Belgium: Selected Issues

This Selected Issues paper analyzes fiscal devolution in Belgium. It examines long-term fiscal strategies for meeting the fiscal burden of population aging. The paper presents estimates of the fiscal costs of population aging in Belgium, and discusses their sensitivity to underlying assumptions. Strategies for meeting the fiscal costs of aging are discussed. Specifically, a strategy of swift debt reduction is discussed to see what budget surplus would be needed to reduce debt and interest payments rapidly enough to finance these costs. Alternative policy options are also discussed to show the consequences of delayed fiscal adjustment.

Abstract

This Selected Issues paper analyzes fiscal devolution in Belgium. It examines long-term fiscal strategies for meeting the fiscal burden of population aging. The paper presents estimates of the fiscal costs of population aging in Belgium, and discusses their sensitivity to underlying assumptions. Strategies for meeting the fiscal costs of aging are discussed. Specifically, a strategy of swift debt reduction is discussed to see what budget surplus would be needed to reduce debt and interest payments rapidly enough to finance these costs. Alternative policy options are also discussed to show the consequences of delayed fiscal adjustment.

III. Regional Economic Development in Belgium22

51. Over the last three decades the performance of the Belgian economy has witnessed sharp geographical disparities. Taken alone, the regional economy of Flanders would be among the best performing European economies; and Flemish real per capita GDP is now more than ⅓ higher than in Wallonia, while 50 years ago it was almost one-fifth lower. It is worth emphasizing, however, that experience within both Flanders and Wallonia has been far from uniform, as some areas in the former have lagged and some areas in the latter are prosperous. Nevertheless, much of the subsequent discussion will be at the level of the Flemish and Walloon régions, reflecting the fact that regional development policy is now vested in the governments of these régions, and also because of data availability.23

52. Apart from their national significance and macroeconomic importance, Belgium’s regional problems offer valuable insights for other countries facing similar disparities. Other countries with regional disparities, notably Italy, Spain, and post-unification Germany, may be able to draw lessons from the Belgian experience. Also, Belgium is now more politically and fiscally decentralized than those of some other countries with regional differences, e.g., the United Kingdom, Italy, and France, and the Belgian experience may therefore offer valuable guidance on the interaction between decentralization and regional development.24 Finally, and related to the latter point, the policy environment within which Belgian regional policy has operated—notably, the absence of an independent monetary policy, the restrictions on fiscal policy through stability programs, and the lack of large scale interregional/national transfer mechanisms—resembles in some respects the one faced by national governments in the euro zone.

53. This chapter attempts to distill the Belgian regional development policy experience. The first section provides some background on regional issues in Belgium. Next, the course of regional policies is charted. Finally, stock is taken of the results achieved thus far. The main conclusions are mixed: while decentralization was accompanied by changes in the focus and delivery of development policies and there is some evidence of convergence, it is clear the much remains to be done.

A. Some Background on Regional Issues in Belgium

54. Considering developments at the broad level of the régions, economic performance and structure has varied substantially over time. For more than a century, starting in the early 1800s, Wallonia was Belgium’s economic powerhouse, with an economy based primarily on coal mining and heavy industry. Since the mid-1950s, however, the economy of Flanders has fared significantly better (Table 1). While some of this reversal initially reflected convergence, Flanders eventually surpassed Wallonia to become the more prosperous and dynamic region.25

Table III.1.

Regional Economic Performance, 1955–97

(Average annual percentage changes, unless otherwise indicated)

article image
Sources: Institute of National Accounts data, compiled in KBC (1999), Eurostat Cronos database, and IMF: International Financial Statistics.

55. It was hoped that increased regional authority, apart from addressing popular demands, would help redress economic imbalances. The 1980s and 1990s saw a substantial decentralization of power (and fiscal authority) from the center to the régions and the linguistic communities. The subsidiarity principle of economic policy would suggest that the differing economic conditions in the régions could be better addressed by granting the relevant subnational governments greater leeway in their economic policies. This was especially salient since at the time it was felt that public resources were inefficiently deployed and insufficiently tailored to specific needs.26

B. A Brief History of Belgian Regional Policy

56. Regional policy typically has several targets and can pursue a mix of strategies (Box 1). The ultimate objective of regional policies to reduce economic disparities between regions by increasing economic growth in poorer areas, and by implication, of the national economy as a whole.

Justification for Regional Policy Intervention

At least five standard reasons have been advanced for regional policy (Armstrong and Taylor, 2000):

  • regional economic disparities are unfair and cause resentment and dissatisfaction;

  • high unemployment in one region tends to persist through negative effects on that region’s human capital, which would not accrue if unemployment could be reduced;

  • regionally concentrated economic growth is inefficient because it creates congestion in the booming regions;

  • regionally concentrated growth and unemployment ratchets up the macroeconomic rate of inflation and unemployment,

  • the flexibility requirements needed to back up a market approach are too onerous: it is argued that it would be politically impossible or ineffective, to, for example, increase mobility incentives or reduce congestion with peak-load road pricing.

57. If one were to conceptualize regional policies in the broadest sense, a symmetry with international development policies would arise. Regional policies can be divided into two broadly distinct approaches: those seeking to attract investment (and increase regional exports) and those seeking to promote indigenous economic development.

  • Typical examples of export-based policies are broad incentives to attract foreign direct investors (often in industries perceived as key), the establishment of industrial parks (targeted infrastructure), and export promotion schemes.

  • Inward-oriented policies are characterized by investment in general infrastructure and human capital, though often their de-facto effect has been to prevent decline in existing industries.

It is important to note that both these approaches seek to move capital to jobs, rather than to encourage labor mobility.

58. Belgian regional policy has used a variety of instruments over time. Belgium was a relative latecomer to regional policy (Vanhoudt and Buyst, 2002), starting at the end of the 1950s with the implementation of economic policies with a regional focus when the coal crisis affected Wallonia. Initial efforts focused on the then-prevalent policy of supporting industry (OECD, 1994). Support was given in the form of interest-rate relief on, and guarantees for, borrowing, a policy tool which remains in the arsenal of regional policy makers. Regional policy was centered on “development regions”—regions with structural adjustment problems—with the aim that the population of such regions was not to exceed 15 percent of the entire population. In the mid-60’s the focus of regional policies was broadened to include regions adversely affected by the decline of the coal-mining industries in the South, and to regions lagging in industrial development. Moreover, national policies to subsidize “strategic industries”, such as textiles and ship building, had distinct regional effects. In the 1970s, much of existing regional policy was formalized into specific legislation, policy instruments were broadened to include capital subsidies, tax exemptions, and targeted measures were introduced to address training and infrastructure needs. Furthermore, small and medium-enterprise development was given a regional dimension. Most recently, regional policy has shifted toward fostering high-tech and knowledge clusters and their associated spin-offs.

59. The change in instruments has been partly driven by changing needs. Throughout the 1970s, central government regional policy initiatives were divided almost evenly between the two major regions (OECD, 1994). However, even under these conditions, the emphasis of regional policies differed across régions, mainly reflecting historical patterns. The severe difficulties in the Walloon heavy industry and coal sectors triggered direct aid to these sectors. Flanders, with a more diversified economy, benefited from investment in new infrastructure, such as the harbor in Antwerp.

60. Moreover, devolution introduced important differences in the targeting of regional support. Starting 1978, the régions were given authority to implement and interpret central government regional policy legislation. In principle, the eclipse of a central government by regional governments can have ambiguous effects on the effectiveness of regional policy. As indicated above, proponents of devolution have argued that subsidiarity ensures more efficient policy intervention. Skeptics, on the other hand, argued that many objectives of regional policies have an externality that can only be internalized at the national level. Thus, sub-national entities would not extend sufficient effort to achieve nationally optimal policy outcomes. From the country-wide perspective, regional policy developed two dimensions, (i) policies used by the different régions across their territory, and (ii) the policies individual régions use to address sub-regional disparities. Figure 1 summarizes some indicators:

Figure III.1.
Figure III.1.

Belgium: Indicators of Regional Policy and Performance

Citation: IMF Staff Country Reports 2003, 050; 10.5089/9781451803181.002.A003

Source: Eurostat, Cronos database and data provided by the authorities.1/ There are no data on the composition of subsidies in Wallonia.2/ Including shared revenue.3/ In percent of overall employment, including “non-market employment”.
  • Wallonia traditionally focused its regional policy on direct aid to support industries in distress, more recently making such subsidies conditional on performance goals (such as investment, employment, and innovation). Apart from being consistent with the inward-oriented regional development model, these policies of course also reflected social needs. Turning to specific regional policies, apart from support to areas with declining industries, Wallonia has also moved to support depressed agricultural zones to encourage diversification. Regarding human capital, Wallonia has relied principally on retraining and active employment measures. More recently, the région has decided to change the orientation of regional policy, as evidenced by the “contract with the future,” and is accordingly focusing more on creating a favorable enterprise environment and high-tech and knowledge clusters.27 Some evidence that this policy shift may be paying off may be seen in the number of per capita patent applications, which between 1995 and 2000 have grown by more than 20 percent a year and are beginning to catch up with those in Flanders and neighboring European countries. The région has also spent comparatively more on public services and public employment. Reflecting these policy choices, in 1997 public primary expenditures per head were some 9 percent higher than in Flanders, mainly reflecting a higher wage bill and transfers (Bisciari, 1998). The higher level of primary spending resulted in a larger stock of debt and larger interest payments than in Flanders, and persistent, albeit declining budget deficits.

  • Flanders, also initially adopted a traditional inward-oriented policy of supporting large enterprises. But direct aid to enterprises was reined in significantly in 1991 (Bisciari, 1998). Emphasis was shifted to creating an enabling environment for the private sector, attracting investment, trade promotion, and assistance to small and medium enterprises (OECD, 1994).28 Flanders attempted to develop general human capital, and the share of the labor force with tertiary or higher education climbed from 11.4 percent in 1986 to 21.1 in 2001 (the Walloon ratios are 11.6 and 19.4 percent, respectively). Arguably, as a result of these policies Flanders was more open and better placed to exploit the gains from the European common market (Capron and Hostelard, 1991). Specific regional policy instruments included research and development zones, which were set up around major universities to foster a high-tech cluster through research links between universities and companies. Similar efforts created so-called “impetus areas” and “unemployment action zones.” Turning to budgetary policies, Flanders focused more on investment rather than consumption and transfers (Bisciari, 1997).29 The fiscal position improved, reaching a surplus in 2000 (Figure 2). One result has been a better credit rating than the other régions.30 The overall size of government as well as government consumption have remained considerably smaller in Flanders. While in per capita terms overall primary expenditures in Flanders are only marginally lower than the aggregated value for Wallonia and the French community, they are considerably smaller when expressed as a percent of regional income.31 In short, Flanders spends more on public investment, but less on wages, subsidies and interest payments. More detailed studies (Dermien et al., 2002 and Bisciari, 1999) confirm this pattern, and in addition indicate that Flanders spends less on regional policy (in part reflecting less of a need to cofinance EU projects), but more on social services, education and culture. On the other hand, Flanders has recently also increased public employment, strengthened environmental regulation, and expanded the social welfare system through the introduction of a nursing care scheme.

  • The federal government’s role in regional policy has been increasingly limited to enforcing compliance with EU regulations on state aid and social and agricultural policies. National policies, such as social security or employment protection, continue to have a “passive” regional dimension to the extent that they effect different geographical parts of the country differently due to, for example, differences in population and employment characteristics.

Figure III.2.
Figure III.2.

Belgium: Indicators of Regional Fiscal Performance 1/

(In Percent of Gross Regional Product)

Citation: IMF Staff Country Reports 2003, 050; 10.5089/9781451803181.002.A003

Source: High Council of Financ; Chronos database.1/ Data for Wallonia also comprise the French Community budget.

61. The EU became an important player in Belgian regional policies. The EU has in the past stepped in to narrow the scope of regional support extended by the national government.32 But it has also significantly increased its own regional spending in Belgium by more than two and a half times between 1989–93 and 1994–99, to a level of €2.1 billion, equivalent to some 0.2 percent of Belgian GDP. Taking into account the national cofinancing mobilized, this resulted in annual allocations of some 0.6 percent of GDP (EU, 1997a). EU regional spending is targeted to meet a set of objectives (Box 2); Table 2 shows the levels of support allocated in fulfillment of these objectives.

EU Regional Policy Objectives in Belgium

EU regional policy support has been directed to specific objectives. Under the 1994-99 five-year program, Belgian regions qualify for the following (see EU, 1997b):

  • Objective 1 (“economic adjustment of regions whose development is lagging behind”): The Hainaut area in Wallonia was eligible for such support in 1994-99, explaining much of the sharp rise in EU regional spending in Belgium.1 The goal was to have per-capita GDP in the Hainaut province grow by 0.5 percentage points more than the EU average. The lion’s share of spending was allocated to direct aid to enterprises, indirect aid through improving the productive environment (diverse measures from supporting exports of small and medium size enterprises to cleaning up derelict sites), and human resources (mainly training the unemployed and other problem groups on the labor market) (EU 2001).

  • Objective 2 (“economic conversion of declining industrial areas”): areas in both Wallonia (Aubange, Hainaut, Liege) and Flanders (Limburg, Turnhout) were eligible for support under this objective. Measures aimed to stimulate endogenous development and continued support to enterprises, but also the improvement of water quality.2

  • Objectives 3 and 4 (“human resource development outside regions eligible for Objective 1”):3 Measures were geared toward improving the labor market prospects of the long-term unemployed and other problem groups on the labor market (ex-convicts, immigrants, the handicapped, and the low skilled).

  • Objective 5a: The target was to speed up the adjustment and modernization of the agricultural sector in the framework of the common agricultural policy (CAP). Some 20 percent of the Belgian agricultural area—covering parts of both Flanders and Wallonia—was designated as “less favored.” Additional funds were allocated to support reduction of the fisheries fleet.

  • Objective 5b (“economic diversification of rural areas”): The focus was to reinforce the endogenous development of the affected regions, in both Wallonia and Flanders.

  • “Community initiatives”: Belgium takes part in all but one of these, participating in those for industrial conversion, small and medium sized enterprises, urban areas, and human resources.

1 For the 7-year period 2000–06, Hainaut has lost its objective 1 status but qualifies for transitional support of €645 million, which is expected to give rise to an equivalent expenditure by the Walloon région.2 Under the 2000-06 program, in Wallonia, the Liege region benefits from EU support to improve the competitiveness of enterprises, while the remainder of Wallonia is eligible for rural programs. Flanders has four program in the provinces of West and East Flanders, Antwerp, and the largest in Limburg. région.3 These two objectives have recently been combined into one “objective 3”, which is also the reference framework for all measures taken under the emolovment chanter of the Treatv of Amsterdam.
Table III.2.

EU Regional Policy Support Allocations, 1989–2006 1/

(In millions of Euro/ECU)

article image
Source: European Union.

The numbers are budgeted figures.

62. The lion’s share of EU support continues to be geared towards human resource development. More funds have recently been allocated to infrastructure and productive environment, but this is largely explained by disbursements of objective-1 funds to Hainaut, which aims to improve the competitiveness of enterprises (⅓ of all total), the attractiveness of the regions—e.g., through cleaning up industrial sites—(¼ of total), prospects for tourism and research facilities (one-fifth each). With the overall share of funds for the improvement of enterprise competitiveness has remained stable. Instruments have shifted from direct aid toward providing risk capital.

C. Taking Stock

Regional development—stylized facts

63. Differences in the sectoral composition of régions have increased (Figure 3). On average, Flanders and Wallonia have grown more dissimilar, accentuating previously existing differences. The economy of Flanders is characterized by higher shares of energy production, transport equipment manufacturing, chemical industry, textiles, airport and harbor services, while the Walloon economy has significantly higher shares of minerals exploitation and primary commodities processing, commercial services, education, government services, and real estate. The discrepancy in sectoral shares has increased in ⅔ of the sectors and the standard deviation of their sectoral differences has edged up from 1.4 in 1987 to 1.5 in 1997, the latest year for which these data are available.

Figure III.3.
Figure III.3.

Belgium: Sectoral Decomposition of Value Added

(In Percent of Total Value Added)

Citation: IMF Staff Country Reports 2003, 050; 10.5089/9781451803181.002.A003

Source: Dala provided by the authorities.

64. Looking at the more disaggregated level of provinces and arrondissements, per capita income appears to have converged over time. Table 3 shows that a high initial per capita income has a negative effect on subsequent growth rules, i.e., over time incomes of richer and poorer regions will tend to converge. At the NUTS 2 level (provinces), this effect is estimated at some 2½ percent per year.33 However, at the NUTS 3 level (arrondissements) the size of this effect declines to ¾ percent and is no longer statistically significant (columns 2a - 2c). There is some evidence that convergence is weaker within Wallonia than Flanders at both the NUTS 2 and NUTS 3 levels (Figure 4).

Table III.3.

Growth Regressions (dependent variable: average real growth of GRP)

(t statistics in parentheses, absolute values)

article image
The numbers indicate regression coefficients of a log linear specification that regresses nominal average annual GRP growth rates on a constant, initial income, and a dummy variable for Wallonia. The t-statistics are in parentheses. Annual data for 1985-99, obtained from Belgostat.
Figure III.4.

Belgium: Regional Growth Regressions 1/

Citation: IMF Staff Country Reports 2003, 050; 10.5089/9781451803181.002.A003

Source: Staff estimates based on Belgostat data.1/ Regression is (log of) growth against initial income.

65. However, the process of convergence is far from complete. As described above, Hainaut was identified as a region in severe distress, and accordingly benefited from special EU structural support under “objective 1 status.” Nevertheless, it still has the lowest per capita GRP in Belgium, and its growth performance fell behind both the Belgian and the Walloon average, with the differential actually widening over the most recent period. While these developments do not necessarily suggest a negative effect of regional policy—the counterfactual in the absence of regional policy may have been even worse—it does indicate that regional policy is not fully achieving its goals.

66. The distribution of per capita income within régions has also become somewhat more equal. In other words, there is some evidence for a convergence in the distribution of GRPs within both Flanders and Wallonia. However, while the statement holds for Flanders irrespective of the level of aggregation, a more nuanced assessment is called for in Wallonia and Belgium as a whole (Figure 5). In both, inequality has declined at the NUTS 2 level, but at the NUTS 3 level inequality has risen in Wallonia, and for Belgium as a whole that increase has cancelled out the decrease in Flanders. Further disaggregation shows that the reduction in variation within Flanders is due to lagging provinces catching up with the better performers: at the Nuts 2 level while in 1977 three out of the top half GRP Belgian provinces (excluding Brussels) were situated in Flanders, in 1999 Flemish provinces took 4 of the 5 top spots; at the NUTS 3 level, in 1977 2/3 of the top 50 percent of arrondissements were located in Flanders, a rate that increased to ¾ in 1999. The reverse is true for Wallonia: for instance at the Nuts 2 in 1977 four of the lowest five per capita income provinces were in Wallonia, but all five were in Wallonia in 1999. The better performance of Flanders with its historically higher inequality conforms to other evidence that countries or regions with more inequality tend to grow faster (see Quah, 1996).

Figure III.5.
Figure III.5.

Belgium: Indicators of Regional Variation of Real per Capita Income

(Coefficient of Variation)

Citation: IMF Staff Country Reports 2003, 050; 10.5089/9781451803181.002.A003

Source: Eurostat Cronos database, and staff calculations.

67. The Flemish and Walloon economies have exhibited differing abilities to weather shocks. Until the late 1970s both Wallonia and Flanders recorded similar investment rates, export growth and labor market performance. However, since then, these developments have diverged:

  • Investment in Flanders recovered quickly, whereas investment as a share in value added never reach its pre-oil shock level in Wallonia.

  • Growth in Flemish exports also picked up quickly, reaching new highs of close to 90 percent of regional value added (compared to some 50 percent after the first oil crisis). The level of Flemish exports is biased upwards because the major national airport and port are located there. In Wallonia, however, the recovery in exports was much delayed and faltered again in the late 80’s with the result that the economy is now less outward oriented than in the early 70s.

  • Labor market performance began to diverge (see Country Report No. 02/43 for a fuller discussion).

68. Productivity and employment rates developed in sharply different directions. In a somewhat crude growth accounting framework, per capita GRP can be decomposed into the product of productivity, the employment rate, and the participation rate:34

GRPPopulation=GRPEmployment*EmploymentActivepopulation*ActivepopulationPopulation

As a result of the developments summarized in Table 4, productivity in Flanders is now some 14 percent higher than in Wallonia, while employment and activity rates exceed Walloon levels by 7 and 8 percent, respectively.

Table III.4.

Belgium: Decomposition of Regional Growth Performance, 1979–2000

article image
Sources: Eurostat Cronos database; and Fund staff calculations.

69. Population growth patterns are, however, converging. In contrast to the adverse relative economic performance, the rate of population decline has slowed in Hainaut—in 1990–95 the population grew—perhaps an indication of perceived improved prospects. More generally, the rates of population growth across regions have become less divergent since 1977, with Flanders again explaining most of the adjustment.35 To the extent that regional policy tries to equalize living conditions and to the extent that internal migration behavior reflects divergent living conditions, this may reflect a positive effect of regional policy.

Searching for explanations

70. Several factors are at play, but they cannot entirely account for the observed divergences. Observers have suggested several explanations: initial conditions, sectoral composition, differing adjustment abilities, productivity or factor endowments, as well as geography. Unfortunately, lack of data to examine the role of these different factors prevents an econometric examination.

71. Initial conditions, notably, different initial economic structure, are unlikely to be the full story. The prolonged reversal of economic fortunes between regions—which has gone well beyond catch-up—suggests that more than adverse initial conditions are at work. The decline in the Walloon coal and steel sector as well as the lighter and smaller-scale industrial structure and traditionally lower wages in Flanders certainly played a role early on. The OECD (1994), has also pointed to effects related to, but not limited to, heavy industry, such as a militant unionized work force, that may have stood in the way of a more speedy adjustment in Wallonia. But these factors should have by now run their course. The shift-share analysis in Table 5 supports this assessment (see also SM/02/47). The Table shows that over the period 1987–97, Flemish value added grew 12.9 percent more than in Wallonia. Of this difference, only 5 percent is attributable to initial sectoral composition, while the majority is explained through better growth performance irrespective of the sector. Compared to Brussels, however, initial composition dominates.

Table III.5.

Shift-Share Analysis of Regional Value Added, 1987–97

article image
Source: Staff calculations based on Eurostat data.

72. It has been suggested that Flanders may have disproportionately benefited from its more developed ties with the Netherlands and Germany. The former boomed in the 1990s and the latter led much of European performance prior to the early 1990’s. Moreover, Flanders borders some of the better performing Dutch regions, with which it also shares a similar sectoral composition (Federal Planning Bureau, 2002). If so, the recent weaker German economy and the better performance of France should imply an improvement in the relative performance of Wallonia. This is, however, not yet clear in the data. Indeed, and notwithstanding the lack of natural boundaries, the Walloon economy is markedly different from its neighboring regions (Federal Planning Bureau, 2002).

73. Some of the explanation lies in economic policies. For example, in Wallonia, the larger size of the government and the correspondingly high tax burden may have hindered the private sector. Similarly, the lower share of state aid to distressed companies in Flanders may have hastened an adjustment process that, in the event, proved inevitable and leveled the playing field for the creation of new businesses. Moreover, national social policy may have contributed to insufficient regional labor mobility. While no firm data exist on mobility, available estimates indicate that, notwithstanding some increase in Wallonia, the région still falls short of Flemish levels (Binon et al. 1998).36 The national unemployment and welfare system, encourages reservation wages relative to underlying productivity levels and employment opportunities in the lagging areas. This may have been compounded by the centralized aspects of the Belgian wage setting mechanism which may have hindered wage differentiation. A comprehensive recent study estimated that Walloon wages are some 7 percent lower than Flemish wages when controlling for worker and job characteristics and the business cycle, which is not sufficient to offset lower productivity levels (Laurent, 1998).

74. Transfers across the régions appear to be diminishing, but, while cushioning differences in economic performance, may have reduced impetus for reforms to regional policies. Studies in the 1970s concluded that Walloon residents benefited from a net annual transfer of close to 5 percent of GDP. More recent studies, however, indicate much smaller, though still significant, transfers (De Boeck and Van Gompel, 1998, van Rompuy and Bilsen, 1993, and KBC, 2000). These occur primarily through the social security system, reflecting differences in age distributions (pensions) and labor-market outcomes (unemployment benefits). The decline over time is due at least in part to reforms in fiscal relations between levels of government instituted by the special law on the financing of régions and communities of 1989 (see OECD, 1992 and Chapter I of this Selected Issues for a description). Moreover, recent decisions and fiscal devolution are likely to further limit transfers. With a more active and better-paid population, Flanders on a net basis supports the national social security system, which has resulted in pressure to decentralize it as well. One factor explaining higher central-government expenditure in Wallonia is the higher share of federal government employment stationed in Wallonia. To the extent that fiscal retrenchment through expenditure savings has effects on the government payroll, this channel for transfers would become less prominent.

D. Concluding Remarks

75. This chapter has essentially reflected past developments. The key issue is, of course the policies to be pursued in the future. It would appear that the distinct policies applied in Wallonia and the EU objective 1 region of Hainaut have not been fully successful. However, there is so far not enough information on the types of policies pursued to perform a rigorous economic policy evaluation. Belgium managed to consolidate public finances and to keep pace with European growth, even as Flanders and Wallonia experienced quite different outcomes. As the two régions also pursued somewhat different economic policies, study of how these policies affected economic growth offers fertile ground. Early indications point to the benefits of policies to open the economy, focus on investment in general human capital, and create a business-friendly environment.

References

  • Armstrong, H., and J. Taylor, 2000, Regional Economics and Policy, Third Edition, Blackwell.

  • Baumol, W., 1986, “Productivity Growth, Convergence and Welfare: What Long-Run Data Show,American Economic Review, 76, pp. 107285.

    • Search Google Scholar
    • Export Citation
  • Barro, R., and X. Sala-i-Martin, 1995, Economic Growth, McGraw-Hill.

  • Binon, F., J.-B. Charles, R. Deschamps, S. Delaigle, J.-C. Jacquemin, O. Lohest, C. Melard, M. Mignolet, H. Sneesens, E. Toulemonde, and B. van der Linden, 1998, “Recherche Relative aux Facteurs Explicatifs de l’Evolution de l’Emploi en Wallonie et de la Divergence Iinterrégionale pour la Période 1974–1995,F.U.N.D.P. and U.C.L. mimeo.

    • Search Google Scholar
    • Export Citation
  • Bisciari, P., 1997, “Vergelijking van de uitgavenbesteding in Vlaanderen en Wallonie,Nieuw Tijdschrift voor Politiek, No. 5, pp. 533.

    • Search Google Scholar
    • Export Citation
  • Bisciari, P., 1998, Federalism et Dépenses Publiques, in: E. De Callatay, (Ed), Les Finances des Régions and Communutes en Belgique, Reflets et Perspectives de la Vie Economique, pp. 33– 60.

    • Search Google Scholar
    • Export Citation
  • Bisciari, P., 1999, Les Enjeux de la Convergence Budgetaire au Sein d’une Belgique Fedérale, Koninklijke Vlaamse Academie van Belgie van Wetenschappen and Kunsten, mimeo.

    • Search Google Scholar
    • Export Citation
  • Capron, H., and S. Hostelard, 1991, “Perspectives d’Intégration Régionale au Marché Interieur,Cahiers Economiques de Bruxelles, No. 131, 3. Trimester, pp. 273303.

    • Search Google Scholar
    • Export Citation
  • Cockx, B., and G. Ridder, 2001, “Social Employment of Welfare Recipients in Belgium: An Evaluation,” in Economic Journal, Vol. 111, (April), pp 32252.

    • Search Google Scholar
    • Export Citation
  • De Boeck, E., and J. Van Gompel, 1998, “Financiele stromen tussen de Belgische gewesten opnieuw bekeken,” in: Vanderveeren, C. and Vuchelen, J. (eds): Een Vlaamse fiscalieit binnen een economische en monetaire unie.

    • Search Google Scholar
    • Export Citation
  • Dermien, V., S. Bastuits, and R. Deschamps, 2002, “Dépenses de la Flanders et de la Wallonie: Comparison des Budgets 2002,CREW Working Paper No. 239.

    • Search Google Scholar
    • Export Citation
  • EU, 1997a, The Impact of Structural Policies on Economic and Social Cohesion in the Union 1989–99, Luxembourg.

  • EU, 1997b, 9th Annual Report on Structural Funds, Brussels, available via internet link www.eu.int.

  • Federal Planning Bureau, 2002, Les Régions Belges en Europe: Analyse Macro-Sectorielle

  • KBC, 1999, Economic and Financial Bulletin, Vol. 54, No. 16.

  • KBC, 2000: Economie and Financial Bulletin, Vol. 55, No. 10.

  • Laurent, S., 1998, Panel Data Analysis of Individual Wages: Belgium 1992-95, mimeo

  • OECD, 1992, Belgium: Economic Survey.

  • OECD, 1994, Report on Regional Development Problems and Policies in Belgium and Luxembourg, Working Paper, No. 3.

  • Quah, D., 1996, “Empirics for Economic Growth and Convergence,European Economic Review, 40, pp. 135376.

  • UWE, 2002, Rapport sur la Situation Economique en Wallonie.

  • Vanhoudt, P. and E. Buyst, 2002, Regional Disparities and (Non-) Convergence in a Federal Monetary Union—Evidence from Belgium, mimeo.

    • Search Google Scholar
    • Export Citation
  • Van Rompuy, P., and V. Bilsen, 1993, Regionalisering van de sociale Zekerheid, Kafholiek Universiteit te Leuven: Leuvense Economische Standpuntern, No. 1993/68.

    • Search Google Scholar
    • Export Citation
22

This Chapter was prepared by Gerwin Bell.

23

The term regional policy as used here concerns economic policies seeking to have differential impact on different subnational geographic areas. This may lead to misunderstandings in the Belgian context, where the largest geographical subnational units (Flanders, Wallonia, and Brussels) are called “régions:” when specifically referring to these, this Chapter will use therefore the term région.

24

See Chapter I, “Fiscal Devolution in Belgium” for a discussion of the fiscal dimension of decentralization.

25

These results are reinforced when GRP data, based on the location of production rather than residence, are corrected for the distortion imparted by workers living outside of Brussels, but commuting there to work.

26

For example, the OECD observed: “… the state budget was kind of a free-for-all in which the allocation of resources gave rise to all kinds of haggling and trade-offs between the various regional entities in the country. Such haggling often meant that resources were wasted, frequently attracting bitter criticism.” (OECD, 1994, p. 31.)

27

There has, however, been some recent controversy about whether the actual budget implementation follows these goals, see UWE (2002).

28

Of course, not all Flemish policies should be emulated: one recent study (Cockx and Ridder, 2001) assesses an active labor market policy scheme in Flanders and finds it a complete failure.

29

Public investment targeted environment, water, and transport infrastructure, including through transfers to local authorities and, recently, public-private partnerships.

30

At present, Flanders holds a AA+ rating with a stable outlook by Standard & Poor’s, the same as Belgian sovereign debt. Brussels is rated a notch lower at AA, also with a stable outlook. Wallonia is rated Aa3 by Moody’s, with a stable outlook.

31

This assessment is based on an aggregated view of the 2001 budget performance. An exact assessment is made difficult by the need to properly account for expenditure by the French community and the Flemish budget in Brussels. The assessment is based on the population census-based allocation mechanism discussed in Dermien et al. (2002).

32

For example, in the early 1980s, the EU reduced the size of “development zones” eligible for direct aid from 40 percent of the population to 36 percent.

33

The convergence rate of 2½ percent (column la, coefficient on initial income) conforms to international evidence that has put typical convergence speed between 2 and 3 percent over a number of country and regional samples (see Baumol, 1986, and Barro and Sala-i-Martin, 1995).

34

Lack of data on regional capital stocks or investment precludes a proper growth accounting exercise. The term “productivity” in the present context thus includes the effects of total factor productivity, capital intensity, and labor productivity.

35

Brussels continued to suffer a population decline throughout the period as residents moved to neighboring Walloon Brabant.

36

A relatively small country with good travel infrastructure, like Belgium, should be well suited for mobility. More detailed longitudinal micro data would be needed to be able to assess the actual situation.

Belgium: Selected Issues
Author: International Monetary Fund
  • View in gallery

    Belgium: Indicators of Regional Policy and Performance

  • View in gallery

    Belgium: Indicators of Regional Fiscal Performance 1/

    (In Percent of Gross Regional Product)

  • View in gallery

    Belgium: Sectoral Decomposition of Value Added

    (In Percent of Total Value Added)

  • View in gallery

    Belgium: Regional Growth Regressions 1/

  • View in gallery

    Belgium: Indicators of Regional Variation of Real per Capita Income

    (Coefficient of Variation)