the following paragraphs reflect mainly the procedures and experiences of the federal government (Regions and Communities have implemented similar changes but started at a different time).
68. One of the main objectives of the Copernicus reforms was to appoint executives with proven capabilities in the civil service or in the private sector. Vacancies for high-level management functions were openly advertised and the recruitment decision was based on an assessment of basic competences by an independent head hunter firm, an assessment of specific knowledge by Selor and an interview with the ministry interested in hiring. In order to limit the number of appeal cases (17), the first assessment has been re-assigned to Selor, with the help of external consultants. Contracts are given for a fixed period of six years and are renewable (but the person must re-apply). Within three months of appointment, each top-manager and his consultants must present the minister with a management plan, which must include a precise definition of the duties, his strategic objectives, the operational objectives and the budgetary resources. Based on this plan, lower level managers work out operational plans for their own departments. Performance targets should be carefully assessed, because experience in other countries has shown that poorly formulated targets can produce adverse effects.
69. Evaluations take place every two years, with the possibility of a premature termination of the contract if performance is found to be really unsatisfactory. In this case, the procedures laid down in the labour code must be respected. A manager whose contract expires after six years is not automatically re-appointed: he/she must re-apply and get through the entire recruitment procedure again unless his/her final assessment note has been "highly satisfactory". Managers do not receive a performance-related premium.
70. In exchange for greater responsibility, the salaries of top-managers have been raised substantially and autonomy has been somewhat increased. High salaries were also viewed as necessary to make the public sector more competitive and to attract proven managers from the private sector. The high remuneration for top-managers does not burden the public pension scheme because all managers, including those recruited from within the civil service, are hired on a contractual basis implying that they are subject to the same conditions as employees from the private sector. In addition, top-managers were offered an additional (second pillar) pension plan, comparable to what exists in the private sector, financed by a contribution of the employer and a personal contribution equal to 1.5% of the monthly wage paid by the top-manager. Nevertheless, generous remuneration packages were apparently not enough to solicit a strong interest for these positions from private sector candidates (Table 2.5).
71. Low interest of the private sector (only 15% of total appointments) may be attributed to the lack of autonomy in the public sector, relative to the private sector. For example, managers have limited control over the personnel budget, limiting their possibilities for reorganising a public service. Managers have to make up a personnel plan, which is subject to approval. Once approved, managers can take all the necessary steps to execute it, taking into account the statutory and contractual rights of the personnel. This low inflow of private sector managers was also one factor (18) prompting the new government to moderate wages of top-managers again, although they remain higher than before the Copernicus reform. Moreover, the lack of experience with quantitative performance indicators also makes it difficult to evaluate ex post the extent to which the strategic objectives have been achieved. As long as it remains difficult to hold managers responsible for outcomes, a lower wage can be justified. However, the priority should be to further develop outcome indicators (as planned by the Budget Ministry) to reinforce the analytical underpinning of the reforms. This is already the case in some public services (RVA/ONEM, Federal Public Service of Interior Affairs, Federal Public Service of Mobility and Transport). In addition, the recently (December 2004) approved new management evaluation process is based on the execution of the managerial and organisational planning, which will force managers into the development of performance indicators. Managers have to prepare a self-evaluation. The use of quantitative performance indicators is therefore necessary. This is also needed to give managers more autonomy, so that they can be held accountable for outcomes.
72. Given the limited interest among private sector managers in assuming a management position in the public sector, the government will have to develop managerial experience in-house and make high quality management training compulsory for new managers. It has already launched a training programme without direct access to management positions but designed to help candidates during selection. Dubbed "Pump", the programme enrols about forty people selected after a written test and a personal talk. Training takes several months. It covers both theory and hands-on experience, including a one-month internship in a foreign country. Trainees must also work for three months as a consultant on modernisation projects and propose reform targets.
73. E-government provides a powerful tool to improve the efficiency with which the government reaches its objectives, including by reducing the administrative burden (see Chapter 7). Electronic applications can generate savings on data collection and transmission, especially through greater sharing of data within and between governments and through the creation of portals allowing citizens and businesses to find information and to make online transactions with the government. A State Secretary for e-government and computerisation has been appointed in 2003. His mission is to co-ordinate and to push the ICT and e-government strategies of the federal administration, thereby contributing to its modernisation. This ICT support will also help Belgium to grow into a knowledge society. The key principles of e-government in Belgium are the unique collection of data, the development of reference registers or "authentic" sources and of unique identity keys (the National Register number for citizens, the company number allocated by the Crossroads Bank for Enterprises for businesses). The sharing of electronic databases not only reduces the administrative burden on enterprises, citizens and civil servants, but also increases the quality of the data managed by the government. Through cross-comparison of information in different databases, the government has an additional investigative tool at its disposal for identifying fraudulent cases. By enabling the automation of manual transactions, such as tax filing, resources can be freed up for more valuable tasks such as better enforcement and more frequent inspections. Other benefits are the modernisation of the administration, the development of new innovative services and the reduction of the digital divide by granting rights automatically, based on the status of a citizen / business without his intervention. This avoids developing services at the front office.
74. Belgium has made considerable progress in its implementation of e-government services since 2001and is now firmly positioned within a large cluster of countries with an e-government maturity score between 50 and 60% (Figure 2.3). E-government maturity is defined as a weighted average of service maturity (70%) and customer relationship management (30%). The Accenture survey assigns a score (1, 2, or 3) for service maturity to each of the 206 services considered at the level of the national government and then computes the average, with the three possible scores defined as: (1) = information only with no possibility for interaction, (2) = one-way or two-way interaction but no possibility to complete end-to-end transactions, (3) = end-to-end transactions Customer relationship management measures the extent to which government agencies manage interaction with their customers (citizens and businesses) and deliver services in an integrated way.
[FIGURE 2.3 OMITTED]
75. The Belgian approach towards e-government has been first to develop and integrate back-office functions before rolling out new services to businesses and citizens at the front end. The focus is predominantly on providing better services to customers in the form of easily accessible information and additional channels for interactions and transactions with the government. The most impressive realisations so far have been:
--The completion of important building blocks for e-government, including a vast secured broadband network connecting the agencies of the federal government (FedMAN), and a Universal Messaging Engine realising data communication between heterogeneous systems, portals and websites of the different federal agencies.
--The launch of a federal portal, www.belgium.be, offering citizens, businesses and civil servants with links to information as well as to useful applications such as the ordering of licence plates and filing of personal income taxes online. Accountants can also file the personal income taxes of their customers (such as independent professionals and company owners online.
--The crossroads Bank for Enterprises (Banque Carrefour des Entreprises, BCE), allocating each company a single unique identification number. Via the BCE, government services will be able to exchange information. The single identification number is necessary to go ahead with administrative simplification (see Chapter 7). The unique "starters" form should give the beginning entrepreneur the possibility to regulate his most important administrative formalities at one single place: "the office for enterprises".
--After a pilot phase in 2003, the roll-out of the electronic identity card has started in 2004. At the latest in 2009, all residents will have such an ID card. This smart card with embedded digital signature will pave the way for new applications such as electronic voting, paying taxes, online access to their files with the government, request of documents, and the secure exchange of information with the government. Moreover, the ID card will leverage the use of eCommerce and eBanking. All private companies can use the ID card as an identification and authentication tool for secured online transactions.
76. The implementation of a successful e-government programme requires large, up-front investments, whereas benefits take time to materialise. Belgium is still in a stage where the development of e-government represents a net-cost to the government. A recent survey organised by the State Secretary for e-government shows that ICT expenditures of the federal government (investments in and development of hardware and software) amount to about 407 million [euro], representing 2.5% of the budget of a public service. One reason why benefits are still low is because the volume of electronic transactions with the government is still limited, especially among citizens, whereas higher volumes would clearly generate economies of scale. The allocation of single ID numbers for companies, along with the BCE, should boost the take-up of electronic transactions between businesses and the government. However, there remain several barriers to a higher take-up of e-government among citizens. The fraction of regular internet users in the Belgian population is still modest (about 49% according to the Belgian internet mapping survey), although the government is actively promoting broadband and access to internet. At the end of 2004, there were about 1.99 million broadband internet connections in Belgium. 82% of Belgian internet users has visited a government website in the last 12 months and one out of five citizens has contacted the government by email in 2003. When the facility to file personal income taxes online became available in 2003, 2% of the population (57 000 persons) used it. This number increased to 169 000 in 2004, an increase of 300% in comparison with 2003. The introduction of electronic ID cards for citizens in combination with a greater availability of services will further stimulate take-up, but the government should consider additional measures to promote its electronic services among citizens, such as well targeted information campaigns (for example in schools) or financial incentive schemes.
77. Some e-government projects that are currently being implemented will generate cost savings. For example, the federal administration is promoting the use of open standards for its administration and open source softwares are seriously taken into account. The usage of open standards and the reusability of ICT applications will contribute towards cost reductions for the administration. In addition, a cooperation agreement between the Federal government and the Regions and Communities on e-government commits all parties to the principle of unique data collection using the same unique identification keys and the same data definition. An interoperability framework is currently being developed in order to enable all governments to interconnect with each other, thereby lowering total expenses for administrations through the realisation of scale economies.
78. Another area in which other countries with more mature e-government programmes and a higher citizens' take-up have reported significant savings concerns tax revenue services. These took various forms, such as the automation of some of the manual processing of tax returns, the reporting of additional income, savings on postage, savings resulting from more queries being handled through self-help services, an improvement in tax collection following a re-allocation of freed-up resources towards tax enforcement. The latter advantage should be highly relevant to Belgium, which has a sizeable black economy. Moreover, the Belgian Ministry of Finance has estimated that the implementation of large-scale electronic filing and treatment of taxes would enable it to partially offset the large natural attrition of its personnel in the coming years. Since this could gradually lead to a reduction of its staff by one quarter, the budgetary margin for the hiring of additional tax inspectors within the existing division charged with investigating fraud and with better electronic fraud surveillance could easily be created.
Improving the outcomes of transportation policies
79. Subsidies by the federal government to the state-owned railway company NMBS/SNCB are still high, also by international comparison (Figure 2.4). These numbers need to be interpreted with caution though. Much of the public financial support to the railways is not notified to the European Commission either because the financing, due to the lack of liberalisation of the sector, is not deemed by Members to constitute State aid under the meaning of Article 87(1) of the Treaty, or because it represents compensation for public services in accordance with Regulation 1191/69. Member states are however required to report overall public expenditure to the sector. Disparities between Member states may reflect different interpretations of the scope of this annual reporting exercise.
[FIGURE 2.4 OMITTED]
80. In the case of Belgium, the total amount of public financing reported to the European Commission is less than the total amount of subsidies given to the national railway company, which is about 1% of GDP. The NMBS/SNCB receives public money for four different reasons: i) to finance new investments in infrastructure and rolling stock (0.3% of GDP); ii) to cover the costs of operation of domestic passenger transport (0.1% of GDP); iii) to cover the costs of maintaining the railway network; and iv) to pay pensions of retired employees of the NMBS/SNCB.
81. Operational subsidies to domestic passenger transport and maintenance of the network remain considerable for several reasons. As pointed out in the previous Survey (OECD, 2003, pp. 139), the NMBS/SNCB is characterised by high operating costs, over-staffing and low efficiency, with losses in all three services (freight transport, international passenger transport (Thalys, Eurostar) and domestic passenger transport) as a result. Subsidies are also motivated by the objective of the government to provide a minimum mobility at an affordable price for everyone. In addition, subsidies are used to reduce the price of public transport services, thereby bringing their price relative to that for private transport services closer to relative marginal social costs. Such subsidies are even set to increase from January 2005, when the government will pay 20% of a rail pass for the section between the private sector employees' residence and workplace, provided that their employers pay the other 80% (such train tickets are already free for public sector employees). However, the question is whether operating subsidies for passenger rail transport are not a very efficient way of ensuring that the relative prices of public and private transport reflect marginal social costs as they encourage excessive mobility by making transport cheaper relative to other goods and services, leading to an increase in overall traffic. Similarly, the other instruments aimed at internalising the external costs of private transport (excise taxes on motor vehicle fuels and taxes on motor vehicle purchases and ownership), the greatest of which is congestion costs, might be less efficient as they are poorly targeted. Economic theory suggests that the best solution consists of internalising congestion costs through road pricing and targeting environmental externalities through taxes on road fuels. In this way, the relative price of public and private transport could be brought into line with marginal social costs without the need for large subsidies. Important progress in this direction has been made recently by a decision to shift taxation from vehicle ownership to use. Excise taxes on fuel are gradually being raised, making the use of a motor vehicle more expensive, whereas the fixed taxation of gasoline vehicles is being phased out (19) and the immatriculation tax is set to disappear as well, making the ownership of a motor vehicle less costly. Fuel taxes are more effective at internalising congestion externalities (Chia, Tsui and Whalley, 2001) than vehicle ownership taxes, so they are currently used to tackle both environmental and congestion externalities. However, the government should consider the feasibility of introducing a road pricing system as soon as it becomes technically feasible and reliable, reducing the need for public transport subsidies insofar as they correct for an inappropriate pricing of road use and refocusing road fuel taxes on environmental externalities (including GHG emissions) only.
82. However, the scope for saving on investment subsidies will be limited because Member States are under pressure to promote railway transport as an environmentally sustainable alternative to private transportation in view of the growing demand for mobility and the need to comply with the Kyoto protocol. (20) This will require additional investments to modernise and extend the public infrastructure network. Public involvement in rail infrastructure projects will remain important because it is very difficult for any private investor to secure a profit on such projects, although public-private partnerships (PPP) could be considered as one amongst a wide range of options, as the European Commission has recently done in the context of the Trans-European Network. In order to accelerate major railway projects, the Belgian government is currently investigating a PPP-formula for a number of infrastructure projects, such as Diabolo (railway access to the National Airport), the modernisation of the Brussels-Luxembourg line and the Brussels portuary terminal. A recent study (Friederiszick, ROller and Schultz, 2003) found that the level of aid has a positive impact on railway efficiency, although aid intensity--defined as aid divided by total operating cost--has a negative impact. This result suggests that aid must be complemented with other means of finance to be effective. Other potential advantages of PPPs are: i) the introduction of private management skills while sustaining a public service ethos within an organisation, ii) lower investment costs to the public sector as the project is co-financed by the private sector, and iii) the potential for efficiency gains in delivery. However, the empirical evidence suggests that most efficiency gains stem not from the private sector involvement as such, but rather from the permanent exposure of potential contractors to competition (Van Den Noord, 2002). In other words, savings are unlikely to be maintained in the long run if the PPP establishes de facto a monopoly position for the incumbent private partner. Another drawback is that a PPP entails a greater exposure of the government budget if the private partner fails.
Box 2.1. Public finance recommendations * Public finances should be put on a sustainable path--defined as one where government programmes can be financed indefinitely at constant tax rates and public debt eventually stabilises as a share of GDP (as a euro area country, this also must be less than 60% of GDP)--so as to avoid future large increases in taxes--and hence in the excess burden of taxation- or deep cuts in social programmes when ageing-related budget pressures rise. This objective could be achieved by increasing the structural budget balance to a surplus of 0.3% of GDP in 2007 and allowing it to rise to 1 1/2 per cent of GDP in 2011 and maintaining it at this level until 2018, as recommended by the High Finance Council. * To this end, the OECD estimates that the government should take consolidation measures, to increase the structural budget balance (abstracting from non-recurring items) by about 1% of GDP by 2007. In view of Belgium's already high tax burden and the adverse effects that this has on economic activity, the priority should be given to expenditure restraint in budget consolidation, and non-recurrent measures should be avoided. * In the medium run, savings in government administrative expenditure could be achieved at all levels of government. Productivity gains can be obtained from greater mobility, the promotion of life-long learning and the development of competencies during the career, the introduction of a new management culture and the implementation of e-government. --Labour mobility should be increased through the creation of a job placement office within the public sector. --Competency premia should be introduced and seniority premia reduced. --High quality management training should be made compulsory for prospective managers. --Managers need to be given more autonomy so that they can in fact be held accountable for outcomes. To this end, outcome indicators (needed to measure performance in a more autonomous system) should be further developed to reinforce the analytical underpinning of the reforms. --The take-up of e-government services by the general public should be promoted in order to reap the benefits of scale economies. * Transport policies should target pollution externalities and congestion costs more efficiently. Recent measures shifting taxation from vehicle ownership to fuel consumption go in the right direction. Moreover, the governments should consider the feasibility of introducing a road pricing system as soon as it becomes technically feasible and reliable, targeting fuel taxation on pollution externalities alone and reducing the need for transport subsidies insofar as they are currently motivated by an inappropriate pricing of road use. ANNEX. 2.A.1. KEY ASSUMPTIONS 2003 2030 Employment rate Global 61.5 68.5 55-64 32.6 42.4 Women 53.0 63.9 Participation rate Global 71.1 73.1 55-64 43.7 50.4 Women 63.8 68.4 Structural unemployment and productivity Productivity and individual wage increases 1.75% after the medium term period (2009) Structural unemployment rate (1) 7.5% (in 2030) Other Welfare adaptation of Pensions (2) 2 scenarios: 0.5% or 0% annually Increase wage ceiling (3) 1.25% annually Real interest rate 4% after 2009 2003-2030 Healthcare expenditure (growth rate) Total expenditures 3.2 Trend evolution 2.3 Demographic effects 0.9 Ageing 0.7 Population volume 0.2 (1.) Includes the older unemployed not looking for work. (2.) Pensions are automatically adjusted for inflation. In addition, the ageing commission assumes in its baseline scenario that there will be political pressure for discretionary real increases averaging 0.5% per year during the period under consideration. The Belgian law does not provide for any automatic real increases. (3.) In the public pension scheme for private sector employees, benefits are related to career earnings but subject to a ceiling. The ceiling increases at the same pace as conventional wages, i.e. at a lower pace than the average wage due to wage drift. Source: HFC. Table 2.1. Replacement rate for a basic typology (1) 2002 2010 2030 Gross replacement rate 1st pillar 36.5 37.1 36.8 Gross replacement rate 2nd pillar (2) 3.5 6.1 12.2 Gross replacement rate total 40.0 43.2 49.0 Net replacement rate total 66.1 68.2 77.2 (1.) Single man retiring at age 65 after 40 years of work with a wage equal to the average wage in the scheme in every year. (2.) The average contribution to the second pillar equals 4.25% of gross salary, starting in 1992. Source: HFC. Table 2.2. Projections for total public health care expenditure 2009-30 2003-30 Total public outlays on health 2.8 3.2 Demographic factor 0.8 0.9 (ageing) (0.7) (0.7) (population) (0.2) (0.2) Non-demographic 2.1 2.3 (GDP) (1.9) (1.9) (1.) Deflated by GDP prices; growth rates. Source: HFC. Table 2.3. Evolution of the structural primary surplus 2001 2002 2003 General government balance 0.6 0.1 0.3 Net interest charges (1) 6.2 5.7 5.2 Net primary balance (2) 6.8 5.8 5.5 Cyclical component 0.6 -0.2 -0.7 Cyclically adjusted net primary balance 6.2 6.0 6.2 Non-recurrent factors 0.4 0.2 1.5 Of which: UMTS sales 0.2 0.0 0.0 Transfer pension liabilities Belgacom 0.0 0.0 1.9 Fiscal amnesty 0.0 0.0 0.0 Funding of Belgian Railway Company 0.0 0.0 -0.4 Real estate sales 0.1 0.2 0.1 Securitisation of tax arrears 0.0 0.0 0.0 Structural net primary balance (3) 5.8 5.8 4.7 Structural general government balance (3) -0.4 0.1 -0.5 2004 2005 2006 (4) (4) (4) General government balance -0.1 -0.4 -0.5 Net interest charges (1) 4.7 4.4 4.1 Net primary balance (2) 4.6 4.0 3.6 Cyclical component -0.4 -0.2 0.1 Cyclically adjusted net primary balance 5.0 4.2 3.5 Non-recurrent factors 0.7 0.4 0.0 Of which: UMTS sales 0.0 0.0 0.0 Transfer pension liabilities Belgacom 0.0 0.0 0.0 Fiscal amnesty 0.1 0.0 0.0 Funding of Belgian Railway Company 0.4 0.0 0.0 Real estate sales 0.1 0.2 0.0 Securitisation of tax arrears 0.0 0.1 0.0 Structural net primary balance (3) 4.4 3.9 3.5 Structural general government balance (3) -0.3 -0.5 -0.6 (1.) Defined as the difference between gross government interest payments and gross government interest receipts. (2.) Defined as the difference between the general government balance and net interest charges. (3.) Defined as the net primary or general government balance corrected for the business cycle and for non-recurrent factors. (4.) Projections OECD. Source: OECD; NBB. Table 2.4. Maximum implied primary spending growth (1) under different growth assumptions Average yearly increases, constant prices Primary Primary spending spending excluding excluding health care Economic Primary health and Ostend/ activity spending care (2) Gembloux (3) 1.93 0.65 0.04 -0.14 2005-2007 2.43 1.20 0.68 0.51 2.93 1.75 1.33 1.15 (1.) Subject to the constraint that the budgetary targets of the Stability Programme are met. (2.) Implied growth of primary spending excluding health care (increasing by 4.5% a year in real terms). (3.) Implied growth of primary spending excluding health care (increasing by 4.5% a year in real terms) and excluding the spending increases decided already in Ostend and Gembloux. Source: NBB. Table 2.5. Origin of newly hired managers Total Origin : Origin Origin Percentage federal other private from government level of sector private (sensu government sector stricto = % Ministers Level N 23 8 10 5 28 N -1 107 59 32 16 18 N -2 28 28 0 0 0 Total 156 95 42 21 15
(1.) The pension scheme for private sector employees is responsible for two thirds of the public pension outlays. There are separate pension schemes for the public sector (14.5% of total spending on pensions in 2003), for independent workers (12% of total spending) and for employees of public enterprises (4% of spending).
(2.) The growth of conventional wages is below the growth of average wages in the economy, the difference being accounted for by wage drift, estimated at 0.5% annually.
(3.) The health index differs from the CPI in that it excludes all products that are harmful to human health from the consumption basket (tobacco, alcoholic beverages, diesel and petrol fuel).
(4.) Past observations are used to estimate the constant elasticity. Its estimate depends on the length of the period considered, but is always found to be greater than one. The Federal Planning Bureau finds an elasticity of 1.6 for the period 1971-2002. In the 70s, the share of the government has been substantially increased, which of course, has had an important impact on the growth rate of government expenditure in that period. Consequently, the high elasticity of 1.6 for the period 1970-2002 can be partly explained by this evolution.
(5.) As the cyclical component is projected to be zero in 2007, the surplus will be equal to the structural surplus.
(6.) Public debt has fallen from its peak of 136.7% of GDP in 1993 to 99.98% of GDP at the end of 2003.
(7.) The transfer amounted to 5 billion [euro], or 1.9% of GDP, and reflected the present value of the liabilities of the Belgacom pension fund, which have been taken over by the federal government.
(8.) Net gains on financial swaps of general government interest payments are included in the data used in the Excessive Debt Procedure and the evaluation of the Stability Programme but are excluded from ESA 1995 data. In the case of Belgium, this results in small differences in some years in data complied according to the two methodologies as net gains on financial swaps linked to public debt represent on average 0.1% of GDP.
(9.) The fiscal amnesty measure (Declaration liberatoire unique, DLU) brought in an additional 500 million [euro], 300 million [euro] less than estimated in the initial 2004 budget but 300 million [euro] more than estimated in the adjusted 2004 budget.
(10.) Measures have been taken in November 2004 to increase income in the health sector by C 160 million (pre-financing, claw back and increases in the reserves of the mutual funds (mutualities).
(11.) The NBMS/SNCB will be broken up into 3 entities, the railway transport company NMBS/SNCB, the infrastructure management company Infrabel, and the holding company NMBS/SNCB Holding which coordinates the two operational companies. Besides these three entities, the Fund for Railway Infrastructure was created, which owns the railway infrastructure and is liable for the corresponding debt of 7.4 billion [euro] (2.5% of GDP). According to ESA95 methodology, this fund would be considered as part of the general government, causing government debt to increase by 2.5% of GDP in 2005.
(12.) The government intends to sell packages of tax arrears to financial institutions at a discount reflecting the fact that some tax arrears are lost. Financial institutions receive all the money that could be recovered by the tax inspection service at the department of finance. This would give them an incentive to closely monitor the efforts made by tax inspectors.
(13.) These would include Regions and Communities in Belgium, which are responsible for important policy areas on their territories.
(14.) Unlike their peers in the private sector, contractual employees in the public sector do not benefit from occupational (second pillar) pension schemes.
(15.) Note that there is a ceiling on the maximal public sector pension that a civil servant can earn.
(16.) The name "Copernic reform" is actually no longer used, but the same philosophy remains in place.
(17.) By law, the selection procedure must take place in front of a mixed French and Dutch speaking jury. As this was not possible in a private head hunter firm, non-selected candidates could use this as an argument in their appeal (to the Cour d'Etat') against the hiring decision.
(18.) The moderation of the management salary is also due to the fact that the distinction between policy and execution has been maintained. The public services are still mainly focused on the execution of policies.
(19.) The "taxe compensatoire d'accises" will be abolished completely by 2007.
(20.) As part of the Kyoto objectives, Belgium has committed itself to reduce greenhouse gas emissions by 7.5% on average during 2008-12 from 1990 levels.
Aberle, G. (2003) "Rail Policy and Rail Subsidies--an Endless History with Weak Results" in: ECMT Fifty Years of Transport Policy 1953-2003.
Accenture (2003), eGovernment Leadership: Engaging the Customer, www.accenture.com.
Accenture (2004), eGovernment Leadership." High Performance, Maximum Value, www.accenture.com.
Arena, M. (2003), Le Monde Bouge, Relevons Ensemble les Defis de la Nouvelle Fonction Publique Federale, Brussels, 16 December 2003.
Bronchi, C. (2003), "The Effectiveness of Public Expenditure in Portugal", OECD Economics Department Working Papers, No. 380, OECD, Paris.
Chia, N.-C, A.K.C. Tsui and J. Whalley. (2001), "Ownership and Use Taxes as Congestion Correcting Instruments", NBER Working Papers, No. 8278.
Commission of the European Communities (2004), State Aid Scoreboard-Spring 2004 update, Brussels.
Federal Planning Bureau (FPB) (2004), Perspectives economiques 2004-09.
Friederiszick, H.W., L.-H Roller and C.C. Schulz (2003), "Evaluation of the Effectiveness of State Aid as a Policy Instrument: The Railway Sector", Draft version 24-10-2003.
High Finance Council (HFC) (2003), Ageing Study Committee (Conseil Superieur des Finances, Comite d'Etude sur le Vieillissement), Rapport Annuel.
High Finance Council (HFC) (2004a), Rapport Annuel 2004, Ministry of Finance, Brussels.
High Finance Council (HFC) (2004b), Ageing Study Committee (Conseil Superieur des Finances, Comite d'Etude sur le Vieillissement), Rapport Annuel.
InSites Consulting (2004), Belgian Internet Mapping, B.I.M. Brussels
Joos, A. and C. Kegels (2004), De administratieve lasten in Belgie voor het jaar 2002, Federal Planning Bureau, Brussels.
Joumard, I., P.M. Kongsrud, Y. Nam and R. Price (2004), "Enhancing the Effectiveness of Public Spending: Evidence in OECD Countries", OECD Economics Department Working Papers, No. 380, OECD, Paris.
Ministry of the Budget and Public Enterprises (2004), Les finances publiques belges pour l'annee 2005, Brussels.
Ministry of Finance (2003), Le Programme de Stabilite de la Belgique (2004-2007), Brussels.
National Bank of Belgium (2002), Rapport 2001: Evolution economique et financiere, Brussels.
National Bank of Belgium (2003), Rapport 2002: Evolution economique et financiere, Brussels.
National Bank of Belgium (2004a), Rapport 2003: Evolution economique et financiere, Brussels.
National Bank of Belgium (2004b), "Economische projecties 2004-2005 voor Belgie", Economisch Tijdschrift 2004(2), Brussels.
OECD (2001), OECD Economic Surveys: Belgium, Vol. 2001/5, OECD, Paris.
OECD (2003), OECD Economic Surveys: Belgium, Vol. 2003/1, OECD, Paris.
OECD (2003), "Managing Senior Management: Senior Civil Service Reform in OECD Member Countries", 28th Session of Public Management Committee, 13-14 November 2003, OECD, Paris.
OECD (2004b), "Taking Stock of Structural Policies in OECD Countries 2004 Assessment", Working Party No. 1 on Macroeconomic and Structural Policy Analysis, OECD, Paris.
Prospero (2004), Een nieuwe impuls voor economische welvaart in Belgie, McKinsey & Company.
Van den Noord, P. (2002), "Managing Public Expenditure: the UK Approach", Economics Department Working Papers, No. 341, OECD, Paris.
Wurzel, E. (2003), "Consolidating Germany's Finances: Issues in Public Sector Spending Reform", Economics Department Working Papers, No. 380, OECD, Paris.