Results of in-depth reviews under Regulation (EU) No 1176/2011 on the prevention and correction of macroeconomic imbalances

Belgium continues to experience macroeconomic imbalances, which require monitoring and policy action. In particular, developments with regard to the external competitiveness of goods continue to deserve attention as a persistent deterioration would threaten macroeconomic stability.

More specifically, a continuing worsening of competitiveness, including its non-cost dimension, may engender negative consequences for the economy. The ability of manufacturing to compete internationally has been hampered, which is reflected in eroding producers' margins and in job destruction. Squaring high labour costs with sustainable job creation and high standards of living requires a push towards products higher up in the global value chains. At the same time, a further decoupling between fast wage and slow productivity growth needs to be prevented. In this respect, Belgium has taken steps that are expected to produce effects in upcoming years. Yet, preserving the manufacturing basis requires more ambitious action, all the more so with reforms unfolding in competitor countries. This is related to labour taxes and making wage formation more responsive to economic and sectorial realities, and addressing persistent problems with regard to labour market functioning. Belgium's high public debt remains a concern for the sustainability of public finances. On the positive side, however, Belgium has managed to stabilize its public debt ratio, is estimated to have met the recommended deficit target in 2013, and is projected to keep the deficit below 3 per cent of GDP. Moreover, long average maturities, relatively reduced interlinkages with the domestic financial sector and a relatively healthy private sector temper risks for the wider economy.

Excerpt of country-specific findings on Belgium, COM(2014) 150 final, 5.3.2014

Executive Summary and Conclusions 7

1.       Introduction  9

2.       Macroeconomic Developments 11

3.       Imbalances and Risks 17

3.1.     Competitiveness 17

3.1.1.  Product specialisation  18

3.1.2.  Export orientation  19

3.1.3.  Cost competitiveness 20

3.1.4.  Functioning of the labour market 28

3.2.     Indebtedness 36

3.2.1.  Recent evolution of the public debt 36

3.2.2.  Risks of short-term fiscal stress 37

3.2.3.  Long-term sustainability risks of the Belgian public debt 39

3.2.4.  Inter-linkages with the financial sector 41

3.2.5.  The public debt and the overall indebtedness of the Belgian economy  41

4.       Specific Topic: The Belgian tax system in the context of macro-economic imbalances 49

5.       Policy Challenges 55

References 58

LIST OF Tables

2.1.     Key economic, financial and social indicators - Belgium   15

3.1.     Manufactured exports by type (% of total) 18

3.2.     Main export markets of Belgium (% of total) 19

LIST OF Graphs

2.1.     Decomposition of potential growth  11

2.2.     Activity rate (2012, %, 15-64y) 11

2.3.     Sectors' share in GVA (%) 12

2.4.     Breakdown of current account balance  12

2.5.     NIIP by sector 13

2.6.     Total debt decomposition (consolidated) 13

2.7.     Decomposition of consolidated credit flows 14

3.1.     Evolution of export market shares for goods (2000=100) 17

3.2.     Decomposition of export market share changes for goods 17

3.3.     Business enterprise researchers per 1000 employees 19

3.4.     Dynamism and competitiveness of good exports (top-10 destinations, 2010-12) 19

3.5.     REER measurements 20

3.6.     Nominal unit labour cost (2005 = 100) 21

3.7.     Evolution of profit margins in the manufacturing sector 24

3.8.     Capital stock/hour worked (2000 = 100) 24

3.9a.  Value added in industry (excl. building & construction, % of total) 25

3.9b.  Value added in manufacturing industry (% of total) 25

3.10.   HICP (annual % change) 25

3.11.   RUEC levels in Belgium and neighbouring countries 28

3.12.   Counterfactual comparison of RUEC in Belgium and neighbouring countries 28

3.13.   Employment rates (2012; %) 29

3.14a.                                                                    2012 employment rate (%) by regions and NUTS2 regions 29

3.14b.                                                               2012 unemployment rate (%) by regions and NUTS2 regions 29

3.15.   Unemployment traps (2012) 31

3.16.   Beveridge curve  32

3.17.   Employment rate by highest educational attainment 33

3.18.   Gross monthly earnings per skill level and age (2011) 33

3.19.   Gross and net public debt (% GDP), BE vs. EA   37

3.20.   Contribution to the change in the public debt ratio (pps. of GDP) 37

3.21.   Public interest expenditure and implicit interest rate  37

3.22.   Spread of selected government bonds vis-à-vis German bunds (10y) 38

3.23.   Gross debt projections (% of GDP) - sensitivity analysis 39

3.24.   Debt decomposition, all sectors (non-consolidated) 41

3.25.   Total debt decomposition (consolidated) 43

3.26.   Breakdown of private sector debt (consolidated) 43

3.27.   Leverage of households 44

3.28.   Debt service to income ratio (households with debt payments) by income  44

3.29.   Debt to assets ratio by income level 45

3.30.   Net assets decomposition, all Sectors (non-consolidated) 47

4.1.     Distribution of taxation and overall tax burden (2011) 49

4.2.     Revenues from property taxes (2011, % GDP) 50

4.3.     Revenues from environmental taxes (2011, %GDP) 51

LIST OF Boxes

3.1.     Report of the Expert Group on Competitiveness and Employment 23

3.2.     Reduction of VAT on electricity for households 26

3.3.     EPL: harmonization of blue and white collar legislation  35

3.4.     The sixth reform of the Belgian State  40

3.5.     The Belgian financial sector 42

3.6.     The Belgian housing market 46

LIST OF Maps

No table of contents entries found.

In April 2013, the Commission concluded that Belgium was experiencing macroeconomic imbalances, in particular as regards developments related to external competitiveness and public indebtedness. In the Alert Mechanism Report (AMR) published on 13 November 2013, the Commission found it useful, also taking into account the identification of imbalances in April, to examine further the persistence of imbalances or their unwinding. To this end this In-Depth Review (IDR) provides an analysis of the Belgian economy in line with the scope of the surveillance under the Macroeconomic Imbalance Procedure (MIP). The main observations and findings from this analysis are:

· The twofold trend of a deteriorating current account balance and worrying export market share losses continues. The orientation towards less dynamic, geographically close export markets and declining cost competitiveness keeps on driving these trends. While product composition is considered a more neutral contributor, a specialization towards products situated at the middle range of value chains has not been able to compensate for these losses. A further expansion of the innovation scope and a deeper penetration of technology in products remain therefore crucial to reconcile sustainable job creation with Belgium's high labour costs. The strongly positive net international investment position mitigates for the risks of a deteriorating external position.

· Labour costs are considered to play a prominent role in the loss in external cost competitiveness. Especially manufacturing activities are confronted with the consequences of fast wage growth on their ability to compete internationally, which is reflected in eroding margins and job destruction. The Belgian authorities have initiated several measures to prevent the relative wage gap from widening, though their effect will only materialize slowly while the scope of the identified challenge calls for the instigation of a higher sense of urgency. As a consequence, narrowing down the gap effectively in order to preserve Belgium's manufacturing basis would require more ambitious and more timely action, including the long-awaited tax overhaul with a rebalancing towards non-labour taxation and making wage formation more responsive to economic realities and sectorial differences.

· Persistent problems with regard to the functioning and the adjustment capacity of the labour market contribute to the overall problem of competitiveness and erode the economy's growth potential. The overall employment rate stagnates at a below average level. Moreover, labour participation of specific target groups and in certain geographical areas is very low, resulting in higher risks of poverty and social exclusion. Driving factors are understood to be the disincentivizing fiscal treatment of labour and its interaction with unemployment benefits, skill and geographical mismatches, suboptimal activation policies, the widespread use of early and pre-retirement schemes and wage rigidities due to the wage setting mechanism. Government measures have not been able to address the persistent malfunctioning of the labour market.

· Public debt remains high, at around 100% of GDP, but a number of factors temper related macro-economic risks and government action prevented a further rise since last year's IDR. Contingent liabilities related to guarantees provided to the financial sector have been reduced, but their level remains sizeable (12% of GDP), entailing the risk of spill-overs from the financial sector to the government sector. A large share of the Belgian public debt is domestically owned, which in case of financial stress on the sovereign would further reinforce the negative feedback loops. Current financing costs are low and the risk of fiscal stress in the short term appears limited thanks to a relatively long average maturity of public debt. However, a sustained period of higher interest rates would have a substantial budgetary cost, complicating the gradual return to a balanced budget. Macro-economic risks related to this high public debt level are mitigated by the healthy state of the private sector: the net asset position of households is very positive and the indebtedness of non-financial corporations is relatively contained when excluding intra-group loans. In sum, while the government carries a large debt, the net asset position of the Belgian economy (+45% of GDP) is positive as compared to a negative net asset position in the euro area (-24% of GDP).

The IDR also discusses the policy challenges stemming from these developments and what could be possible avenues for the way forward. A number of elements can be considered:

· Wage formation is a strand for which the Belgian authorities could unfold a more ambitious reform agenda. A first element would be the tightening of the Law of 1996 in order to prevent future problems from arising, including by the inclusion of a direct link between margins for wage growth and productivity and the introduction of a higher degree of sectorial differentiation. Secondly, several options would render the practice of automatic wage indexation less harmful in times of economic distress by reconciling short-term concerns of individual purchasing power protection with the longer-term concern of safeguarding overall employment.

· The current approach of repeated, small reductions in the parafiscal pressure on labour could be replaced by a global rethinking of the design of Belgian taxation. This would entail a revenue-neutral shift in the tax burden from labour to other sources of revenue, both at the federal and the regional level. VAT and PIT carry important scope for tax base widening through the reconsideration of high tax expenditures. In addition, an increase in the standard VAT rate may be considered as this would generate substantial revenues with limited spill-over effects. Other areas that may be explored in a global redistribution of the tax burden include (recurrent) property taxation, the use of currently underdeveloped aspects of environmental taxation (including the fiscal treatment of company cars) and the taxation of certain types of financial income.

· Squaring Belgium's high labour costs with sustainable job creation and the conservation of a high standard of living would require the strengthening of non-cost competitiveness through a reorientation towards products with a higher technological content than is currently the case. To nurture the innovative prowess of the Belgian economy and facilitate the growth of innovative companies, policymakers could further enhance the emphasis on support to clusters and aim for a simplified subsidy structure in order to stimulate business spending on research and development. Another crucial factor to accelerate the transition towards a knowledge-based economy will be to assure the labour market of sufficient, aptly skilled workers in order to address skill mismatches.

· The reduction of the public debt burden continues to represent one of the country's main policy challenges in the medium to long term, all the more given the lower growth potential of the Belgian economy, the high projected impact of ageing in Belgium and the already high level of taxation. So far, strategies to reduce the public debt level were partly built around the sale of assets. However, further expenditure-based deficit reduction in itself, in line with the commitment by Belgium to reach its Medium Term Objective in 2016, will remain necessary in order to put the debt ratio on a sustainable downward path. In this respect, the increased fiscal decentralisation requires a rigorous implementation of agreed fiscal coordination arrangements and a balanced contribution by all layers of government to the consolidation. Also the implicit debt associated with an ageing population would have to be addressed by further pension reforms in order to prevent new increases of the debt level in the medium term. Simultaneously, as a tax shift away from labour to other sources would help boost employment and growth, it would be supportive for deficit and debt reduction. Making the taxation of savings gradually more neutral with respect to the type of financial product could diversify the channels through which the large financial assets of Belgian households are allocated to the real economy.

On 13 November 2013, the European Commission presented its second Alert Mechanism Report (AMR), prepared in accordance with Article 3 of Regulation (EU) No. 1176/2011 on the prevention and correction of macroeconomic imbalances. The AMR serves as an initial screening device helping to identify Member States that warrant further in-depth analysis to determine whether imbalances exist or risk emerging. According to Article 5 of Regulation No. 1176/2011, these country-specific “in-depth reviews” (IDR) should examine the nature, origin and severity of macroeconomic developments in the Member State concerned, which constitute, or could lead to, imbalances. On the basis of this analysis, the Commission will establish whether it considers that an imbalance exists in the sense of the legislation and what type of follow-up in terms it will recommend to the Council.          

This is the third IDR for Belgium. The previous IDR was published on 10 April 2013 on the basis of which the Commission concluded that Belgium was experiencing macroeconomic imbalances, in particular as regards developments related to external competitiveness and public indebtedness. Overall, in the AMR the Commission found it useful, also taking into account the identification of an imbalance in April, to examine further the persistence of imbalances or their unwinding. To this end, this IDR takes a broad view of the Belgian economy in line with the scope of the surveillance under the Macroeconomic Imbalance Procedure (MIP).

Against this background, Section 2 starts with an overview of the general macroeconomic developments while Section 3 looks more in detail at how the earlier identified imbalances and their underlying risks have evolved. This is followed by a specific focus on the tax system in Section 4 and policy considerations in Section 5.

Potential growth down from pre-crisis level

Notwithstanding a cautious improvement of economic performance in the second half of 2013, growth is not projected to reconnect with pre-crisis rates. The growth potential of the Belgian economy is currently estimated a little below 1%, about half of the pre-crisis level, and would rise only modestly over the medium term. While the contribution from all components has fallen over the past decade, the very low gains from total factor productivity are particularly notable. This points to problems regarding the capacity to engender innovation and efficiency gains.

Potential growth would also benefit from an improvement in the dismal activity rate, which reached 66.9% in 2012. Belgium's low activity rate relates to a number of lasting economic problems identified in previous vintages of the European Semester. Factors such as high labour taxation, persistent unemploy­ment and inactivity traps, and modera­tely effective active labour market policies are understood to put a brake on labour supply. Conversely, on the demand side job creation is hampered mainly by high labour costs. These labour costs are also a key element within the context of deteriorating external cost compe­titive­ness, which has become more prominent and pressing in recent years and is discussed in section 3.1. To address the underlying drivers, the govern­ment took a first series of measures in recent years. These include incentives to take up work by means of altered unemploy­ment benefits and the tax treatment of the lowest wages, the restriction of access to early retirement and reductions in social security contributions.

While broadly stable, social indicators suggest some areas for concern. Belgian elderly continue to face a somewhat higher risk of poverty or social exclu­sion than on average in the EU. For children this risk has been rising as well. Furthermore, the early school leaving rate has increased in recent years. The number of people living in households with low work intensity is above the EU average and increasing. This contrasts with a still low in-work at-risk-of-poverty rate, which underscores how inducing higher employment would not simply raise the economic potential, but would also help to shield against risks of social exclusion.

The Belgian economy has been reorienting towards the services sector. Between 2000 and 2012 the industry (excl. construction) shed 110,000 positions while service sectors added 477,000 jobs, mainly in professional, scientific and technical activities, and administrative and public services. There are no signs of this shift slowing down immediately. This transformation of the economic landscape also comes to the fore when looking at the sectorial decomposition of gross value added (GVA) with the manufacturing and services sectors respecti­vely losing and gaining around 7 pps. between 1996-2012. As regards the income decomposition of GVA a stable distri­bution among compensation of labour on the one hand and capital on the other hand appears, hove­ring respectively around 58% and 42% of GVA.

External position remains strong, despite wea­kening current account balance

Whereas positive net exports prevented a sharp downturn in 2012-13, this disguises a longer-term trend of a deteriorating current account balance. According to estima­tions by Commission services, the current account deficit (BoP definition) continued to widen in 2013. Aside from this widening deficit, also the longer-term evolution of the current account catches the eye. A surplus of around 5% of GDP at the turn of the century dwindled gradually until 2008, when it turned negative([1]). Estimates for the cyclically-adjusted current account balance (i.e. the level that would prevail if both the domestic and trading partner economies were at potential output) indicate that only a fraction of the deterioration can be attributed to cyclical factors.

The deterioration in the current account is predominantly driven by the downward evolution of the goods balance. The latter deteriorated by 6.5 pps. over 2002-12, reflecting both a price and a volume effect. The substantial surplus of about 2% of GDP which trade in services continues to record has not been able to compensate for the downward trend at the goods side and seems to have stabilised recently following years of steady expansion. The slow shift in recent years of the capital income balance from a modest surplus to a limited deficit has its origin mainly in higher net outflows linked to direct investments and reflects the effect of a low yields environment on the asset structure.

At almost 50% of GDP, Belgium's overall external position as measured by its NIIP can still considered to be sound. The Net International Investment Position (NIIP) is the balance between external financial assets and liabilities. Whereas the current account balance equals flows, the NIIP can be interpreted as the stock indicator of a country's external position. A stabilisation of the current account near its present level with nominal GDP growth at around 3.5% would entail a gradual reduction in the NIIP, though the latter would nevertheless remain at comfortable levels. Even in case the long-term trend is confirmed and a more substantial current account deficit would emerge, the accumulated net external assets imply that, at the current juncture, Belgium's current account deficit can be considered sustainable in the medium term.

While external sustainability does not seem to be a major risk, a closer look to the robust NIIP reveals that internal equilibriums are neverthe­less affected. The strongly positive NIIP reflects a comfortable creditor status by the private sector, offsetting the structural debtor position of the public sector. The current account corresponding to the difference between domestic savings and domestic capital formation, the deterioration over the past ten years went hand in hand with lower savings and broadly stable investments. The decrease in total savings has in turn been primarily driven by lower household savings with consump­tion growing faster than disposable income.

Domestic debt mainly a matter of public debt

Following a sustained reduction in the debt burden until 2007, public debt has risen again to about 100% of GDP. Behind this reversal of debt dynamics lie a deterioration of public finances and the repeated need for interventions in the banking sector. While the recent return to primary surpluses and the sale of assets is expected to have stabilised public debt, non-negligible contingent liabilities to the financial sector and high projected ageing costs still threaten the medium-term outlook of public finances.

Private sector indebtedness does not appear to be a major concern but should be read against a background of high public debt. Non-financial private corporate debt is high in non-consolidated terms but when national intra-company loans are excluded, debt is comparable to the EU average. Household debt is close to the EU average and is mostly mortgage related. Last year's IDR presented a thorough assessment of the housing market and found that a potential house price correction would most likely not provoke major macroeconomic repercus­sions. Indeed, the fast house price increase between 2000 and 2011 was not accompanied by a higher interest burden for households, nor by an excessive house supply. Furthermore, households' average annual redemption effort has remained broadly stable as a consequence of wages growing faster than loan burdens and a preference for financing at fixed rates. There is also no practice of remortgaging in Belgium. Finally, risks for the economy appear contained with a low default rate, which, at 1.2% in 2013, has remained stable in recent years.

Strengthening of financial sector continues

After recovering somewhat, credit growth has been declining in the last two years but remains above the EU average with lending rates to non-financial corporations among the lowest in the EU. This lower credit growth reflects on the one hand declining financing needs and on the other hand a tightening of credit standards, driven by new regulation and capital requirements in the financial sector.

The financial sector has continued to gradually strengthen and total financial assets have decli­ned significantly. The continuing restructu­ring in the sector resulted in a rebalancing from external and inter-bank activities towards more traditional activities and the domestic market. Capital injections have strengthened capital ratios but profitability is still low and non-performing loans have increased amid weak economic activity.

In May, Belgium will stage federal and regional elections. While this may create a reformist window of opportunity for the next government, the stalemate at the federal level following the last two elections (2007 and 2010) with a caretaker government in charge for respectively 6 and 18 months, gives rise to concerns that Belgium may face a new period of political procrastination. This would result in reforms being further adjourned and may trigger renewed financial market pressures as was the case end-2011 when the spread on 10-year Belgian government bonds relative to German bonds rose to 360 base points.

3.1.      Competitiveness

As highlighted in section 2, Belgium's trade balance for goods has been displaying a steady decline. While the current deficit level by itself does not point to an emerging imbalance, the evolution is nevertheless symptomatic for an underlying problem of weakening competitiveness, which has been identified as an imbalance in earlier IDRs and is the subject of section 3.1. Addressing this core issue would therefore also soften the negative trend on the current account.

One of the most prominent manifestations of Belgium's deteriorating competitiveness is the dwindling of export markets shares over the past decade. These losses did not coincide with the economic downturn of 2008. Between 2003 and 2012 they fell on average by 3.2% annually with only 2007 and 2009 seeing (modest) gains. Moreover, the pace has accelerated in recent years with an average annual decline by 5.3% during 2009-12 for goods and services and 5.9% for goods. As highlighted in last year's IDR, market shares for services have been broadly stable, but account only for a limited share in total exports.

Most EU15 countries have been recording export market share losses but for Belgium these losses have been more pronounced. The country has seen its share in total European exports declining (see Graph 3.1). As a consequence, the Belgian experience cannot be interpreted as a mere reflection of shifting global (or regional) economic clout. In fact, Belgium's share in global economic activity, which could be considered a rough proxy for expected export share evolution given Bel­gium's open economy, has remained remarkably stable and even gained some ground in recent years. In this respect, economic performance over the past ten years did probably not fully reflect underlying dynamics given the worsening external position.

Export growth falling short of world export growth can be rooted in three broad factors. (i) a product specialisation in goods for which demand is progressing relatively slowly, (ii) a geographical orientation towards slower growing markets or (iii) the inability to compete effectively internationally, i.e. a loss in competitiveness. A decompo­si­tion of export growth gives insight into the importance played by these different drivers. A common and instructive approach in this regard is a 'constant market share analysis'. The results of such a basic shift-share analysis for Belgian good exports for the respective periods 1997-02, 2002-07 and 2007-12 are visualized in Graph 3.2. Belgian exports of goods grew faster in 1997-02 than global imports, while they fell short in the subsequent periods. The initial gain in export market shares was predominantly driven by competitiveness (in a broad sense). Subsequently, the contribution by competitiveness factors turned negative, triggering a steady decline in export market shares since the turn of the century. Geographical orientation seems to have become another important driver behind overall export losses, indicating a disproportionate orientation towards less dynamic export markets. The third subcomponent, product composition, shows a slightly negative contribution for all periods considered. This implies that the Belgian export configuration is broadly aligned with global patterns, but also that Belgium fails to specialise in products characterized by a more dynamic demand, which could have limited overall export market share losses.

The following sections take a closer look at the drivers behind the observed deterioration in Belgium's performance on international markets. Product specialisation, export orientation, cost-competitiveness and labour market aspects are discussed respectively.

3.1.1.   Product specialisation

Though factor prices usually play an important role in a company's ability to compete success­fully on international markets, the kind of products that are fabricated also matters. Producing and trading a range of (finished) products with a higher value added makes it easier to pass on labour costs to final prices as competition for this type of products is mostly based on non-price aspects such as quality (ECB, 2013). As mentioned, Belgium's product composi­tion does not appear as particularly problematic, but at the same time it has not been able to compensate for deteriorating competive­ness or weak export market growth. Recent data confirms last year IDR's finding that Belgian companies export relatively little capital goods, less than half of the share of the average euro area country (see Table 3.1). Conversely, intermediate goods have a higher share in manufactured exports compared with other euro area countries, which is also the case for imports. This exposes Belgian exports relatively more to international competition on the basis of cost factors. A high share of intermediate goods is for example exported to Germany([2]) with Belgian exporters having to compete with companies from newer EU member states and emerging markets. These countries are generally better positioned to compete on cost terms and, contrary to Belgium, have seen their share in German intermediate imports rising over the past decade.

The balanced findings of last year's section on Belgium's innovation capacity remain valid. Total R&D spending has risen in recent years and reached 2.2% of GDP in 2012; up from 1.9% in 2002 but still falling considerably short from the EU2020 target of 3%. Compared with Scandina­vian countries and Germany a substantial gap remains due to comparatively lower R&D spen­ding in Belgium by both businesses and public entities. It should be noted though that enhanced public support for R&D has borne fruit (FPB, 2013), which bodes well for a further intensifica­tion of business R&D outlays. Nevertheless, busi­ness R&D expenditures are highly concentrated in a small number of (international) companies in essentially one sector (pharmaceuticals), which narrows the base for innovation. In the case of research, labour costs are generally not the most important factor. Moreover, the issue has been addressed by initiatives such as the federal payroll tax incentive to lower R&D personnel costs. Yet, for actual manufactu­ring activities that could derive from performed research, labour costs are of higher importance. Finally, R&D and the develop­ment of activities with a higher value added are hampered by skill availability, which helps to explain a relatively low penetration of researchers in overall employment (see Graph 3.3). While Belgium disposes of an overall qualified work­force, the number of graduates in science, mathe­matics, engineering and technology (STEM) remains low.([3]) The creation of substantial high-tech em­ployment in Belgium over the past decade (Vives, 2013), highlights the importance of adequate in­flows of accurately qualified workers, all the more as research points to the high potential of job creation in the slipstream of  high-tech employment.

Competition in services in Belgium is limited by a number of regulatory barriers. Service provi­ders face obstacles when entering the market and offering services, for example different authorisa­tion systems are in place at regional level. The retail sector continues to be subject to a number of regulations and operational restrictions which hinder its development and means retail prices in Belgium are more than 10% above the euro area average.

3.1.2.   Export orientation

As indicated by the above shift-share analysis, Belgium's orientation to less dynamic markets has been hampering export performance. Earlier analysis pointed to the high concentration of exports to European countries, particularly the neigh­bouring ones. Graph 3.4 highlights that Belgium is mainly exporting towards markets that are growing slower than global imports and that it has also been experiencing competitiveness losses vis-à-vis its competitors on these markets.

Nevertheless, the importance of traditional export markets has gradually decreased. This has been especially true in recent years with exports becoming more diversified and the direct relevance of markets such as for example India, China or Turkey rising rapidly (see Table 3.2). While, on the one hand, this reflects slower export growth to traditional markets, on the other hand, export growth to emerging markets has also accelerated in general, implying that the gradual reorientation does not merely reflect a relative shift as a conse­quence of sluggish import growth across Europe.

Moreover, the share of exports going to Euro­pean countries in general and to neighbouring countries in particular, is less important when data are corrected for re-exports. The latter are high for countries with large international ports such as Belgium and the Netherlands. This also comes to the fore when exports are measured in value added terms (NBB, 2013). In 2009, the last year for which data are available, the EU27 represented 64.6% of total value added exports as compared with 75.5% when looked at gross exports. Especially the relevance of the three neighbouring countries falls notably, while EU12 countries but also Italy and the UK become more prominent, as do for example the United States with third countries' exports comprising significant Belgian value added.

Nevertheless, even when measured in value added, Belgian exports have been losing global market shares. Between 1995 and 2009 the loss amounted to 26%. This compares to more shallow EA12 and EU27 losses of respectively 15% and 3% over the same period.([4]) 

3.1.3.   Cost competitiveness

A. Labour costs

An appreciating real effective exchange rate (REER) points to a loss in relative cost competitiveness. As the deterioration of the current account and the loss in export market shares have their origin in the export of goods, the REER deflated by unit labour costs in the manufacturing sector can be considered the most appropriate gauge of cost competiveness for Belgium.([5]) Graph 3.5 indicates a sharp deterioration of relative labour cost competiti­veness in the manufacturing sector since the turn of the century; i.e. an appreciating REER. Other, less narrow measure­ments of cost competitiveness are also included in Graph 3.5: REER deflated with consumer prices and with export prices. All three run broadly parallel until 2006 when a decoupling appears. The REER deflated with ULCs in the manufacturing sector seems, however, the most relevant for mentioned reasons.

REER correction through nominal exchange rate adjustment is no longer possible for Belgium as the euro exchange rate reflects dynamics at the aggregate euro area level. As a consequence, and as recent experience by some countries confirms, a too divergent evolution of a country's relative competitiveness could generate imbalances which in turn may trigger strained corrections or result in welfare losses if left unaddressed. Therefore, it is paramount to assure that competitiveness developments are aligned with those in trading partners.

Relative wage dynamics are a key determinant of Belgium's cost competitiveness. Two features characterize the Belgian wage-setting system: a national, two-year wage norm which determines the maximum real wage increase in the private sector based upon which negotiations take place at lower levels, and the quasi universal practice of automatic wage indexation based on the so-called 'health index'.([6]) As highlighted in previous IDRs, wage costs and productivity have decoupled in recent years. While still high in absolute terms, average labour productivity growth in 2007-12 has been around zero. At the same time, nominal wage growth has advanced fast under the impetus of wage indexation, driven in turn by high price pressures. As Graph 3.6 shows, prior to the 2008-09 crisis Belgian ULCs rose faster than those in the aggregate euro area and main trading partner countries, though not alarmingly. As of 2009, however, a marked deviation emerges compared with the EA17, in spite of accelerated ULC growth in leading trading partner Germany. This reflects measures in other countries such as Spain to curb ULC growth. Current projections indicate that ULC growth in Belgium would be lower in 2014-2015 as compared to preceding years. However, this will be insufficient to correct for the deviation accumulated in the past. If left unchecked, this deviation may further reinforce the long-term trends of capital deepening through automation or outright delocalisation of activities, both of which have a negative impact on employment with less productive labour being pushed out of the labour market (see below).

Belgium is one of the Member States where the practice of linking wages to cost-of-living increases is most widespread. This has the merit of ensuring social peace but also carries negative economic implications. While wages in other countries do follow domestic price evolutions over the medium term as past experience has demonstrated, the absence of an automatic adjustment makes purchasing power corrections dependent on wage negotiations. As Bodart & Shadman (2013) point out, the outcome of such negotiations will be a function of the business cycle and the situation on the labour market, making adjustment in the short term generally less rigid than is the case in Belgium. This helps to explain why hourly wages grew by 3.3% on average annually between 2007 and 2012 in Belgium, compared with 2.3% in Germany and the Netherlands and 2.6% in France and the euro area. Wages in Belgium have been adjusted faster to (higher) price pressures, despite the economic environment. In the short term, wage indexation in its current form may thus affect competitiveness, leading to job and export market share losses which may become permanent over the longer term as a new equilibrium arises (Bodart & Shadman). Moreover, as Bogaert & Robette (2013) and Bodart & Hindriks (2013) point out, lower incomes have suffered considerable purchasing power losses despite the indexation mechanism.

When in November 2012 the Belgian govern­ment stated its goal to close the accumulated wage gap with its main trading partners by 2018, there was no agreement on the precise size of this gap. The latter is commonly measured for policy goals in terms of the evolution of hourly labour costs in the private sector since 1996 (when the Law on Competitiveness and Employment was adopted) compared to Germany, France and the Netherlands.([7]) However, this measurement only includes direct reductions of social security contributions, not the 'wage subsidies' to which the Belgian authorities have taken increasingly recourse over the past decade. In addition, only an aggregated calculation of the wage gap is used, regardless of sectorial dispari­ties. Finally, when looking at hourly labour costs in isolation, productivity dynamics are put aside. These shortcomings, in addition to the arbitrariness of the year of reference, have stirred debate on exactly how large the effective wage gap is. In an attempt to straighten out the lasting controversy, the authorities set up an Expert Group, which published a first report in July 2013. Though it did not manage to end the conceptual muddle that has blurred the debate with regard to labour costs developments in Belgium, the report discusses numerous relevant issues (see Box 3.1). The Belgian government has commissioned a follow-up report by June 2014.

As the Expert Group report points out, in parti­cular Belgium's industrial sectors and more specifically manufacturing activities are faced with a serious problem of deteriorating cost competitiveness. This is well illustrated by the evolution of profit margins. The average cross-industry margin has been hovering between 25% and 30% in recent years. While this compares favourably to the situation in France, it is below margins realized in Germany and the Netherlands, even if the difference with the latter two is limited and constant. Moreover, the stability of the average margin disguises a trend of deteriorating performance by manufactu­ring activities, with margins notably below those in Germany and the Netherlands, where they have been increasing in recent years (see Graph 3.7). If (all) wage subsidies are taken into account, margins improve somewhat, though the negative trend remains, as does the significant gap with the Netherlands and Germany, where wage subsidies play only a marginal role.

Belgium combining high hourly labour costs with a high level of productivity, especially in manufacturing, raises the question to what extent high wages drive high productivity. Indicators point to the presence of capital deepening in Belgium ([8]) as can also be seen in Graph 3.8.([9]) This suggests that the high aggregate level of productivity is to some extent the result of labour being replaced by capital inputs and a steady shift towards activities at the higher end of the salary scale with ever more capital intensive production techniques being applied and the average wage rising together with the share of qualified labour. This, however, has negative effects on employment, with less skilled labour being pushed out of the labour market and a higher risk of labour market mismatches. The higher level of productivity thus partly explains Belgium's low labour participation, in particular among low-skilled workers (see section 3.1.4)

What is more, the capital deepening of the Belgian manufacturing industry seems not to have put a brake on the long-term trend of deindustrialisation. Graph 3.9 shows how the (manufacturing) industry in Belgium has lost relatively more ground in terms of total value added than in other countries. Although a comparable loss is seen in France, the Netherlands and Germany have been able to stop and even reverse the downward trend, demonstrating that it is not inescapable. Belgium's share in the total value added in the euro area fell by 11% between 2000 and 2012, a less bad performance than France (-20%), but compared to +12% and +9% for Germany and the Netherlands.

The already high degree of capital deepening lowers the scope for future productivity gains, which erodes potential growth. This is of particular concern as market services have not been able to realise a strong productivity growth and the attained level is still lower than in the industry (NBB, 2013). It may also entail the risk of accelerated delocalisation in the future.

The strong bias towards job creation in the non-tradable sector with the help of wage subsidies and public sector jobs has been compensating for steady net job destruction in the industry. The crisis years only accelerated a process of net job losses in most industrial sectors. This has been especially true for the car assembly industry, textiles, metallurgy, chemicals and non-ferrous metals, all labour-intensive sectors (Vandekerckhove, Struyven & Heylen, 2013).

B. Government action with regard to cost competitiveness

In order to avert the relative wage difference with neighbouring countries from widening further, the Belgian authorities imposed a freeze of real wages (i.e. zero wage norm) on the social partners for 2013-14, a decision which will most likely have to be extended in 2015-16. Other actions to curb the wage gap included a series of modest reductions in social security contributions for employers – as elevated labour costs are rooted in high taxation – and method­logical interventions on the measure­ment of price pressures by the health index.([10]) Price pressures as measured by the standardised HICP have on average been 0.4 pp. higher than in neighbouring countries over the past decade (see Graph 3.10). This difference can to a large extent be attributed to energy products, with an average price differ­rence of 1.3 pps. over the same period. Fluctua­tions of international energy prices have a more pronounced effect on the Belgian economy([11]) with the impact of an oil shock on competitiveness being larger than in the neighbouring countries (Bodart & Shadman, 2013).

Government measures to foster competition and increase transparency on domestic energy markets have, however, reversed this price differential since mid-2012. This has resulted in overall inflation converging with euro area inflation and the average price growth in neigh­bouring countries (see Graph 3.10). This reduced inflationary pressure may make it somewhat easier for Belgium to achieve wage moderation given the reduced impact of wage indexation on total wage growth. However, like in past years, core inflation continues to reach a higher level than in neighbouring countries. This could be attributed both to suboptimal functioning in certain product markets (EC, 2013) and to second round effects from a widespread practice of price indexation across sectors (e.g. rent prices, insurance policies).

At the end of 2013 the Belgian authorities announced an inter-federal 'Competitiveness and Employment Pact'. At the federal level, this included a reduced VAT rate on electricity (see Box 3.2) as well as three additional structural reductions of social security contributions (in 2015, 2017 and 2019) of EUR 450mn each, one third of which would be allocated to linear reductions, one third to low wages and one third to sectors exposed to international competition and whose growth potential is at risk because of the evolution of labour costs compared to productivity.  These (prospective) reductions would be accompa­nied by some smaller measures targeting among others SMEs, shift work and young unemployed. As part of the Pact, the Regions also announced some initiatives, including a EUR 125mn budget to reduce wage costs for employees under 30 and above 55 in Flanders.

Many of the announced measures carry an important, though unfunded, future budgetary impact. This is the case for the triple reduction by EUR 450mn of employer's contributions([12]) as well as the lowered VAT rate on electricity if made permanent. Putting in place accompa­nying financing measures would have made them more credible and, even more important, reconci­led the outgoing coalition's ambition to improve cost competitive­ness during the next legislature – the triple reduction has been stipulated through legislation – with the already very tight budgetary margins during that period.

Aside from measures to correct past excessive wage growth, the government has unsuccess­fully sought to alter the Law of 1996 to prevent future decoupling of wages and produc­tivity. Several reform options are reported to have been discussed but negotiations have stalled as a conse­quence of disagreements within the coalition government and the reform has eventually not been included in the 2013 Competitiveness Pact. As a consequence, the reform is likely to be post­poned until after the May elections. In case of a renewed protracted period of government negotia­tions, this would mean that the new Law may not be in place in time to be applicable to the 2015-16 Inter-Professional Agreement. One of the reform options discussed would introduce a more stringent method to set the wage norm with social partners being legally obliged to factor in past relative wage evolutions in order to set the norm in such a way that the accumulated wage gap would be progress­sively narrowed. Another proposed modification would require the government to intervene in a legally binding way whenever partners do not come to an agreement or fail to comply with the stricter stipulations of the Law. More severe sanctions would also be introduced for employers in case of non-compliance with the wage norm by lower-level collective agreements.

If adopted such reforms would make the existing wage bargaining framework more coercive, though there would still be important shortcomings. A first one is the lack of a link between wage setting and productivity develop­ments in order to safeguard employment over the longer term. Yet, the existence of sizeable sectorial differences with regard to productivity means that introdu­cing productivity in the wage setting at the inter-professional, national level would seriously distort figures. Incorporating it at lower levels would therefore be more appropriate. This would require a more decentralized procedure than currently applied with the determination of a national wage norm. Such decentralisation would enable industries and companies to square wage policies more aptly with their performance and allow for labour inputs to be reallocated more gradually from declining activities towards sectors in expansion.

Furthermore, even if the national wage norm would be made more stringent, the wage gap could still widen as a result of the pervasive practice of automatic wage indexation applied at the sectorial level. While the current frame­work caps real wage growth, it does not interfere with the freedom of social partners to conclude wage indexation agreements at the level of their preference. As a consequence, an important part of the wage growth margin falls beyond the scope of the Law on Competitiveness and Employment and is excluded from wage negotiations, which encumbers the economy's adjustment capacity. To mitigate the risk of inflationary pressures fuelling nominal wage growth beyond the rise in relative produc­tivity levels vis-à-vis neighbouring countries, one could consider a system where the financial equivalent of increases in the health index is only partially allocated to nominal wage increases while the remaining part goes to other forms of (non-wage) compensation and/or invest­ment in training and human capital. All-in agreements capping total wage growth would produce similar results.

C. Energy costs([13])

Energy being a key input in many production processes, its relative direct costs represent a factor of competitiveness for the industry. As a consequence, in addition to labour costs, also energy inputs may be behind the observed loss in cost competitiveness. However, another equally important factor is the intensity of use. In order to provide a more comprehensive assessment of the role that energy plays with respect to industrial competitive­ness, both factors need to considered by looking at Real Unit Energy Costs (RUEC), i.e. the amount of money spent on energy (including non-commodity elements such as taxes) in order to obtain one unit of value added. For the manufactu­ring sectors the level of RUEC in 2011 in Belgium is comparable to that in the Nether­lands, but higher than in France and, particularly, Germany (Graph 3.11). According to estimations this was still the case in 2013.([14])

Disentangling the change in RUEC during the period 1995-11 by means of a shift share analysis shows that in Belgium the increase in RUEC in the manufacturing sector is for a comparatively large part driven by restructu­ring and interaction effects. This indicates that to a larger extent than in the neighbouring countries, there has been an increase in the shares of manufacturing industries with (i) relatively high initial energy costs or with (ii) energy costs that rose rather strongly. This suggests that it is the production structure, rather than the energy unit costs, which causes the comparatively high RUEC in Belgium. This is confirmed by counterfactual analysis which shows that, under the assumption that manufacturing sector shares in neighbouring countries would be identical to the actual shares in Belgium, RUEC in the manufacturing sector would actually be higher in the neighbouring countries than in Belgium in 2011 (see Graph 3.12).

3.1.4.   Functioning of the labour market

The impact of the crisis on employment has been relatively contained. This is partly due to the stepping up of short-time working schemes and a more intensive use of job support and activation measures in recent years. Unemployment rose from 7.6% in 2012 to 8.4% in 2013, compared to the EU28 average of 10.9%. However, labour market performance is characterised by vast disparities among population sub-groups (see Graph 3.13) and areas (see Graph 3.14).([15]) As a result, tight labour markets and growth-hampering skill shortages in certain areas and sectors coexist with persistently high unemployment in others.

These deep-rooted labour market problems continue to result in a chronic underutilisation of labour. Activity and employ­ment rates remain stagnant and below the EU28 average (72.5% vs. 76.5% and 67.2% vs. 68.2% respectively in 2012). Long-term unemployment as a percentage of total unemployment is consistently high and has been persistently higher than the EU28 average for several years. It is only recently, mainly as a result of the heavy toll of the crisis on the labour market of a number of member states, that the long-term unemployment rate has converged with that in the EU28 (45.9% in 2012 vs. 46.1%).

The low-skilled, the young, the elderly and third-country nationals are the sub-groups least active in the labour market, with levels well below the EU average (see Graph 3.13). According to estimations by the OECD (2013) the fiscal impact of bringing employment rates for third-country nationals on par with nationals would be close to 1% of GDP, the highest among EU countries. Furthermore, as can be read from Graph 3.14, high geographic employment and unemploy­ment disparities persist, with for instance Brussels registering an unemployment rate four times that of the neighbouring province of Flemish Brabant.

The middling performance of the Belgian labour market is rooted in a number of factors. These  include 1) the largest (and still increasing) tax wedge on labour in the EU, which, through its interaction with the benefit system, creates sizeable unemployment and inactivity traps for most categories of workers (cf. below); 2) considerable vertical and horizontal skill mismatches([16]) hampering the efficient allocation of resources in the labour market; 3) only moderately effective activation policies; 4) a widespread though diminishing use of early and pre-retirement schemes and 5) the wage setting system leaving limited room for differentiation between industries, taking away the possibility for more productive sectors to use wages to attract qualified labour. In addition, some common practices such as rewarding sector seniority with higher salaries, potential severance payments and early retirement options continue to discourage mobility.

These various factors contribute to modest potential growth by undermining labour utilisation and its optimal allocation, with adverse effects on overall productivity, the adjustment capacity and, ultimately, job creation. In addition, the labour market's underperformance puts a strong pressure on the social security system and on public finances in general. In the context of the macroeconomic imbalances procedure, the focus will be on unemployment benefits and active labour market policies, which are more closely linked to public finances and potential growth.

In order to increase incentives to take up work the Belgian authorities have enacted a reform of the unemployment benefit system and taken targeted measures to reduce the tax wedge on labour. Thus, for example the so-called 'work bonus' for the lowest wage earners has been increased several times.([17]) While it can be expected that these measures will somewhat reduce the unemployment and inactivity traps at the bottom of the pay scale, their effect is not yet fully visible in the latest available data.([18]) Nevertheless, it is clear that the historic unemployment traps in Belgium are both sizeable and pervasive. According to the latest figures provided by the OECD tax and benefits model([19]), they are among the highest in the EU.

Regarding the unemployment trap, out of 30 categories of workers identified in the OECD model, Belgium was scoring better than or in line with the EU average in four categories in 2001, eight categories in 2006 and in only six categories in 2012 (see Graph 3.15). In 2012, the unemployment trap for low-wage workers earning 67% of the average wage when resuming work was above the EU average for all family compositions, in particular for single persons and two-earner couples with the main earner gaining 100% of the average wage.

High unemployment traps have not escaped political attention. Over the past few years, policy makers have attempted to tackle the problem by means of targeted measures aimed at the bottom of the pay scale.  Actions taken in past years have indeed led to a decrease of the unemployment traps of very low salaries (up to 50% of the average wage) for all family types. However, the trap has widened for workers earning 67% of the average wage and above, with lone parents or singles being again those suffering from the highest financial disincentives to work. All in all, the only categories where Belgium scores better than the EU average are very low income households (i.e. 33% or less of the average wage).

When comparing to the three neighbouring countries, the picture is more mixed. According to a study by the Central Economic Council (2012), the net replacement rate in Belgium varies strongly according to the salary level, with the lowest salaries being more protected than in neighbouring countries and salaries as of 100% of the average wage less protected than in all three neighbouring countries.([20]) When taking into account top-up social assistance benefits, the average family of four will be better protected in Germany and the Netherlands than in Belgium in the case of 67% of the average wage. For 100% of the average wage, such families will be better protected initially in the neighbouring countries than in Belgium though benefit decreases reverse the situation after 23 months in the Netherlands and 25 months in France and Germany.

The recent reform of the unemployment benefit system introduces some important novelties in the design of the unemploy­ment insurance. The rate at which regular unemployment benefits are reduced over time has been made steeper as of November 2012. To this end, the net replacement rate was increased for the first six months (to 65% of the gross salary capped at a reference gross wage of EUR 2,370.8 for the first three months, and 60% of the same reference wage for the next three months) and remains at the same level as before the reform for the following six months (60% of the gross salary capped at EUR 2,209.6). After this first year, unemployment benefits decrease in several steps at a pace which depends on household composition and the length of the previous contribution period. After between 16 and 48 months, job seekers fall back on a lump sum slightly higher than social assistance benefits.([21]) Exceptions apply to temporary unemployed, certain categories of elderly unemployed or with long tenures, and unemployed with disabilities.([22]) At the same time, eligibility conditions have been relaxed by lengthening the reference periods used for the calculation of previous work experience (from 18, 27 and 36 to respectively 21, 33 and 42 months) and by assimilating to salaried work also periods of activity carried out in the framework of labour market re-integration programmes. In addition, activation policies and availability to work requirements have been stepped up. The reforms also limited the period that young people over 18 years entering the labour market can receive an 'insertion allocation' to a maximum of three years.([23])

In order to increase labour participation, it is crucial to find the right balance between, on the one hand, combining well-designed benefit systems (with decreasing benefits over time) with appropriate job search requirements and, on the other hand, effective job search assistance and training opportunities.([24]) The reform of the unemployment benefit enacted by the Belgian authorities constitutes a step towards such an appropriate balance. However, the complexity of the Belgian benefit calculation method might hamper the incentive effects of the overall system. Also, the gradual extension of the coverage of activation policies to include elderly age groups and the rising unemployment rate increases demand for timely follow-up, job search guidance and retraining services by the public employment services, which might worsen existing capacity issues.

Inactivity traps are also large and among the highest in the EU. For those who disqualify from the unemployment benefit system, or do not have access to it but are rather covered by social assistance, inactivity traps are notably above EU average for all categories of workers and family types. Data for 2012 show that very low wage earners at 30% of the average wage suffer from very high financial disincentives to work, notably if they are single parents, single persons and one-earner couples both with and without children.([25]) A study by the Public Social Welfare Centre (2012) confirms that single parents with dependent children earning close to the minimum salary will still receive more money and benefits under social assistance than when taking up work and are thus strongly discouraged from active job searching. In addition, large low-wage traps exist for most categories of workers and family types, which do not encourage low-skilled workers to work more hours or to make efforts to progress and upgrade their skills due to the loss of the benefits incurred when increasing their salary level. This in turn has an impact on competitive­ness through lower productivity growth.

Belgium's labour market is also characterized by a high level of mismatch between skills and jobs. As can be observed from the evolution of the Beveridge curve (see Graph 3.16), matching efficiency did not worsen significantly after the crisis.([26]) In the period 2008-09, vacancies fell and unemployment grew, before the mismatch was exacerbated in 2010 with higher vacancies and higher unemployment. In 2011 and 2012, unemployment decreased and vacancies increased before returning to around the 2010 level in mid-2013.

A large and persistent mismatch nevertheless exists between skills and jobs, with demand for the latter concentrated on high-skilled profiles, against a largely untapped offer of low-skilled workers. Furthermore, it appears that the mismatch is particularly severe in Brussels, even if it affects the whole country (Zimmer, 2012). As illustrated in Graph 3.17, more than 80% of the active population with tertiary education is employed in Belgium, against an employment rate over 60% for medium-skilled workers and less than 40% for low-skilled workers.

Even though the problem has been recognized by the government, initiatives taken in order to increase the number of graduates in sciences and engineering and redirect youth to technical professions did not bear fruit so far. Belgium still shows a below average rate for graduates from mathematics, science and technology studies compared to the EU (1.26% vs. 1.68% of aged 20-29 in 2011). However, it would benefit both workers and the economy if students would be directed towards these fields considering that the salary difference between low/medium-skilled and high-skilled for the same age category is sizeable (see Graph 3.18) and a higher proportion of high-skilled labourers in the total workforce would increase labour productivity.

Activation policies also matter to increase the matching between skills and jobs. Although Belgium is one of the countries with the highest expenditure on active labour market policies in terms of GDP([27]), yields remain unresponsive as can be concluded from the still below average employment rate, though at the same time the unemployment rate has remained largely constant even in the face of the financial and economic crisis. The latest reform of unemployment benefits also encompassed activation policies with a quicker and more intensive follow-up of job seekers. This is expected to improve the take-up of jobs by unemployed persons and better orientate job seekers towards sought skills.

The negative effects of unemployment and inactivity traps on labour market participation are recognised by the Belgian authorities. Steps are indeed taken to address the issue([28]) but, while their full effect still needs to materialise, it is doubtful that the enacted measures will be ambitious enough to fully solve the issue. While the unemployment trap at the bottom of the pay scale indeed somewhat decreased, large unemployment traps remain pervasive for households with slightly higher wages.

Given the already sizeable decreases of social security contributions on the lowest wages, this approach offers limited room for additional tax wedge reductions. In that respect, a thorough reform of the tax system would probably be more effective (see section 4). A tax shift from labour to other bases could indeed help making work more rewarding and incentivize unemployed persons towards re-entering the labour market. Additional adaptations could also be considered to reduce disincentives to take up work for second-earners, for example by reducing the average effective tax rate for second earners with respect to single individuals or by shifting (part of) cash-based child support to in-kind benefits in order to facilitate labour market participation. Finally, the implementation of the 6th state reform and the ensuing further regionalisation of activation policies provide the opportunity for a comprehensive review of the effectiveness of the various existing measures. This opportunity can be seized to rationalise the existing activation and employment incentive schemes and integrate them in a coherent policy tailored to the needs of the regional labour markets.

Conclusion on external competitiveness

Belgium continues to lose export market shares at a worrying pace and faster than most other EU countries. This trend is partly driven by the traditional focus towards geographically close, but slower growing, markets. While there are signs that the latter factor is slowly improving, the other important driver, cost developments, continues to burden the ability of Belgian manu­facturers to compete successfully internatio­nally. This loss in cost competitiveness surfaces through indicators such as profit margins and job creation, which in turn threatens the potential for sustainable growth. As a consequence, the loss of competitive­ness is still considered a threat to the stability of the Belgian economy over the medium term.

In recent years, the Belgian authorities have initia­ted a series of actions to address this imbalance, in particular as regards wage cost developments but also for example regarding the stimulation of innovative activities in order to align the product mix better with labour costs. However, the degree to which this will improve Belgium's competitive­ness position depends on developments in other countries. Reforms being enacted across Europe, Belgium needs to pay attention not to fall behind. This would require both more ambitious and timely action, including a comprehensive tax overhaul to rebalance taxation towards non-labour sources, reforming the wage setting system to allow for more differentiation and cyclicality, and addressing lasting mal­functions of the labour market.

3.2.      Indebtedness

In April 2013, the Commission concluded that Belgium was experiencing macroeconomic im­ba­lances with regard to its indebtedness, espe­cially concerning the implications for the real economy of the high public debt. In this section we will look at the recent evolution of the public debt level and further assess the potential risks.

Last year's In-Depth Review also concluded that the indebtedness level of the private sector, which is high in terms of non-consolidated debt, did not point to emerging risks given the still reasonable consolidated level. The high diver­gence between both levels could be explained by the high degree of intra-group loans of non-financial corporations, fostered by advantageous tax regimes, which do not constitute a macro-economic imbalance. In this section, we will reassess the validity of this conclusion.

3.2.1.   Recent evolution of the public debt

According to the European Commission's 2014 Winter Forecast, the consolidated public debt reached almost 100% of GDP at the end of 2013([29]), which was also the level at the end of 2012. This is substantially above the MIP threshold of 60% and also above the EA aggregate (at 95% of GDP in 2013). However, despite massive interventions in the financial sector and a deficit above the 3% of GDP threshold from 2009 to 2012, the debt increase since the start of the financial and economic crisis is substantially less pronounced in Belgium than in many other Member States and then in the euro area on aggregate (see Graph 3.19).

Since the increase in the debt level went hand in hand with the accumulation of assets in the financial sector, the net debt increase was more contained. It increased by 8pps. of GDP since 2007 (compared with over 20pps. in the euro area). The net debt ratio has remained broadly stable since 2011 and stands now at around the level observed in 2005.

In 2013, the general government's primary balance turned positive for the first time since the start of the crisis. However, the negative snowball effect continued, with interest expendi­ture – although historically low – exceeding nominal GDP growth (see Graph 3.20).

In order to respect its own commitment to keep the debt ratio below 100% of GDP in 2013([30]), the government sold several assets such as the investment portfolio of Royal Park Investments (0.2% of GDP), the participation in BNP Paribas Fortis (0.8% of GDP) and a claim on future rents of the Berlaymont building (0.2% of GDP). KBC reimbursed a loan of 0.5% of GDP. However, these operations have partly been offset by negative stock-flow adjustments, such as the contribution to the EFSF and ESM (0.6% of GDP).  Moreover, the sale of assets only improves the gross debt level, while their effect on the net debt is neutral.

At unchanged policy, the gross debt ratio is forecast to rise further in 2014, to 100.5% of GDP, mostly due to stock-flow adjustments (contribution to the EFSF and ESM), and to start declining in 2015. Current fiscal consolida­tion commitments, if implemented, should ensure a steady reduction of the Belgian public debt in the coming years. However, the reduction in the debt ratio remains to be hampered by low inflation, subdued GDP growth, and, in the longer term, by the impact of ageing (see section 3.2.3).

The following sections will put the high public debt level of Belgium in perspective.

3.2.2.   Risks of short-term fiscal stress

Despite the high level of its public debt, Belgium seems to have rapidly regained market confidence after the sudden increase in risk premia on Belgian government bonds towards the end of 2011. The country currently enjoys relatively low financing costs (see Graph 3.21). By refinancing its debt at low rates, the Belgian sovereign has been able to drastically reduce its interest burden, even after debt levels started to rise again.

Today, interest rates on Belgian bonds are historically low and spreads vis-à-vis German bonds converged to those of core group countries (see Graph 3.22). On top of actions by the ECB and political progress in the architecture of EMU, the regained market confidence could probably also be attributed to Belgium's good past track record in implementing large fiscal consolidation plans, notably in the '80s and '90s ([31]). Also the relatively healthy state of the private sector (see below) could have played a role. However, as a kind of swing state between core and peripheral member states, Belgium could see its risk premium rapidly rising in case of renewed financial turbulence in the euro area or domestic political instability.

Currently, Belgium does not seem to face a risk of fiscal stress in the short term. The average life to maturity of the (federal) debt portfolio ([32]) is relatively long, at 7.6 years at the end of 2013 (Belgian Debt Agency, 2014). The Belgian government used the current juncture to refinance the outstanding debt at low rates and pre-financed part of the 2014 financing needs. The 12-month and 60-month refixing risk([33]) of the federal debt decreased from 20.3% and 56.8% at the end of 2012 to 15.8% and 51.3% at the end of 2013  (Belgian Debt Agency, 2014).

Although risks of short term fiscal stress seem to be contained, a sustained period of higher interest rates on government debt would have a substantial impact on Belgium's public finances (see also Graph 3.23). Higher interest expenditure would have to be compensated by a higher tax burden or expenditure cuts elsewhere which could in turn impact economic activity. In addition, inte­rest payments to non-residents would deteriorate the balance of primary income and thus would contribute to the deterioration of the external position of the country. As discussed in previous sections, competitiveness problems are already exerting pressure on the current account balance.

Increased financing costs for the government would also pass through to the private sector, driving up financing costs for domestic financial institutions and ultimately for non-financial companies and households, which would have a negative impact on investment and innovation. Another channel to the real economy goes through the cost of mortgage loans with a floating interest rate, which is legally bound to the market interest rate of Belgian bonds. Lastly, the high public debt level also reduces the capacity of public finances to face potential adverse econo­mic shocks and thus increases the vulnerability of the economy as a whole.

3.2.3.   Long-term sustainability risks of the Belgian public debt

On top of the high public debt level, Belgium's public finances also face an above average projected impact of ageing, with ageing costs projected to rise by almost 5 pps. of GDP by 2030.([34]) At unchanged policy ([35]), this is projected to bring the debt level to almost 127% of GDP by 2030, while the euro area aggregate debt ratio is projected to decrease according to similar simulations. Moreover, these long-term projections are very sensitive to changes in the underlying macro-economic scenario. A 1 pp. increase in the interest rate assump­tions or 0.5 pp. lower GDP growth would bring the debt level to 136% of GDP (see Graph 3.23). On the other hand, adequate progress towards Belgium's Medium Term Objective (MTO)([36]), as required by the Stability and Growth Pact, would put the debt on a sustained downward path, arriving at around the 60% of GDP threshold at the end of the period.

In this context, it should be noted that Belgium has already a relatively high tax burden. Hence, the fiscal space to service a higher debt or reduce the debt burden through revenue-increasing measures is limited. However, there is room to make the tax system less growth distortive, notably through a shift in taxes away from labour (see section 4).

On the other hand, public spending in Belgium is highly decentralized. This is a particular challenge for debt reduction, which is mainly situated at federal level. In addition, also the increase in ageing costs will mainly occur at federal level. The recent state reform and revision of the special law which organizes the financing of regions and communities (see Box 3.4) tries to address this challenge, among others by indexing the amounts of financial transfers from the federal to sub-federal level only partially to GDP growth.

Lastly, the sustainability of public debt is also determined by the economy's growth potential. Section 2 highlighted the current low potential growth of the Belgian economy, especially due to a gradual erosion of the contribution of total factor productivity since the beginning of the nineties. Also the loss in competiveness and the inefficien­cies in the labour market discussed in the previous sections render the high debt level even more problematic, as they weigh on growth prospects, and in turn make it more difficult to put the ratio on a downward path.

3.2.4.   Inter-linkages with the financial sector

Belgium still carries substantial contingent liabilities due to guarantees granted to the financial sector, although they have been substantially reduced since last year's IDR. In December 2013, outstanding guarantees stood at EUR 45.4bn (12% of GDP), compared to EUR 59.6bn in 2012([37]). The maximum risk currently amounts to EUR 56bn (14% of GDP) ([38]), mainly related to the guarantee scheme for Dexia. The sale of the investment portfolio of Royal Park Investments (a special purpose vehicle created in the context of the Fortis rescue operation) and of the participation in BNP Paribas Fortis, as well as the reduction of a loan to KBC  made the sovereign less exposed to financial sector developments. Belfius, one of the four large banks on the Belgian market, is still fully state-owned.

The interaction between the sovereign debt and the financial sector also plays in the opposite direction, in particular through the substantial public debt holdings on the financial sector's balance sheet. Holdings of Belgian bonds by the domestic financial sector increased strongly between 2008 and 2012, at the expense of foreign bond holdings. While this increased concentration on the home market may have acted as a buffer du­ring the height of the sovereign debt crisis, it also reinforced the potential spill-over effects between the public debt and the financial sector in case of a decrease in the market value of government bonds.

The trend of increasing exposure to the domestic public sector has been partly reversed in 2013. Since the end of 2012, domestic exposure of credit institutions fell from EUR 69bn to EUR 62bn and stabilized at around EUR 60bn for insurance companies. However, the above-described negative feedback loops remain a source of vulnerability in case of renewed financial stress.

The two main challenges faced by the Belgian financial sector are meeting new regulatory requirements and boosting the low profitability (see Box 3.5).

3.2.5.   The public debt and the overall indebtedness of the Belgian economy

Belgium is not only characterized by a high public debt, but also by high private sector indebtedness, at least in non-consolidated terms. In 2013, the non-consolidated debt level of the private sector stood at 243% of GDP, slightly below its 2012 peak level, but well above the euro area aggregate of 160% of GDP (see Graph 3.24). This high level is mainly due to the high indebtedness of non-financial corporations (at 188% of GDP).

However, last year's IDR concluded that while the private sector debt was high in non-consolidated terms, it did not point to emerging risks given the still reasonable consolidated level, which is comparable to the euro area aggregate (see Graph 3.25) and fairly stable since 2008.([39]) For Belgium, the difference between both ratios is relatively large (around 100% of GDP), due to large intra-company lending

(Continued on the next page)

Box (continued)

of non-financial corporations, which have been fostered by advantageous tax regimes, such as the system of coordination centres which was replaced by a more general allowance for corporate equity (ACE) in 2006. While this allowance made debt financing less attractive for corporations, which may explain the drop in the consolidated debt ratio prior to the crisis, it also constitutes an incentive for intra-group loans, which are visible in the non-consolidated ratio.([40])

It should be noted that also consolidated debt figures include some intra-group loans, notably cross border lending and borrowing inside multinational groups. Loans granted by foreign firms (excluding MFI) to Belgian non-financial corporations, which accounted for almost 40% of GDP in 2012, could serve as a proxy for this intra-group lending. As shown by Bruggeman and Van Nieuwenhuyze (2013), the increase in such loans between 2005 and 2012 (+17 pps. of GDP) can explain much of the rise in the consolidated debt ratio of non-financial corporations between 2005 and 2012 (+26 pps. of GDP). During the same period, bank lending to non-financial corporations only increased by 5pps. of GDP.

Lastly, the steady increase in the liabilities of non-financial corporations has been accompa­nied by an accumulation of assets. Net liabilities of this sector stand at around 90% of GDP, which is below the euro area aggregate of 95% of GDP. Also other indicators of Belgian corporations' financial health do not point to significant sustainability risks. As shown already in last year's IDR, the rise in the debt-to-GDP ratio is not mirrored by an increase in debt-to-assets or debt-to-equity ratios of non-financial corporations.

The risks related to high public indebtedness are also somewhat tempered by the good financial position of Belgian households. Household indebtedness is lower (55.8% of GDP) than in the euro area (63.9% of GDP). However, it has continued its increase in recent years (see Graph 3.27), while in the euro area as a whole it has stabilized since 2009. Given that household debt is mostly mortgage-related, this divergent evolution could be explained by the fact that, contrary to most other euro area member states, the Belgian housing market has not seen a price correction so far. Last year's IDR argued that a possible downward correction of housing prices would only have a limited macro-economic impact (see Box 3.6 for an update on the housing market).

At the end of 2013, the National Bank of Belgium took some measures to address potential risks of housing market overvaluation: an add-on of 5pps. to banks' own risk weights for Belgian mortgage loans; a horizontal review of banks' internal risk based models for Belgian mortgage loans; a self-assessment and reporting by banks on compliance with the European Banking Authority's opinions on good practices for mortgage lending and on good practices for the treatment of mortgage owners with payment difficulties.

Fewer than half of households are indebted and three in ten Belgian households have a mortgage loan. In the lowest income quintile, around 25% of households are indebted and 38.5% of their income is dedicated to debt service, which is relatively high from a European perspective (see Graph 3.28).

The overall increase in household indebtedness has been mirrored by growing financial assets, resulting in a relatively low and stable debt-to-financial assets ratio. This ratio stands at 20.4%, compared to 31% in the euro area as a whole. Moreover, the debt to asset ratio is rather equally distributed in Belgium (see Graph 3.29).

Overall, net financial assets of households are highly positive (around 220% of GDP) and are now close to pre-crisis levels. The recent recovery is mostly related to the rise in equity prices, the volume effect having played less of a role due to the slight reorientation of the household investor's profile to less risky assets.([41]) Net financial assets of Belgian households are substantially higher than the euro area aggregate of 136% of GDP and the three main trading partners, with a ratio of 192% for the Netherlands, 139.4% for France and 126.5% for Germany. However, this favourable position of households may change in the long term, notably due to reduced savings as a consequence of population ageing.

Net financial assets are somewhat more evenly distributed than in the euro area on average. The highest income quintile owns 61.2% of net financial assets as compared to 67.7% for the euro area aggregate. Moreover, the share of households owning negative financial assets is smaller in Belgium.

However, this more even distribution of financial assets between income groups is not fully confirmed by the distribution of total net wealth. While 70% of Belgian households own their main residence as compared to 60% in the euro area (NL: 57.1%; FR: 55.3%; DE: 44.2%), the share of home owners in the lowest income quintile (45%) is around the euro area aggregate (47%) (Du Caju, 2013). On the other hand, the share of households with negative wealth is smaller in Belgium (2.7%) than in the euro area (4.8%), and the three main trading partners (NL: 11.7%; DE: 7.4%; FR: 3.9%).

In conclusion, while the public sector carries a substantial net debt, the net asset position of the Belgian economy is highly positive (+45% of GDP) and markedly above the euro area aggregate (-24% of GDP), notably thanks to the net assets of households, which more than offset the net liabilities of the public sector and the non-financial corporations (see Graph 3.30).

The total level of taxation – or public expendi­ture – reflects to a large extent collective choices regarding public facilities such as the organisa­tion of welfare provisions, education and public infrastructure. The level of taxation therefore needs to be appreciated against public output. Yet, comparable overall tax burdens may bring about quite divergent consequences for economic development as the distribution across different types of taxation influences taxpayers in dissimilar ways. In this respect, different types of taxation can be ranked from less to more 'growth-friendly' with potential distortive effects increasing with the overall tax burden. Put concisely, taxation in Belgium is characterized by a high overall level with the tax burden skewed towards labour (see Graph 4.1) and high nominal rates in combination with narrow tax bases given the presence of significant tax expenditures, in particular for VAT and personal income taxation (PIT).

The high tax burden on labour is due to high employers' social security contributions (SSC) and high personal income tax rates, including for medium income earners. Employers' SSC account for the highest share of the tax wedge. The high tax burden on labour goes hand in hand with weaknesses in the labour market discussed in section 3.1.4. It is in this context that European recommendations have called upon Belgian authorities to address the suboptimal tax design by pursuing a general tax shift away from labour.([42]) Wage costs being one of the drivers behind the loss in external competitiveness discussed higher, this tax reshuffle may help to address this imbalance and increase the low activity rate that characterizes the Belgian labour market. This would in turn support the country's growth potential, which connects with the second broad challenge of assuring public finance sustainability. Given that budgetary margins generated in upcoming years will be needed in the first place to bring down the fiscal deficit, this implies that in the short term tax reforms will need to be broadly revenue neutral and assure stable revenue flows. Hence, a tax shift is currently the most realistic way to alleviate labour taxation.

Despite the repeated call for a comprehensive review of the tax system, no substantial progress has been made so far.([43]) A consensus on the need to make more substantial inroads on labour taxation seems nevertheless materialising among political parties – though the financing side of tabled proposals appears less developed – and progress seems feasible following the May 2014 elections at federal and regional levels. This could be especially true as the devolution to the regions of certain tax competen­ces regarding personal income taxation and targeted social security contributions within the framework of the sixth reform of the state (see Box 3.4) may create a window of opportunity. This focus aims precisely to look at potential ways forward.

Redesigning the Belgian tax system

Two approaches (or a combination of both) could be considered with regard to a lowering of labour taxation. Lower taxation upstream by a reduction in employers' SSC or lower taxation downstream by reducing PIT or employees' SSC.([44]) This enhances respectively labour demand and supply which are both hampered by high taxation. Reducing employers' SSC for low-skilled jobs carries the highest potential for job creation according to Breemersch & Konings (2013), who also highlighted the need for sufficient accompanying measures (e.g. training) to overcome mismatches. QUEST simulations have comparable findings for targeted SSC reductions (EC, 2013).

Regardless of the approach chosen, compensa­ting revenues from alternative sources will have to be found in order to ensure revenue neutrality. Provided the goal is to make a serious dent in labour costs, a multipronged approach will be needed.([45]) Making the tax system more efficient implies that compensating revenues should have their origin in tax bases that are stable over the economic cycle, are broad enough and are less distortive from an economic point of view than labour taxation. Economic distortion is understood to increase from taxation on property and consumption (VAT, excise duties, environmental) over labour income to corporate income (Arnold, 2008).

Property taxation

There seems to be scope to restructure and possibly increase property taxation, mostly a regional competence in Belgium. While total property taxation is already relatively high compared to other countries (see Graph 4.2), this form of taxation, in particular recurrent taxes on real estate or the taxation of value added on land property, is considered among the taxes least detrimental to growth given the rather inelastic tax base (Johannesson Lindén & Gayer, 2012; OECD, 2010). However, it also appears ahead of least popular taxes.

On the one hand, recurrent taxes on immovable property could be increased. Aside from a stable source of revenue, their distortive effect is limited as already highlighted.  Recurrent taxes currently represent only around 40% of total property taxation. In this respect, cadastral values will have to be updated as these estimates of imputed rent (the net annual revenue generated by renting a building or plot of land) form the basis for recurrent taxes on immovable property in Belgium.  While Belgian law stipulates that imputed rents have to be re-estimated each decade, taxes are currently based on indexed rent estimations for 1975. This way the tax base is significantly underestimated: the value assessed through imputed rents is on average less than half the market value (Høj, 2009). Moreover, over the last 40 years, the relative value of property has changed drastically, even within municipalities, giving the tax a regressive character in some situations, also considering the existence of tax reductions in case of a low imputed rent.

Dated imputed rents also serve as a benchmark to determine whether a property can benefit from lower registration rights. The obligatory registration of renting contracts for private accommodation since 2007 may provide the authorities with a good database for re-assessing imputed rents. Considering that regions are the recipients of property taxes with the imputed income only playing a minor role for PIT at the federal level, transferring the responsibility over the re-assessment of imputed rents to the regional level would allow regions to determine both tax rates and their basis([46]) and to employ these for the development of more holistic housing policies.

This may involve, on the other hand, lower revenues from transaction taxes (regional registration rights in Belgium). Currently, these are the highest of all EU27 countries as seen in Graph 4.2. Lowering them further or generalizing the carry-over of paid duties as already exists in Flanders([47]) may stimulate labour mobility and would facilitate switching property according to effective needs. This may improve the proper functioning of the property market, which may over time also facilitate price corrections in certain areas and markets. Revised registration taxes would be one element in the broader revision by regions of the fiscal treatment of ownership and acquisition of real estate.

Environmental taxation

Environmental taxation is currently under­developed in Belgium.([48]) Revenues are among the lowest across the EU. This is mainly due to low revenues from energy taxation (see Graph 4.3). Cross-border fuel tourism may partly explain these low revenues, but also other elements are under­stood to play a role: lower excise duties on diesel than on petrol in combination with a higher share of diesel cars, and low taxes on heating fuels. Heating support to vulnerable households could potentially be organized more efficiently through general welfare payments, rather than through the current generalized support to energy consump­tion. In addition, Belgium does not apply an auto­matic indexation of environmental taxes.([49]) Index­ing excise duty levels to inflation would prevent an erosion of tax revenues and would help to maintain the impact of the tax on agents' behaviour.

Another important environmental issue is the preferential treatment of company cars (and fuel cards) in the tax system. Despite measures to make the tax base of company cars (which are more widespread than in other countries) CO2 dependent, the private use of company cars conti­nues to be heavily subsidised. This encourages car ownership, affects driving and commuting habits and imposes welfare costs on the society. Finally, projections for greenhouse gas emissions by 2020 indicate that Belgium will miss its 15% reduction target by 11 pps. (EC, 2013).

While this underpins the argument in favour of a fiscal greening and an increase seems feasible through, among others, higher excise duties, road pricing or kilometre-driven charges, revenue expectations need to be qualified somewhat. Indeed, environmental taxes are 'Pigovian' taxes: corrective consumption taxes which aim to internalise externalities. As these taxes target a socially desired behaviour, they may stifle an already relatively narrow tax base – ambitions of moving towards a 'low-carbon economy' would resort the same effect. As a consequence, effective revenues may disappoint in the long-run and additional sources will need to be found for tax shift purposes. Furthermore, higher taxes on energy consumption need to be reconciled with the need to safeguard competitiveness of the Belgian industry, which is rather energy-intensive in composition as underscored in section 3.1.3.

VAT

The efficiency of VAT revenues in Belgium seems subject to improvement. In 2011 the country had a VAT revenue ratio (actual revenues vs. revenues that would theoretically be raised if VAT was applied at the standard rate to all final consumption) of 47.9%, compared to a EU27 average of 49.4% and 55.6% and 55.1% in respectively Germany and the Netherlands (EC, 2013). Studies point out that this efficiency gap is not due to higher VAT evasion as compared to other countries (CASE/CPB, 2013) but reflects the fairly widespread application of VAT exemptions and reduced rates. These tax expenditures amoun­ted to 2.4% of GDP in 2012([50]) and apply to a relatively broad group of products, affecting nearly 30% of the tax base. This suggests considerable scope for efficiency gains.

The standard VAT rate is 21%, comparable to rates in neighbouring countries. In addition, two reduced rates of 6% and 12% and a zero rate exist.

- 0%: applicable to newspapers and periodicals;

- 6%: applicable to primary and socially important goods and services. This intro­duces a certain degree of progressivity in the VAT given the higher importance of such goods in the total consumption of lower income house­holds. A comparable reduced tariff for basic goods exists in most other European countries;

- 12%: applicable to e.g. restaurant meals and social housing.

Reducing the number of rates and allocating products to the most fitting category would go a long way in improving VAT collection. A first element would most likely be the reconsideration of the appropriateness of maintaining the 12% reduced rate.  Some items currently taxed at this rate (e.g. social housing) seem to qualify for the reduced VAT rate, though for most items there does not seems to be sufficient justification for not applying the standard rate.([51]) The main item in this regard is the reduced VAT for restaurant meals, introduced as recently as 2010 and carrying a budgetary cost of 0.1% of GDP.([52]) The 0% rate accorded to newspapers and periodicals represents a comparable cost for Belgian public finances. At the same time, the primary or social character of certain items currently taxed at 6% appears dubious (e.g. fresh flowers, sport and amusement tickets, original pieces of art, antiquities,…), creating scope for applying the standard rate for these products.

In addition, also an increase in the standard VAT rate could be contemplated. VAT has proven a stable revenue source in the past, amounting to 6.9-7.1% of GDP during 2004-11. An additional advantage is the broad tax base, meaning that a slight increase in the standard rate may yield considerable additional revenues with probably only modest changes in consumer behaviour. In the context of its budgetary exercise for 2013, the Belgian government already contem­plated an increase of the standard VAT rate from 21% to 22% (and a simultaneous reduction of the 6% rate), though this option was not retained in the end. Exemptions for services by bailiffs, notaries and lawyers have been lifted in recent years, while the decision to subject household electricity consumption to the reduced rate of 6% (see Box 3.2) goes against the idea of a reorientation towards consumption taxes.

When looked at in isolation, measures to increase VAT efficiency may curb its progressiveness. However, value added taxes can be considered a blunt instrument for redistribution as reduced rates do not allow targeting low-income households by discriminating between identical consumption by lower and higher income groups.([53]) Yet, from a fairness point of view it is the degree of redistribution of the overall tax system that matters. Furthermore, increasing VAT revenues in order to reduce employers' SSC for low-wages([54]) would enhance job creation in this segment and increase disposable income.

Clearly, a tax shift in this direction would have the biggest effect on external competitiveness provided the resulting price effects do not (fully) transpire in wage growth. As De Mooij & Keen (2012) highlight,  a shift towards VAT can accele­rate rebalancing but has essentially short- to medium-term effects as over a longer time horizon  the price effect would trickle through in wages. In particular within the Belgian context of automatic indexation of wages a (partial) neutralization of a temporary uptick in inflation would probably be required.

PIT

Personal income taxation is characterized by high marginal rates as applicable rates increase rapidly already at the lower end of the wage scale. This creates disincentives for the uptake of (extra) work and is reflected in the high labour market traps discussed in section 3.1.4. A wide­ning of tax brackets could be financed through a revision of the many tax expenditures that exist in personal income taxes and which eat into the tax base.

Though not all tax expenditures are unwarran­ted or inefficient, a rationalization seems never­the­less necessary as they are higher and more numerous in Belgium than in neigh­bouring countries. They have been rising fast over the past decade, reaching 2.4% of GDP in 2012. Aside from broadening tax bases, cutting tax expenditures contributes to raising transparency and fairness while reducing distortions, and lead to a favourable trade-off with growth in both the short- and long-run (Carnot, 2013) given the distortive effect they often resort through the creation of dead-weight losses. While generally favouring higher incomes, tax expenditures do not fundamentally reduce the overall progressivity of the Belgian PIT (Decoster, 2013). A comprehen­sive revision would assess for every tax expendi­ture whether it is (i) opportune in terms of policy goals, (ii) effective in moving towards these goals and (iii) justifiable in terms of budgetary impact. Such a revision seems to impose itself anyhow as several important tax reductions will become regional competences following the sixth reform of the state (see Box 3.4), all the more as budgetary means will only be partly transferred and the cost of several of these reductions has been rising fast. Examples of costly tax expenditures which are likely to give rise to distortive effects include reductions for mortgage loans (0.33% of GDP in 2012, compared to 0.11% in 2008), private pension savings (0.14% of GDP), and service vouchers (0.06% of GDP), though there are a plethora of small measures that contribute to the complexity of PIT and could possibly be reconsidered.

Another way to broaden the tax base of PIT is pushing back the 'corporatisation' of personal income by subjecting income of (quasi) indivi­dual companies to progressive PIT rates. As highlighted in Valenduc (2011) the significant difference in marginal tax rates between personal and corporate income taxation has led to the creation of many vehicles that, according to their judicial statute, are companies but in reality often not aim at expansion and generate little to no employment.

A more far-reaching reform would entail the fiscal globalisation of all revenues of private persons. This would subject all income from labour, capital, property, social benefits etc. to a progressive tax rate and has been put forward by Castanheira et al. (2014) as the initial globalisation of revenues has been gradually eroded by government measures with the result than only labour income is currently taxed progressively.

CIT

Corporate income taxes are considered among the most growth-unfriendly forms of taxation as indicated higher and form therefore no alterna­tive for labour taxation. Nevertheless, also CIT qualifies for a base broadening, which would create space to reduce the nominal corporate tax rate significantly. At 33.99% the latter is among the highest in Europe. However, a long list of tax deductions puts the effective tax rate a little over 26%, compared to 34.2% in France, 28.2% in Germany, 27.5% in the Netherlands and 24.9% in Luxemburg (ZEW, 2012).

A noteworthy tax expenditure in Belgian CIT concerns the deduction for venture capital or allowance for corporate equity (ACE). It was introduced in 2006 to stimulate the self-financing capacity of companies and the attractiveness of Belgium for multinationals after the phasing out of the coordination centre regime. While the ACE has improved solvability of non-financial companies, its (gross) budgetary impact has risen fast (from 1.1% of GDP in 2008 to 1.6% in 2012) while 30% of the deductions is done by companies with no employment (Valenduc, 2013). Therefore, the Belgian authorities have altered the system repea­tedly in recent years. There has been a gradual reduction in the applicable interest rate (with an addition for SMEs), bringing it more in line with market rates for Belgian government bonds. The possibility to carry-over unused deductions has been abolished since 2013. A minimum corporate income tax (the so-called 'fairness tax') has also been introduced for large companies that pay out dividends from profits which were untaxed as a consequence of ACE use and/or by the carry-forward of losses. Still, reserving the use of ACE for new, reinvested capital flows would render the system more sustainable. In addition, the specific anti-abuse rules could be reviewed to ensure their effectiveness.

Other taxes

Taxation for most financial income has been increased and harmonized during consecutive rounds of budgetary consolidation in recent years though a number of exemptions remain. A first one is the exoneration on saving accounts (0.12% of GDP in forgone revenues in 2012) with savings taxation at 0.2% of GDP, among the lowest in the EU. Another remaining exemption concerns realised value added on financial assets held by households. Finally, income from property renting to private persons could be taxed on the basis of effective revenues instead of cadastral values given the problems with this tax base discussed supra.([55])

Excise duties on alcohol and tobacco, sometimes referred to as 'sin taxes', have been increased repeatedly in recent years. Given the narrow tax base, the possibilities of cross-border shopping, and the change in consumer behaviour to be expected from increases, the scope for additional revenues seems more limited.

Finally, collection costs stood at EUR 1.36 per EUR 100 of taxes in 2011 (OECD, 2013). This places Belgium in the upper spectrum of the EU. Reducing these relativity high costs towards the EU27 average of EUR 1.13 may contribute to a lowering of labour taxation.

Conclusion

The high level of tax expenditure weighs on the complexity and efficiency as well as the fairness of the Belgian tax system. As a consequence, tax bases for VAT and PIT provide ample scope for widening in the context of a tax shift away from labour as to address the loss in competitive­ness. A general approach of tax base broadening carries the advantage of a more efficient allocation of resources, which in turn would benefit the economy's growth potential over time. In addition, an increase in the standard VAT rate may be considered as this would generate substantial revenues with limited spill-over effects. Other areas to be explored in a global, revenue-neutral redistribution of the tax burden include (recurrent) property taxation, the use of currently underdeve­loped aspects of environmental taxation and certain types of financial income.

A comprehensive rethinking of the Belgian tax sys­tem could be complicated by the scattered fiscal competences between the federal level and regional entities. Each level could nevertheless take substantial action within its respective compe­tences while ensuring appropriate co-ordination. In this respect the sixth reform of the state may even facilitate the realisation of a tax shift given the increase of regional surcharges in PIT and the transfer of a list of tax reductions, both of which could be employed for targeted SSC reductions, another new regional competence (see Box 3.4).

The analysis carried out in section 3 confirms that competitiveness and public indebtedness confront the Belgian economy with important macro-economic challenges. These were already identified under the MIP in preceding IDRs and relevant policy responses were reflected and integrated in the country-specific recommenda­tions (CSR) adopted by the Council in July 2013. The evaluation of progress in the implementation of those recommendations will take place in the context of the assessment of the National Reform Programme and Stability Programme under the European Semester. Against this background, this section discusses different avenues that could be envisaged to address the above-mentioned challenges.

External competitiveness

With regard to the first broad challenge of external competitiveness different avenues need to be distinguished given the composite nature of the observed loss.

In order to allow for the high export market concentration to diversify and take advantage of faster growing markets, the authorities may aim to further enhance conditions for manufac­turing companies to partake in or to expand their export activities. Scope for wider export involvement seems significant among the many Belgian SMEs. Also the big regional differences in cross-border trade suggest potential for improve­ment. Access to (long-term) financing and the inherent risk of default are generally understood to represent the main barriers for companies to cross borders. While companies often face inadequate prospects to fund their expansion by means of private capital, at the same time, record amounts stand idle on Belgian saving deposits. Policy makers may consider ways to mobilise this sleeping capital for more productive means. Assisting companies in moving beyond the small domestic market would also allow for their natural expansion, which in turn would augment the potential for job creation.

Market diversification may also result from an alteration of the product mix engendered in Belgian facilities. As sequential IDR-analysis highlighted, exports are geared towards the medium level in terms of value added. Squaring Belgium's high labour costs with sustainable job creation and the conservation of a high standard of living will require a reorientation towards products with a higher technological content than is currently the case. To nurture the innovative prowess of the Belgian economy some potential ways forward were already outlined in last year's IDR. A first one is a further enhancement of support to clusters. Given the importance of SMEs for the Belgian economy as well as their general underrepresentation in export activity, bringing together compatible SMEs within specialized clusters could compensate for a loss in traditional, mass production. Creating the necessary conditions to allow for innovative firms to expand remains another general challenge. Financial incentives put in place by Belgian authorities (e.g. payroll tax incentive and tax credit to lower R&D personnel and investment costs) are laudable initiatives to help innovative firms to expand and Belgium to move closer to the 2020 target of 3% of GDP spending on R&D. Access to subsidies is, howe­ver, still considered complex and time-consuming, while available support is highly fragmented. Also, in order for research performed in Belgium to lead to actual product development in the country, attention could be paid to labour taxation as the elevated level discourages manufacturing activities. Aside from financial initiatives, a further administrative modernisation and simplifi­cation could be pursued through for example a deeper penetration of ICT solutions considering that administrative burdens affect smaller companies relatively more than bigger ones.

Relative dynamics of productivity-adjusted wages are an important element with regard to eroding competitiveness. In this respect, wage formation is a strand for which the Belgian authorities could demonstrate more ambition in order to restore Belgium's relative cost competiti­veness and prevent a further decay of the country's manufactu­ring basis.

A first element to consider is the Law of 1996 and the wage norm defined under this Law. Unsuccessful attempts to increase the effectiveness of the law mean that the topic could be taken up again by the next government. So far, the political debate on reform options has focused on the methodology used to arrive at the norm, its legal status, the automatic correction mechanisms provided for in the current legal framework and the role of the government in the entire process. Discussions on how to establish a direct link between margins for wage growth and productivity evolutions seem not to be on the agenda, although persistent decoupling between both threatens employ­ment.

Introducing a more decentralized procedure of wage negotiation would enhance the overall resi­lience and functioning of the Belgian economy. Determining a national wage norm ignores important differences observed between sectors. If wages would be entirely determined at a lower, i.e. sectorial, level, companies would be enabled to square wage policies more aptly with their performance. Furthermore, a higher degree of sectorial differentiation would allow for labour inputs to be allocated away more gradually from declining activities towards more dynamic sectors and avoid disruptions. 

Even in case of a reform of the Law of 1996, wage indexation means that a renewed wide­ning of the wage gap cannot be excluded. The widespread practice of some form of automatic wage indexation carries a high symbolic value in the eyes of both its supporters and its detractors. Suggestions for refor­ming this character­istically Belgian feature, including repeated calls in this direction by European institutions, have stirred stiff resistance by some players and are perceived as a dismantlement of the Belgian welfare state.  However, the tangible advantage in the short term of swift purchasing power correc­tion is offset by risks of creeping employment erosion over the longer term with insufficient versatility in times of economic distress affecting competitiveness. 

Certain reforms would carry im­portant advan­tages and render the system more sustainable in terms of employment. Indexation could for example be limited to a certain wage level (i.e. indexing salaries in nominal terms compared to the current, regressive proportional adjustment). This would at the same time guaran­tee effective purchasing power and better reconcile wage devel­opments with economic realities. Other options for using the entire margin and safeguard cost competitive­ness more accurate­ly include the possibility to allow employers to use part of the increase in the index for non-wage compensation such as pension plans or to invest these means in human capital. Concluding all-in agreements remains another viable option.

Given the practice of automatic wage indexa­tion, domestic price developments are an important aspect behind nominal wage growth. The notable efforts by the Belgian authorities to enhance the functioning of product markets, in particular for energy and telecommunica­tion, seem to bear fruit as reflected in price developments. Continued monitoring in order to make these gains permanent remains warranted, especially when considering that non-commodity price components (e.g. energy distribution costs) have tended to rise fast in the past. Making sure that all network tariffs reflect actual costs and are incentive-based remains therefore crucial. Expanding attention to the functioning of service sectors may be another goal.

The current approach of repeated, small reduc­tions in parafiscal pressure on labour could be replaced by a global rethinking of the design of Belgian taxa­tion. Employment growth would be supported by a broader approach of reduced labour taxation as compared to the current emphasis on measures to reduce cost of hiring for a limited number of workers in often non-tradable activities such as the catering sector. A broad rethinking would involve a shift in the tax burden from labour to other sources of revenue, both at the federal and the regional level. While a broad consensus has developed on the need of such a reorientation, no effective action has been initiated. Further­more, the source of compensating revenues is more subject to disagreement. VAT and PIT carry important scope for tax base widening. In addition, an increase in the standard VAT rate may be considered as this would generate substantial revenues with limited spill-over effects. Other areas to be explored in a global redistribution of the tax burden include (recurrent) property taxa­tion, the use of currently underdeveloped aspects of environmental taxation (including the treatment of company cars) and the taxation of certain types of financial income. A point to be reckoned with concerns the distributional effects a rethinking of the tax system may induce and how overall pro­gressivity of the tax system would be safeguarded.

In order to boost the growth potential of the economy it will be crucial to increase the labour participation rate of underrepresented groups. Although in recent years the government has taken measures to increase incentives to work and reduce skill mismatches, the results are still inconclusive.  Persistently high levels of skill mismatches suggest that active labour market policies are not playing their role. In this regard, it is crucial to strike the right balance between, on the one hand, time-responsive benefits and appropriate job search requirements and, on the other hand, effective job search assistance and training opportunities. In addition, the tax reshuffle discussed above, could aim to lower disincentives for the uptake of (more) work, including by second-earners. The uptake of work by the latter category may also be assisted by a rethinking of the current cash-based child support towards a service-based system.

Regarding job search assistance and training, the complexity of the system and the increased capacity need for effective assistance from the public employment services (PES) will have to be addressed.  The latest reform of the state will provide regions with the leverages to approach regional labour market rigidities in a more tailor-made way, for example through job orientation, retraining and activation policies. As a conse­quence, appropriate working means for PES needs to be assured. Finally, more concrete steps could be taken to stimulate the uptake of STEM-studies and alleviate labour market tensions.

Public indebtedness

The high public debt level and the possible implications for the real economy remain one of the main challenges for Belgium. Debt reduction would not only reduce the risks associated to the public debt, but would also provide the authorities with more latitude to implement a fiscal policy aimed at improving the competitiveness of the country, as well as to face unexpected develop­ments in other economic sectors, such as financial markets. So far, strategies to reduce the public debt level were partly built around the sale of assets.

However, further deficit reduction in itself, in line with the country's commitment to reach its Medium Term Objective in 2016, will remain necessary in order to put the debt ratio on a sustainable downward path. While the deficit is estimated to have been brought below the 3% of GDP threshold of the Stability and Growth Pact([56]), a large fiscal effort (3% of GDP) remains to be made over the coming years in order to reach the agreed targets. The high overall tax burden in Belgium and the above-mentioned competitiveness challenge faced by Belgium imply limited room for deficit and debt reduction by overall tax increases. Hence, deficit reduction will have to come primarily from spending restraint.

While Belgium's state structure became gra­dually more decentralised over the last decades, the historically accumulated public debt remained largely at the federal level. The recent reform of the state devolves further spen­ding responsibilities and tax powers to regions and communities. In this context, clear and transparent fiscal arrangements will be necessary. The sixth reform of the state includes a revision of the Special Finance Act which rebalances the financial flows between the federal and sub-federal authori­ties. A recent cooperation agreement implementing the European 'Fiscal Compact' provides some mechanisms for budgetary coordination. A rigo­rous implementa­tion and a balanced contribution by all layers of government will be beneficial in order to achieve a successful consolidation strategy.

Also the implicit debt associated with an ageing population will need to be addressed by further pension reforms in order to prevent new increases of the debt level in the medium term. Measures to strengthen the growth potential of the economy would also contribute to the sustaina­bility of public finances.

Lastly, while prospective consolidation needs limit the scope for overall tax decreases, they do not prevent a tax shift that is neutral from a revenue point of view (cf. supra). A tax shift away from labour may boost employment and growth and could thus even be supportive for deficit and debt reduction. Finally, making the taxation of savings more neutral with respect to the type of financial product could diversify the channels through which the large financial assets of Belgian households are allocated to the real economy. This would have to be implemented in a gradual way, in order to minimize the disruption to financial institutions and the financial system.

Arnold, J., 'Do Tax Structures Affect Aggregate Economic Growth? Empirical Evidence from a Panel of OECD Countries', OECD Economics Department Working Papers, No. 643, OECD Publishing, 2008.

Belgian Debt Agency, 'Borrowing requirements and funding plan', 10 December 2013.

Belgian Debt Agency, 'Review 2013, 2014 Outlook', January 2014.

Belgian Government, 'Stability Programme of Belgium', April 2013.

Belgian Government, 'Verslag over de uitvoering van de begroting 2013 in antwoord op de beslissing van de Raad 2013/370', December 2013.

Bodart, V., Hindriks, J., 'Les inégalités d'inflation selon l'âge et le revenu', Regards économiques 102, IRES, 2013.

Bodart, V., Shadman, F., 'Indexation et compétitivité en Belgique', Regards économiques 107, IRES, 2013.

Bogaert, H., Robette, F., 'Effets d'un choc pétrolier sur la structure des revenues en présence du mécanisme belge d'indexation', 20ième Congrès des économistes belges de langue française, CIFoP, 2013.

Breemersch K., Konings J., 'De terugverdieneffecten van een lastenverlaging op arbeid', Vives/K.U.Leuven, 2013.

Bruggeman, A., Van Nieuwenhuyze, Ch., 'Size and dynamics of debt positions in Belgium and in the euro area', NBB, 2013.

Carnot, N., 'The composition of fiscal adjustments: some principles', ECFIN Economic Brief, European Commission, 2013.

CASE/CPB for DG TAXUD, 'Study to quantify and analyse the VAT Gap in the EU-27 Member States, Final Report', European Commission, 2013.

Castanheira, M., Colmant, B., de Callataÿ, E., de Streel, A., Pestieau, P., de Laminne, I. (red.), 'Pis­tes et éclairages économiques, Vers de nouvelles orientations structurelles pour la Belgique', 2014.

Cedefop research arena (Cedra), 'Skill mismatch - Identifying priorities for future research', Cedefop working paper No 3, 2009.

Central Economic Council, 'Note documentaire: Le revenu des chômeurs, Partie I : Comparaison des régimes de chômage en Belgique et dans les pays voisins', 2013.

Central Economic Council, 'Rapport technique du secrétariat sur les marges maximales disponibles pour l’évolution du coût salarial', 2013.

Cherenti, R., 'Les pièges à l'emploi', Fédération des CPAS, 2012.

'Council Recommendation on Belgium’s 2013 National Reform Programme and delivering a Council opinion on Belgium’s Stability Programme for 2012-2016', 2013.

Decoster A., Spiritus, K., 'Wie draagt de last van een btw-verhoging?' Flemosi, 2012.

Decoster, A., 'Vrijdenken over een belasting­hervorming', De Gids, 2013.

Du Caju, P., 'Structure et répartition du patrimoine des ménages : une analyse menée sur la base de la HFCS', Economic Review, September 2013, NBB.

Dumont, M., 'The impact of subsidies and fiscal incentives on corporate R&D expenditures in Belgium (2001-2009)', Federal Planning Bureau, Working Paper 1-13, 2013.

Duprez, C., Dresse, L., 'The Belgian economy in global value chains', Economic Review, September 2013, NBB.

ECB, 'Going Beyond Labour Costs', Compnet Policy Brief 01/2013, 2013.

European Commission, 'Assessment of the 2013 National Reform Programme and Stability Programme for Belgium', Commission Staff Working Document, 2013.

European Commission, 'Benchmarking Unemploy­ment benefit Systems', DG ECFIN, 2012.

European Commission, 'Employment and Social Developments in Europe 2013', DG EMPL, 2014.

European Commission, 'Energy Economic Developments in Europe', DG ECFIN, 2014.

European Commission, 'Fiscal Sustainability Report', DG ECFIN, 2012.

European Commission, 'In-Depth Review on Belgium', DG ECFIN, 2013.

European Commission, 'Labour market developments in Europe', DG ECFIN, 2013.

European Commission, 'Tax reforms in EU Member States 2012 - Tax policy challenges for economic growth and fiscal sustainability', 2012.

European Commission, 'Tax reforms in EU Member States 2013 - Tax policy challenges for economic growth and fiscal sustainability', 2013.

European Commission, 'Winter 2014 European Economic Forecast', DG ECFIN, 2014.

Eurostat, 'Labour Force Survey', 2012.

Geens, K., 'Uitvoering begroting 2013', Presentation 30 January 2014.

Goos, M., Hathaway, I., Konings, J., Vandeweyer, M., 'High-Technology Employment in the European Union', Vives/K.U.Leuven, 2013.

Groupe d'experts Compétitivité et Emploi, 'Analy­ses des deux propositions de réformes: diminution de la TVA sur l'électricité et réductions de charges salariales dans des zones spécifiques', Groupe d'experts 'Compétitivité et Emploi' (GECE), 2013.

Groupe d'experts Compétitivité et Emploi, 'Coût salarial, subventions salariales, productivité du travail et effort de formation des entreprises', Rapport au Gouvernement du Groupe d'experts 'Compétitivité et Emploi' (GECE), 2013.

High Finance Council, 'Het Belastingsbeleid en het leefmilieu', 2009.

Høj, J., 'How to reform the Belgian Tax System to Enhance Economic Growth', OECD Working papers No. 741, 2009.

'Inventaire 2012 des exonérations, abattements et réductions qui influencent les recettes de l'état', annex to the 2014 draft budget, Belgian Chamber of Representatives, 2013.

Johannesson Lindén, A., Gayer, C., 'Possible reforms of real estate taxation: criteria for successful policies', European Economy Occasional Papers, No. 114, European Commission, 2012.

Keen, M., de Mooij, R., 'Fiscal devaluation as a cure for Eurozone ills – could it work?', contribution to www.vox.eu, 2012.

LIME Working Group, 'Assessment of house price dynamics', 2013.

National Bank of Belgium, 'Annual Report 2012', 2013.

OECD, 'Detailed description of employment protection legislation, 2012-2013', from OECD EPL Database, update 2013.

OECD, 'International Migration Outlook 2013', 2013.

OECD, 'OECD Employment Outlook 2013', 2013.

OECD, 'Tax Policy Reform and Economic Growth', 2010.

OECD, 'Enhancing the inclusiveness of the labour market in Belgium', 2013.

Valenduc, C., 'Politique fiscale et réformes structurelles, Reflets et perspectives de la vie économique', 2011/3, Tome L, pp. 149-163, De Boeck Supérieur, 2011.

Valenduc, C., 'Réformes fiscales, soutenabilité budgétaire et croissance équitable', 20ième Congrès des économistes belges de langue française, CIFoP, 2013.

van Ewijk, C., Lukkezen, J., Rojas-Romagosa, H., 'An early-warning indicator for debt sustainability', contribution to www.vox.eu, 2013.

Vandekerckhove, S., Struyven, L., Heylen, V, 'Heeft tewerkstelling in de industrie nog toekomst? Over de arbeidsmarktdynamiek van een noodlijdende sector', Over.Werk 2/2013,  Steunpunt WSE, 2013.

ZEW, 'Effective tax levels using the Devereux/Griffith methodology', Project for the European Commission, TAXUD/2008/CC/099 2012. Situation per 01/07/2012.

Zimmer, H., 'Inadéquations sur le marché du travail', Economic Review, September 2012, NBB.

([1])  According to the National Accounts (NA) approach the current account has been close to balance in recent years. While the current account (NA) has on average been 1.6 pps. more positive than the current account (BoP) during 2002-12, it also features a long-term declining trend.

([2])  In 2000 16.9% of intermediate goods went to DE, rising to 18.7% in 2006 and 18.9% in 2012.

([3])  In 2011 12.7% of the 20-29 year old population had a tertiary graduate in science and technology, compared to the EU average of 15.2%. Moreover, Belgium is one of the only countries to have seen this ratio fall as compared to 2007, when it still reached 14%.

([4])  Based on OECD/WTO data on trade in value added. The most recent year for which data are available is 2009.

([5])  This REER measurement uses the nominal effective exchange rate versus a group of 36 countries representing 80% of Belgian exports in 2012. DE, FR & NL represent 45% of Belgian exports and are often used as a proxy for general developments.

([6])  The health index differs from the HICP in that it excludes the price evolution of alcoholic beverages, tobacco products and motor fuels.

([7])  See preceding In-Depth Reviews for a more detailed discussion. The Central Economic Council's latest Technical Report (December 2013) puts the hourly wage gap at 4.8% at the end of 2012 and projects it to shrink to 3.8% by the end of 2014.

([8])  Capital deepening, a rising capital stock per labour hour, involves a situation of expanding activity with rising productivity per worker.

([9])  For reasons of data availability, data used are gross fixed capital formation in the private sector and hours worked for the total economy. When hours worked in the private sector are used for Belgium, the trend remains unchanged.

([10]) A number of revisions were already introduced as of 01/01/2013 (see Assessment of the 2013 NRP and SP for Belgium, 2013). A second set of revisions applies since 01/01/2014 and concerns the composition of the commodity basket, the way in which the price of specific products is computed and the index calculation method as such. With regard to the latter, chain-linked weights have been introduced in national inflation calculations, allowing for future changes in consumer behaviour to be faster reflected in the health index.

([11]) Correlation between overall HICP and energy price inflation in BE (2000-13): 0.89, compared to 0.73 for the EA, 0.66 for DE, 0.72 for FR and 0.34 for NL.

([12]) The impact on the Social Security budget of the repeated reductions in employers' contributions will be compensated through higher transfers from the general budget ('alternative financing'), which still makes it unfunded in aggregate terms.

([13]) This section draws to a large extent on Energy Economic Developments in Europe, European Commission, DG ECFIN, 2014.

([14]) The RUEC estimates for 2013 are obtained as follows. Per country two preliminary estimates are produced by means of logarithmic extrapolation of the change in the periods 1995-11 and 2007-11 respectively. The reported estimate is calculated as the average of the two preliminary estimates.

([15]) From 31.4% for a low qualified 3rd country national to 82.7% for a highly qualified Belgian national (LFS, 2012, 20-64 years).

([16]) Vertical skill mismatch refers to a situation where the level of skills a worker possesses is higher or lower than is required in the job. Horizontal skill mismatch refers to a situation where workers have the appropriate qualification level but different skills than required for the job they occupy (Cedefop, 2009).

([17]) The ‘work bonus’ consists of a reduction of employees’ social security contributions and of a personal income tax credit for the lowest wages. Both elements have been reinforced in several steps in January and April 2013 and in January 2014.

([18]) The latest reform of the unemployment benefit system (cf. infra for more details) has entered into force in two phases.  The reformed 'insertion allowance' scheme applies since 01/01/2012, whereas the new rules on unemployment benefits have entered into force as of 01/11/2012. The full impact of these reforms is therefore not yet taken into account in the latest OECD data. 

([19]) OECD Tax and Benefits database.  

([20]) With the exception of a single person who is better protected in Belgium than in Germany until reaching 120% of the average salary.

([21]) EUR 474.4 cohabitant; EUR 711.6 single; EUR 948.7 with dependent child. 

([22]) More precisely, the following categories are excluded: unemployed persons with a minimum of 20 years of employment (23 years from 01/01/2014); unemployed persons with children; single unemployed persons from 55 years old and above; and unemployed persons with reduced work capacity of 33%.

([23]) This period can be extended in some cases by up to six months.

([24]) See Employment and Social Developments in Europe 2013 (2014).

([25]) OECD Tax and Benefits database.

([26]) The Beveridge curve captures the relationship between aggregate unemployment and vacancies.

([27]) According to the Eurostat LMP database, Belgium is spending 3.7% of GDP on all LMP categories, surpassed only by Denmark and Spain; from which 1.38% of GDP on active LMP measures (categories 2-7: activation measures for the unemployed and other target groups including the categories of training, job rotation and job sharing, employment incentives, supported employment and rehabilitation, direct job creation, and start-up incentives).

([28]) On top of the already enacted reinforcements of the work bonus, future increases have been decided upon in the framework of so-called Competitiveness Pact. More specifically, the federal government decided to reinforce the tax dimension of the work bonus by three times EUR 50mn in 2015, 2017 and 2019.

([29]) This estimate is broadly similar to the most recent official estimate from the national authorities (Geens, 2014).

([30]) 2013 Stability Programme of Belgium

([31]) This is also shown in a CPB paper by Van Ewijk et al (2013), based on an 'at-risk' indicator which takes into account the past policy responses. According to this paper, this 'at-risk' indicator for Belgium is one of the lowest of the sample.

([32]) The federal debt represents 90% of the general government debt.

([33]) The share of outstanding debt which matures in a given time period or which is subject to changes in interest rates (because of a floating interest rate).

([34]) DG ECFIN internal projections.

([35]) These projections start from the European Commission 2013 Autumn Forecast, with the no-policy change assumption translated into a structural primary balance kept constant (excl. ageing costs) at the last forecast year (2015). The baseline scenario is based on the following macro-economic assumptions for the long term: potential GDP growth is assumed to rise to 1.6%. Inflation and change of GDP deflator are assumed to stabilize at 2% in the medium term. Long-term interest rates are assumed to rise to 5.1% and short-term rates to 4.2% (See also European Commission, 2012).

([36]) The MTO of Belgium is set at a surplus of the government balance in structural terms of 0.75% of GDP.

([37]) Source: December 2013 EDP reporting by Belgium and 2013 Stability Programme of Belgium.

([38]) i.e. the combined ceiling of all guarantees granted to financial institutions.

([39]) A statistical break (implemented in 2013) explains part of the jump in 2008.

([40]) For a detailed explanation, see 2013 IDR on Belgium.

([41]) Based on data from the National Bank of Belgium (Belgostat).

([42]) "Establish concrete and time-specific proposals for shifting taxes from labour to less growth-distortive tax bases notably by exploring the potential of environmental taxes for example on diesel, heating fuels and the taxation of the private use of company cars." Council Recommendation on Belgium's 2013 National Reform Programme, 2013.

([43]) A parliamentary commission has reflected on tax reforms and presented its report in February 2014. It takes a broad view on taxation but does not put forward specific recom­mendations on how a future government may proceed.

([44]) This is the approach taken by the Belgian 'work bonus' for low wages.

([45]) The total wage cost in the private sector is about EUR 200bn: a reduction by 1% requires in other words EUR 2bn (0.5% of GDP) in compensating revenues.

([46]) Regions set the basic rate of recurrent property taxation, while provinces and municipalities add surcharges to it. Since the devolution of property tax powers to the regional level in 2002, regions can decide to use another property tax base. They lack, however, the authority to re-estimate the current tax base.

([47]) This would still be suboptimal as transferability would probably only apply within individual regions.

([48]) The High Finance Council (2009) analysed environmental tax aspects, including policy advice.

([49]) Moreover, energy products have been excluded from the 8% increase in excise duties as of 1 August 2013.

([50]) See Inventaire 2012 des exonérations, abattements et réductions qui influencent les recettes de l'état, annex to the 2014 draft budget, Belgian Chamber of Representatives, 2013.

([51]) Aside from the reduced rate of 12%, there is also a so-called 'parking rate' of 12%, applicable to products such as cokes, charcoal and lignite. While forgone revenues are limited, the appropriateness of this rate can be questioned from an efficiency point of view, the more given the polluting effect of these products.

([52]) The ineffectiveness of a reduced VAT rate for restaurant services has already been demonstrated for other countries.

([53]) According to estimations by Decoster & Spiritus (2012), an increase in the standard rate from 21% to 22%, even without a lower reduced rate, is mostly borne by higher (expenditures) deciles.

([54]) This shift from SSC to VAT has also a coordination advan­tage in the sense that, contrary to for example environmen­tal taxation, full control over VAT is situated at the federal level, which after the sixth reform of the state remains responsible for implemen­ting structural SSC reductions such as for low wages.

([55]) On a separate but related note, it should be highlighted that the fact that the method for determining the level of property income from abroad for tax purposes (i.e. on the basis of effective rent) is less favourable than that applied for equivalent income deriving in Belgium raises a problem in terms of compliance with fundamental European freedoms. As a consequence, the Commission has been asking Belgium to revise its taxation of property income, in whichever way it sees fit, to put an end to this situation. European Commission - IP/12/282, 22/03/2012.

([56]) European Commission 2014 Winter Forecast.