EUROPEAN COMMISSION

Brussels, 22.5.2017

COM(2017) 531 final

REPORT FROM THE COMMISSION

Belgium

Report prepared in accordance with Article 126(3) of the Treaty

REPORT FROM THE COMMISSION

Belgium

Report prepared in accordance with Article 126(3) of the Treaty on the Functioning of the European Union

1.    Introduction

Article 126 of the Treaty on the Functioning of the European Union (TFEU or the Treaty) lays down the excessive deficit procedure (EDP). That procedure is further specified in Regulation (EC) No 1467/97 on speeding up and clarifying the implementation of the excessive deficit procedure 1 , which is part of the Stability and Growth Pact (SGP). Specific provisions for euro area Member States under EDP are laid down in Regulation (EU) No 473/2013 2 .

According to Article 126(2) TFEU, the Commission has to monitor compliance with budgetary discipline on the basis of two criteria, namely: (a) whether the ratio of the planned or actual government deficit to gross domestic product (GDP) exceeds the reference value of 3 %; and (b) whether the ratio of government debt to GDP exceeds the reference value of 60 %, unless it is sufficiently diminishing and approaching the reference value at a satisfactory pace.

Article 126(3) TFEU provides that, if a Member State does not fulfil the requirements under one or both of those criteria, the Commission has to prepare a report. This report must also “take into account whether the government deficit exceeds government investment expenditure and take into account all other relevant factors, including the medium-term economic and budgetary position of the Member State”.

This report, which represents the first step in the EDP, analyses Belgium's compliance with the deficit and debt criterion of the Treaty, with due regard to the economic background and other relevant factors.

On 18 May 2016, the Commission issued a previous report under Article 126(3) TFEU, as Belgium did not make sufficient progress towards compliance with the debt reduction benchmark in 2015. The report concluded that, after the assessment of all relevant factors that might justify the prima facie lack of compliance, the debt criterion as defined in the Treaty and in Regulation (EC) No 1467/1997 should be considered as complied with at the time. Those relevant factors included notably: (i) the unfavourable economic conditions which made the respect of the transitional debt rule particularly demanding; (ii) the expectation that compliance with the required adjustment towards the medium-term budgetary objective (MTO) was broadly ensured; and (iii) the expected implementation of ambitious growth-enhancing structural reforms in line with the authorities' commitment, which was expected to contribute to debt reduction in the medium to long term.

Data notified by the Belgian authorities on 31 March 2017 3 and subsequently validated by Eurostat 4 show that the general government deficit in Belgium reached 2.6% of GDP in 2016, while debt stood at 105.9% of GDP, above the 60% of GDP reference value. For 2017, the notification planned a deficit of 1.7% of GDP and a debt ratio of 105.6% of GDP, while the 2017 Stability Programme, received by the Commission on 28 April 2017, plans a deficit of 1.6% of GDP and a debt ratio of 105.2% of GDP. Eurostat has made a reservation on the quality of the data reported by Belgium in relation to the sector classification of hospitals. Eurostat considers that, under ESA 2010, government-controlled hospitals in Belgium should be classified inside the government sector, which is currently not the case. A possible future reclassification would entail a limited increase in government debt of around 0.3 pp. of GDP.

The notified data shows that Belgium made insufficient progress towards compliance with the debt reduction benchmark in 2016 (see Table 1). The change in the structural balance is estimated to have been 0.1% of GDP in 2016, compared to a required minimal linear structural adjustment (MLSA) of 2.2% of GDP 5 . Moreover, in 2017 and 2018, Belgium is forecast not to comply with the debt reduction benchmark as its debt-to-GDP ratio is expected to remain 2.7 pps. and 2.1 pps. of GDP above the debt benchmark according to the Commission 2017 spring forecast. On the basis of the scenario included in the 2017 Stability Programme the gap would be 0.8 pp. of GDP in 2017 and just 0.1 pp. in 2018, with the Stability Programme planning compliance with the debt criterion as of 2019. The difference with the Commission forecast is due to a deficit reduction that is 0.3 pp. higher in 2017 and 1 pp. higher in 2018 as the Commission forecast works on the basis of a no-policy change assumption whereas the Stability Programme includes the planned effort. That difference highlights the importance of deficit reduction for compliance with the debt criterion.

Overall, Belgium's insufficient progress towards compliance with the debt reduction benchmark in 2016 provides evidence of a prima facie existence of an excessive deficit for the purposes of the Stability and Growth Pact before, however, considering all factors as set out below.

The Commission has therefore prepared this report to comprehensively assess the departure from the debt reduction benchmark and the excess over the Treaty reference value in order to examine whether the launch of an excessive deficit procedure is warranted after all relevant factors have been considered. Section 2 of the report examines the deficit criterion. Section 3 examines the debt criterion. Section 4 deals with public investment and other relevant factors, including the assessment of compliance with the required adjustment path towards the MTO. The report takes into account the Commission 2017 spring forecast, released on 11 May 2017, and the Commission's evaluation of subsequent macroeconomic and fiscal developments.

Table 1. General government deficit and debt (% of GDP)

2.Deficit criterion

Belgium’s general government deficit widened from 2.5% of GDP in 2015 to 2.6% in 2016. According to the Commission 2017 spring forecast, the deficit would narrow to 1.9% in 2017, hence respecting the 3% of GDP Treaty reference value. In 2018, the Commission forecast projects the deficit to rise again to 2.0% of GDP under a no-policy-change assumption.

The multiannual trajectory included in the 2017 Stability Programme puts forward a deficit reduction to 1.6% in 2017 and to 0.7% in 2018. The difference in 2017 between the Commission forecast and the Stability Programme stems from a number of measures that have not been included in full in the Commission forecast, e.g. fiscal regularisation, the activation of long-term ill or anti-fraud measures. For 2018 the difference is caused by the fact that the Commission projections only include measures that have been sufficiently detailed.

Belgium thus complies with the deficit criterion as defined in the Treaty and in Regulation (EC) No 1467/97.

3.Debt criterion

General government gross debt increased from 87% of GDP in 2007 to 106.7% of GDP at the end of 2014, an increase of almost 20 pps. of GDP. In 2015 it fell back to 106% thanks to a downward stock-flow adjustment linked to the repayment of support granted to a financial institution. Debt broadly stabilised in 2016 as a small primary surplus compensated for an upward stock-flow adjustment with the snowball effect (the interest-growth rate differential) about neutral. Debt dynamics are discussed in more detail in Section 4.3.

The Commission forecast expects a mild debt reduction in coming years, to 105.6% of GDP in 2017 and 105.1% in 2018. The annual downward impact of 1.5 pps. of GDP rendered by primary surpluses and the snowball effect are projected to be partially offset by upward stock-flow adjustments in 2017-2018. The Commission forecast was finalised on 25 April 2017, before the sale of part of the Belgian State's participation in BNP Paribas on 3 May 2017. That divestment represents around 0.5% of GDP. When adjusting the Commission forecast for this downward stock-flow adjustment (assuming proceeds are fully used for debt reduction), general government debt would be 105.2% of GDP in 2017 and 104.6% in 2018. The gap towards the debt reduction benchmark would narrow slightly in 2017 (from 2.7 pps. to 2.6 pps.) and remain unchanged in 2018 at 2.1 pps.

According to Belgium's 2017 Stability Programme the debt ratio would decline to 105.2% at the end of 2017 and to 103.4% in 2018. The difference from the Commission's projection at unchanged policy stems from a lower planned headline deficit and smaller stock-flow adjustments with nominal growth assumptions broadly similar. In addition, the scenario included in the Stability Programme does not include the partial divestment from BNP Paribas. Adjusting for that element (under the assumption proceeds are fully used for debt reduction) would bring general government debt to 104.7% of GDP in 2017 and to 102.9% in 2018. The gap towards the debt reduction benchmark would slightly narrow to 0.7 pp. in 2017 (from 0.8 pp.) and remain unchanged in 2018 at 0.1 pp.

Table 2: Debt dynamics

Following the abrogation of the excessive deficit procedure in June 2014, Belgium was subject to a three-year transition period to comply with the debt reduction benchmark. That transition period started in 2014 and ended in 2016. In order to ensure continuous and effective progress towards compliance during the transition period, Member States should respect simultaneously the two conditions below:

a.    First, the annual structural adjustment should not deviate by more than ¼% of GDP from the MLSA ensuring that the debt reduction benchmark is met by the end of the transition period;

b.    Second, at any time during the transition period, the remaining annual structural adjustment should not exceed ¾% of GDP (unless the first condition implies an annual effort above ¾% of GDP, which is the case for Belgium).

Based on the Commission 2017 spring forecast, a structural effort of 1.4% of GDP would have been required in 2015 to make sufficient progress towards meeting the debt reduction benchmark, which was significantly above the adjustment towards the MTO of at least 0.6% of GDP recommended by the Council in July 2015. The structural balance is estimated to have improved by 0.6% of GDP in 2015, so that the first condition is not respected.

In view of the shortfall in 2015, the remaining required adjustment in the last year of the transition period, 2016, reached 2.2% of GDP. The structural balance is estimated to have improved by 0.1% of GDP in 2016. As a result, Belgium did not comply with the debt reduction benchmark by the end of the transition period.

The analysis thus suggests that prima facie the debt criterion for the purpose of the Treaty and Regulation (EC) No 1467/1997 is not fulfilled based on the 2016 outturn data and the Commission 2017 spring forecast as well as the 2017 Stability Programme before, however, consideration is given to all relevant factors set out below.

4.    Relevant factors

Article 126(3) TFEU provides that the Commission report “shall also take into account whether the government deficit exceeds government investment expenditure and take into account all other relevant factors, including the medium-term economic and budgetary position of the Member State”. Those factors are further clarified in Article 2(3) of Regulation (EC) No 1467/97, which also provides that “any other factors which, in the opinion of the Member State concerned, are relevant in order to comprehensively assess compliance with the deficit and debt criteria and which the Member State has put forward to the Council and to the Commission” need to be given due consideration.

In case of apparent breach of the debt criterion, the analysis of the relevant factors is particularly warranted given that debt dynamics are to a larger extent influenced by factors outside the control of the government than is the case for the deficit. This is recognised in Article 2(4) of Regulation (EC) No 1467/97, which provides that the relevant factors shall be taken into account when assessing compliance on the basis of the debt criterion irrespective of the size of the breach. In that respect, at least the following three main aspects need to be considered (and have been considered in the past) when assessing compliance with the debt criterion given their impact on the debt dynamics and sustainability:

1.    adherence to the MTO or the adjustment path towards it, which, is supposed to ensure sustainability or rapid progress towards sustainability under normal macroeconomic circumstances. As by construction the country-specific MTOs take into account the debt level and implicit liabilities, compliance with the MTO or the adjustment path towards it should ensure convergence of the debt ratios towards prudent levels at least in the medium term;

2.    structural reforms, already implemented or detailed in a structural reform plan, which are expected to enhance sustainability in the medium term through their impact on growth, thereby contributing to bring the debt-to-GDP ratio on a satisfactory downward path. Overall, adherence to the MTO (or the adjustment path towards it) alongside with the implementation of structural reforms (in the context of the European Semester) is expected under normal economic conditions to bring debt dynamics on a sustainable path through the combined impact on the debt level itself (through the achievement of a sound budgetary position at the MTO) and on economic growth (through the reforms).

3.unfavourable macroeconomic conditions, and in particular low inflation, which can hamper the reduction of the debt-to-GDP ratio and make compliance with the SGP provisions particularly demanding. A low-inflation environment requires a Member State to achieve more demanding structural adjustments to comply with the MLSA during the transition period and negative inflation surprises may contribute to the upward revisions of the required MLSA over time. In addition, the debt reduction benchmark assumes by construction that GDP deflator growth returns to the long-term average value of 2% by 2021, which makes compliance with the forward-looking debt benchmark particularly demanding. Under such conditions, adherence to the MTO or the adjustment path towards it is a key relevant factor in assessing compliance with the debt criterion.

In view of those provisions, the following subsections consider in turn (1) the medium-term economic position; (2) the medium-term budgetary position, including an assessment of compliance with the required adjustment towards the MTO and the development of public investment; (3) the developments in the medium-term government debt position, its dynamics and sustainability; (4) other factors considered relevant by the Commission; and (5) other factors put forward by the Member State.

4.1.    Medium-term economic position

Cyclical conditions, potential growth and inflation

The Belgian economy proved to be rather resilient following the global economic recession in 2009. GDP quickly regained pre-crisis levels, thanks to strong economic growth in 2010 and 2011. That recovery period was followed by stagnation, though, with flat GDP growth in 2012 and 2013. In 2014 and 2015, economic activity rebounded and growth reached 1.7% and 1.5% respectively. It fell back to 1.2% in 2016 with the terrorist attacks of March 2016 as well as the security situation in the months before and after those events considered to have had a negative, though transitory, impact on the Belgian economy. According to the Commission 2017 spring forecast, economic growth is expected to pick up again, reaching 1.5% in 2017 and 1.7% in 2018 on the back of domestic demand, more specifically private consumption and investment.

Potential growth estimates for Belgium are rather low, at 1.2% on average over 2014-2018. The slowdown compared to the pre-2009 situation is broad-based as it reflects the continuation of a long-term trend of declining gains in total factor productivity (which is estimated to have stabilised at a low level in recent years), a decline in the contribution of labour to potential growth (due to a slower growth of the working age population) and somewhat lower capital accumulation. The negative output gap narrowed to -0.3% in 2015 compared to a trough of -1.6% in 2013. It widened again slightly in 2016 as growth fell short of potential but is expected to narrow in 2017 and 2018, when the output gap would be closed.

As was the case for other euro area Member States, Belgium experienced a protracted period of low domestic price growth. The GDP deflator rose by 0.7% in 2014 and by 0.9% in 2015 compared to the historical average of around 2%. It accelerated to 1.6% in 2016. The relatively low GDP deflator until 2015 has had an important impact on the evolution of the debt-to-GDP ratio in past years and increased the structural adjustment required to assure that the debt ratio stays on a firm downward path as required by the forward-looking debt benchmark 6 . As a result, the MLSA for Belgium (2.2% of GDP in 2016, see Table 1) was substantially higher than the fiscal effort of 0.6% of GDP recommended by the Council in July 2016. In view of the above-mentioned cyclical circumstances, such a high effort may be neither feasible nor desirable. Moreover, the primary balance was also impacted by those cyclical conditions, which fed through in public debt. Economic conditions thus partly explain non-compliance with the MLSA in 2016.

However, the ongoing improvement in macroeconomic conditions means that they can no longer be regarded as a major mitigating factor in explaining the gap compared to the forward-looking debt benchmark (2.7% in 2017). The GDP deflator is expected to reach 1.8% in 2017 as price growth moves closer to the long-term average, with nominal growth rising to 3.4% both in 2017 and 2018.

Table 3: Macroeconomic and budgetary developmentsa

Declining interest rates have created a supportive context for budgetary consolidation. The implicit nominal interest rate on Belgian public debt has fallen continuously over the past two decades and that trend has accelerated in recent years. As a consequence, total interest expenditure by the general government has continued to decrease as a share of GDP. Between 2007 and 2016 interest expenditures fell by approximately 1.1 pps. of GDP. These windfall gains amounted to 0.2 pp. in 2015 as well as in 2016, with similar gains expected in 2017-2018. Against that background of falling interest expenditure, the change in the structural balance in 2016-2017 according to the Commission forecast (+0.1% and +0.6% respectively) is accompanied by a deterioration in 2016 in the structural primary balance (-0.1%) and a lower improvement in 2017 (0.3%). The sensitivity analysis in the 2017 Stability Programme highlights how a linear increase of the yield curve by 100bp would imply 0.1% of GDP higher costs in 2017, rising to 0.4% of GDP in 2020 7 , though relative to a baseline of falling interest payments. It underscores the risks inherent to a consolidation strategy that leans significantly on windfall gains stemming from lower interest expenditures.

Structural reforms

In its Communication of 13 January 2015, the Commission strengthened the link between effective implementation of structural reforms, investment, and fiscal responsibility in support of jobs and growth, within the existing rules of the SGP.

The 2017 Country Report for Belgium concluded that the country had made some progress in addressing the 2016 country-specific recommendations. Substantial progress was made with regard to the reform of the legal framework on wage negotiations, the so-called Law of 1996 on the promotion of employment and the safeguarding of competitiveness. The reform approved by Parliament in March 2017 strengthens the preventive aspect of the law in order to avoid that in the future the growth of hourly labour costs would again outpace that in the three main trading partners. To that end, social partners need to respect a safety margin when setting the ceiling for real wage increases, to anticipate potential higher domestic inflation or lower wage growth in the reference countries. Moreover, the altered Law requires the progressive reduction of the hourly wage gap built up prior to 1996. The new law also prevents that reductions of social security contributions are fully translated into wage increases. If new gaps were nevertheless to occur, enhanced correction mechanisms should ensure an automatic narrowing. Those reforms implemented an important element of the country-specific recommendations and allow for a steady continuation of competitiveness gains achieved over the past years, which in 2016 led to the conclusion that Belgium was no longer experiencing macroeconomic imbalances.

The Country Report also saw some progress regarding the functioning of the labour market. Robust job creation, in particular in the private sector, has been fuelled by labour tax cuts and wage moderation efforts. Progress with regard to other recommendations since July 2016 has been less tangible. However, over the course of the last few years important progress has nevertheless been made on a number of fronts. These relate primarily to pensions and taxation. As has been discussed in previous Article 126(3) reports, Belgium has been reforming its public pension system in recent years. In order to raise the effective retirement age, early exit possibilities have been reduced by further tightening the standard eligibility requirements for both early and pre-retirement. The legal retirement age will also rise from 65 to 66 in 2025 and to 67 in 2030. The long-term impact of that set of measures is discernible in the latest projections of the Ageing Working Group: pension expenditures are projected to rise by 1.3 pps. of GDP between 2013 and 2060, compared to 3.3 pps. before the most recent reforms. The difference is mostly due to the pension reform itself (-1.6 pps. of GDP), while other measures, such as the temporary suspension of indexation and lower public employment (-0.4 pp. of GDP), will also curb expenditure growth. Those more positive ageing projections allowed Belgium to lower its MTO under the SGP from a structural surplus of 0.75% of GDP to a balanced budget in structural terms.

Although already enacted reforms have thus substantially reduced the projected rise in public pension spending, curbing the expected increase in age-related spending further through additional reforms would improve fiscal sustainability in the long term 8 . Moreover, there are signs of a strong shift from early exit through the pension and unemployment systems to the sickness and disability schemes, which would partially offset the projected gains from the enacted reforms. Spending on sickness and disability has been rising rapidly: from a broadly stable level of around 1.2% until 2007 to 1.9% in 2016. By the end of its term in 2019, the government intends to lay the groundwork for the introduction of a credit-based public pension system as of 2030. Once fully implemented such a system would allow for automatic adjustment mechanisms in response to demographic or economic developments.

Within the framework of a multi-annual tax reform initiated in 2016, the tax pressure on labour is being reduced. Both employers and employees benefit from measures to narrow the tax wedge. On the one hand, the statutory rate of employers’ social security contributions will be gradually reduced to 25% by 2018, with lower effective rates for low to medium wages. In addition, wage subsidies for shifted and night labour have been increased. SMEs and independents are exempted from social security contributions for the first newly hired employee, while contributions are reduced for the next five employees. On the other hand, the take-home pay of employees is being increased through a combination of increases in the lump sum allowance for professional expenses, the work bonus, the tax exempted amount, and a reshuffling of tax brackets. The announced labour tax cuts represent EUR 11.5 billion by 2020, or 2.2% of GDP. While the reduction of the tax pressure on labour goes in the direction recommended repeatedly by the Council in the past, the tax reform package has not been designed in a budgetary neutral way (even when accounting for the positive effects on growth and employment). That factor contributed to the budgetary slippage in 2016 discussed below and weighs on the budget outlook for future years as well. In addition, despite the recent reforms, the Belgian tax system remains hampered by widespread distortions which narrow tax bases and contribute to the system's complexity. Taxation also continues to lean heavily on labour taxation, even after the recent reforms, and additional tax reductions at the lower end of the pay scale would contribute to reducing unemployment and low wage traps for second earners, singles and single parents.

The 2017 Country Report listed a number of other structural bottlenecks of the Belgian economy and elements which hamper economic and budgetary policy making. Those challenges were reflected in the country-specific recommendations issued by the Commission on 16 May 2017. Aside from the obligations stemming from the Stability and Growth Pact, those recommendations call for more effective fiscal coordination between the several Belgian government levels; further tax reforms to reduce complexity and distortive features of the system; better targeting of labour market and education policies on the most disadvantaged groups; improved functioning of certain service markets; fostering both tangible and intangible investment, inter alia by reorienting overall government spending in favour of investment.

4.2.    Medium-term budgetary position

Headline, structural balance and adjustment towards the MTO

Belgium's headline deficit increased from 2.5% of GDP in 2015 to 2.6% in 2016. The revenue-to-GDP and the expenditure-to-GDP ratios fell by 0.7 pp. and 0.5 pp. respectively. The 2016 headline deficit was impacted by additional expenditure related to asylum-seekers and security measures.

In the 2016 stability programme, the Belgian authorities revised their MTO to a balanced budget in structural terms, down from a structural surplus of 0.75% of GDP. The MTO appears sufficiently stringent under what can be considered as normal economic conditions to ensure debt rule compliance in the medium and long term. In the 2017 stability programme, Belgium confirmed the MTO but postponed the planned achievement of it from 2018 to 2019. According to the Commission forecast, it would require a structural improvement of 1.6% of GDP over 2018-2019. At the same time, the Commission forecast expects the structural balance to deteriorate by 0.3 pp. of GDP at unchanged policy in 2018, the last year of the Commission projections, while according to the High Council of Finance the implied deterioration in 2019 represents 0.4% of GDP 9 . As a result, achieving the MTO in 2019 will require substantial additional measures and a strict execution of the budget in view of implementation risks towards the end of the current legislative period.

The 2017 Stability Programme indicates that the budgetary impact of the exceptional inflow of refugees and security-related measures has been significant, and provides adequate evidence of the scope and nature of those additional budgetary costs. According to the Commission, the eligible additional expenditure linked to the exceptional inflow of refugees and security measures amounted to 0.08% and 0.05% of GDP, respectively, in 2016. The provisions defined in Article 5(1) and Article 6(3) of Regulation (EC) No 1466/97 allow that additional expenditure to be catered for, in that the inflow of refugees as well as the severity of the terrorist threat are exceptional events, their impact on Belgium's public finances is significant and sustainability would not be compromised by allowing for a temporary deviation from the adjustment path towards the MTO. The required structural improvement in 2016 has consequently been reduced from 0.6% of GDP to 0.47% of GDP. In 2017, the additional security-related expenditure is currently estimated at 0.01% of GDP. A final assessment, including on the eligible amounts, will be made in spring 2018 on the basis of observed data as provided by the Belgian authorities for 2017.

Based on outturn data and the Commission forecast, the growth of primary government expenditure, net of discretionary revenue measures and one-offs, exceeded the expenditure benchmark – corrected for the impact of unusual events – by 0.6% of GDP in 2016, pointing to a significant deviation. The structural balance is estimated to have improved by 0.1% of GDP in 2016, 0.4% of GDP below the recommended effort of 0.47% of GDP, suggesting some deviation. The expenditure aggregate was negatively impacted by higher than expected inflation in 2016. Whereas the reference growth rate for the expenditure benchmark is based on a deflator of 1%, the actual GDP deflator used for the structural balance amounted to 1.6% of GDP, closer to the long-term average. The impact on expenditure growth from higher than anticipated inflation transpires in the fact that the automatic indexation of social benefits and public sector wages occurred four months earlier than expected in the Commission 2015 autumn forecast at the time of the draft budget for 2016. The impact is estimated at around 0.2% of GDP. Correcting for this brings the deviation for the expenditure benchmark at 0.4% of GDP, with both pillars providing the same signal and leading to a conclusion of some deviation from the recommended adjustment path towards the MTO in 2016. In 2015-2016 together both pillars suggest a significant deviation took place. However, when correcting the expenditure benchmark for the above-mentioned higher than anticipated inflation in 2016 and the structural balance for revenue shortfalls in 2015 compared to standard elasticities – due to low wage and price growth – (discussed in the spring 2016 assessment), both pillars point to some deviation in 2015-2016.

In 2017, the growth rate of the expenditure aggregate is projected to exceed the expenditure benchmark by 0.4% of GDP, pointing to a risk of some deviation. At 0.6% of GDP the improvement in the structural balance is in line with the recommended structural adjustment. Over 2016 and 2017 taken together, however, the expenditure benchmark points to a risk of significant deviation, with an average deviation of -0.5% of GDP. The average deviation for the structural balance over the same period amounts to -0.2% of GDP according to the Commission forecast, indicating a risk of some deviation. When correcting for the impact of unforeseen inflation in 2016 discussed supra, the deviation for the expenditure benchmark in 2016-2017 narrows to 0.4% of GDP, still above the threshold for significant deviation. The remaining difference with the average gap for the structural balance reflects the impact of the decline in interest expenditure in both years. That windfall improves the reading of the fiscal effort based on the structural balance but does not affect compliance with the expenditure benchmark, which is therefore considered to reflect more appropriately the underlying fiscal effort. As a result, the overall assessment points to a risk of a significant deviation from the recommended adjustment path towards the MTO over 2016 and 2017 taken together, but the projected deviation can still be corrected in 2017.

Public investment

Public investment peaked at 2.5% of GDP in 2012, due to the investment cycle at local level. For the same reason, it decreased again to 2.4% of GDP in subsequent years and to 2.3% in 2016. Investment grants to non-financial corporations (among others for railway infrastructure, hospitals, and elderly homes) are not included in those figures. Over the forecast horizon, investment is projected to return to 2.5% of GDP due to large investment projects at regional level and an acceleration of local government investment in the run-up to municipal elections in 2018. Since 2009, public investment has been lower than the general government deficit, though the latter is expected to fall below the investment ratio again as of 2017.

The federal government has been pursuing a 'National Investment Pact', which aims to accelerate investment in key areas by mobilising public and private means as well as by identifying obstacles for private investment. Regions and Communities can join the initiative.

4.3.    Medium-term government debt position

Debt dynamics

Between 1997 and 2007, Belgium's government debt-to-GDP ratio decreased by 36 pps., thanks to sizeable (although gradually declining) primary surpluses. That trend of sustained debt reduction was halted by the financial and economic crisis of 2008. At the end of 2007, Belgium's general government debt stood at 87% of GDP. It rose to 106.7% of GDP in 2014, an increase of almost 20 pps. It compares to an increase of 27 pps. in the euro area.

Graph 1. Breakdown of change in debt ratio 1997-2018 (pp. of GDP)

The main drivers behind the increase between 2007 and 2014 were the upward snowball effect (+10.5 pps.) and stock-flow adjustments (+9.4 pps.), with the erstwhile primary surpluses gone (see Graph 1). The snowball effect reflects how interest spending generally surpassed nominal growth since 2008. Yet, at 1.4 pps. on average, the annual upward impact of that dynamic in 2008-2015 was similar to that in 1997-2007, as the denominator effect of lower nominal growth was offset by the nominator effect stemming from lower interest spending in terms of GDP. The latter ratio continued to decline after 2007 as continuously declining interest rates compensated for an increasing debt ratio. A further decline in interest spending in recent years resulted in a slightly downward snowball effect in 2016 for the first time since 2011.

The substantial debt increase due to stock-flow adjustments occurred predominantly in 2008 and 2011, when the Belgian State had to intervene in the financial system. In 2008 authorities had to step in to save Fortis, KBC, Dexia and Ethias. In 2011 the Belgian State acquired Dexia Belgium, the current Belfius bank. The recovering of part of the financial sector bailout resulted in downward stock-flow adjustments representing 3.4% of GDP in 2012-2015. There was no change in 2016. Remaining participations include a share of 7.8% in BNP Paribas – 10.3% before the partial divestment in May 2017 –, 100% of Belfius, 100% of insurer Ethias (including stakes of regional and local authorities), and 50% of Dexia bank. Dividends paid by BNP Paribas and Belfius represented 0.1% of GDP in 2016.

The positive stock-flow adjustment of 0.2% of GDP in 2016 was mainly the result of interest rate swaps, the difference between accrued and paid interest and regional loans for social housing.

While the accumulated impact on the debt ratio from the primary balance has been about zero since 2008, the contrast with the situation prior to 2008 is striking. Between 1997 and 2007 the attainment of substantial primary surpluses allowed for an annual debt reduction of around 5 pps. In that respect, the disappearance of the primary surplus can be considered the main driving force behind the increase in the debt ratio since 2007. It highlights how the return to substantial primary surpluses is a precondition for putting debt on a clear downward trajectory and complying with the debt reduction benchmark.

According to the Commission 2017 spring forecast, the debt-to-GDP ratio would fall by 0.3 pp. in 2017. A primary surplus of 0.7% of GDP and a sizeable downward snowball effect of 0.8% of GDP as a result of rising nominal growth and decreasing interest expenditures are largely offset by upward stock-flow adjustments stemming from loans at the regional level for the financing of social housing investment and interest rate swaps. 10 The same trend is expected in 2018 when debt would decrease to 105.1% of GDP at unchanged policy.

Interest expenditure

In line with the general trend in the euro area, interest rates on Belgian debt instruments are at historical lows. The ten-year bond yield stood at 0.8% at the end of April 2017. The spread between Belgian and German bonds has been broadly stable for several years. It averaged 50, 32, 37 and 48 basis points in 2014, 2015, 2016 and the first four months of 2017 respectively, compared to a maximum of 366 basis points at the end of November 2011. The implicit interest rate declined steadily in recent years, from 4.6% in 2007 to 2.7% in 2016. It is projected to decline further to 2.3% in 2018.

Debt sustainability

Belgian authorities have been using those favourable market conditions to refinance the outstanding debt against much lower rates at considerably longer maturity. The average maturity of long-term issuance rose to 17.5 years in 2016 (14.1 years in 2015 and 15 years in 2014) with an average weighted yield of 0.8% (0.9% in 2015 and 2.2% in 2014). As a result, the average life to maturity of the total federal debt portfolio 11 rose to 8.7 years at the end of 2016 and 9.2 years at the end of April 2017 12 . It is the longest ever and compares to around 6 years until 2009 and 8 years at the end of 2015 13 . The 12-month and 60-month refixing risk 14 of the federal debt decreased from around 20% and 57% at the end of 2012 to around 19% and 43% at the end of 2016 15 . Currently, Belgium does not appear to face a risk of financial stress in the short term. If interest rates were to start rising, the high debt level implies a substantial hike in interest expenditure over time, though the high average life to maturity means that that hike would materialise only gradually.

The sensitivity to potential shocks in nominal growth and interest rates as well as the unfavourable starting point result in high sustainability risks in the medium term. At unchanged policy, the debt level is projected to decline to 102% of GDP by 2027 16 . A 1 pp. increase in the interest rate assumptions or 0.5 pp. lower GDP growth would bring the debt level to 108-109% of GDP in 2027. Adequate progress towards Belgium's MTO, as required by the Stability and Growth Pact, would put the debt on a sustained downward path, arriving at 80% of GDP by 2027. However, the fiscal effort required for reaching the MTO is substantial, considering that the structural deficit is estimated at 2.0% of GDP in 2018 at unchanged policy. Moreover, rising expenditure might require additional measures once at the MTO.

Lastly, the sustainability of public debt is also determined by the economy's growth potential. As described above, the gradual decline of total factor productivity growth since the beginning of the 1990s has lowered potential growth. It underscores the importance of implementing structural reforms in order to boost potential growth. Progress with regard to reforms was discussed in Section 4.1.

4.4.    Other factors considered relevant by the Commission

Among the other factors considered relevant by the Commission, particular consideration is given to financial contributions to fostering international solidarity and achieving the policy goals of the Union, the debt incurred in the form of bilateral and multilateral support between Member States in the context of safeguarding financial stability, and the debt related to financial stabilisation operations during major financial disturbances (Article 2(3) of Regulation (EC) No 1467/97).

Rescue operations in the financial sector explain part of the debt increase since 2007 as discussed in section 4.3. The direct cumulative debt impact of those operations reached almost 7% of GDP in 2011 but declined to around 3.5% of GDP as of 2015 due to the sale of some of the acquired assets as well as the reimbursement of loans. Contingent liabilities related to guarantees granted to the financial sector all relate to Dexia. Awaiting full resolution, the Belgian State guarantees 51.4% of Dexia's liabilities. Those guarantees reached 8.7% of GDP at the end of 2016, up from 7.7% at the end of 2015.

Article 12(1) of Regulation (EU) No 473/2013 requires that this report considers also "the extent to which the Member State concerned has taken into account the Commission's Opinion on the country's Draft Budgetary Plan, as referred in Article 7(1)" of the same Regulation. The Commission Opinion on Belgium's draft budgetary plan for 2017 pointed to a risk of non-compliance with the provisions of the SGP in 2016-2017 and invited the authorities to implement all planned measures within the national budgetary process and to ensure that the 2017 budget complies with the SGP 17 . The federal budget was adopted by Parliament on 22 December 2016 without major changes compared to the Draft Budgetary Plan. In March 2017 the federal government carried out a budget review, which largely consisted of updated assessments of earlier announced measures and of underlying assumptions, with some additional spending measures announced as well. The overall impact of the March review on the Commission projections is therefore limited.

4.5.    Other factors put forward by the Member State

On 9 May 2017, the Belgian authorities transmitted a letter with relevant factors in accordance with Article 2(3) of Regulation (EC) No 1467/97. The analysis presented in the previous sections already broadly covers the key factors put forward by the authorities.

The authorities notably argue that the MLSA for Belgium was excessively demanding, referring to the Commission reports under Article 126(3) TFEU of 27 February 2015 and 18 May 2016 which stated that such an effort was neither feasible nor desirable. Secondly, the authorities refer to the continuation of the structural reform agenda by the Belgian government, in line with what is mentioned in Section 4.1. On top of reforms in the area of competitiveness, the federal government highlights the tax shift away from labour in 2015-2020 as well as planned changes to the pension system such as a reform of assimilated periods, the extension of the second pension pillar to all workers and the introduction of a partial pension.

In their letter, the Belgian authorities also point to the fact that the debt ratio has fallen in 2015-2016, for the first time since the economic and financial crisis of 2008, and would continue decreasing as of 2017. The letter highlights the upward impact that weak growth, support to financial institutions, lending to Greece and contributions to the EFSF/ESM have had on the debt ratio. The letter announces that the recent sale of part of the stake in BNP Paribas will reduce the debt level by almost 0.5 pp. of GDP from 2017 onwards. According to the authorities' calculations, the budgetary path presented in the 2017 Stability Programme should enable compliance with the debt criterion by 2019 at the latest. The authorities also invoke exceptional refugee-related costs and security measures. While they are neutralised with respect to the adjustment path towards the MTO, they have an impact on the headline balance and thus on public debt developments.

5.    Conclusions

General government gross debt stood at 105.9% of GDP at the end of 2016, well above the 60% of GDP reference value. Belgium did not make sufficient progress towards compliance with the debt reduction benchmark in 2016. Moreover, the Commission forecast does not expect Belgium to comply with the debt reduction benchmark either in 2017 or in 2018, based a no-policy-change assumption. This suggests that before consideration is given to all relevant factors, the debt criterion as defined in the Treaty does not appear to have been fulfilled prima facie in 2016. In line with the Treaty, this report also examined the relevant factors.

Unfavourable economic conditions in the recent past partly explain non-compliance with the debt criterion in 2016. However, economic conditions have been improving and are no longer considered a strong mitigating factor in explaining Belgium's gap compared to the forward-looking dimension of the debt reduction benchmark in both 2017 and 2018 according to the Commission forecast.

On the basis of the Commission 2017 spring forecast Belgium was found to be broadly compliant with the required preventive arm adjustment towards the MTO in 2016. The same holds with regard to 2017. However, for 2016 and 2017 taken together Belgium is considered to be at risk of non-compliance given the accumulated deviation. That projected deviation can, however, still be corrected in 2017.

Belgium has been making progress in implementing the structural reforms announced since the beginning of 2015, notably in the area of pensions, competitiveness and taxation. For several of those reforms progress is considered substantial. They are expected to contribute to enhancing the economy's growth potential and reducing the risks of macroeconomic imbalances, thereby having a positive impact on debt sustainability in the medium to long term. The non-budgetary neutral nature of the tax reform undertaken has worsened the budgetary position, though. In a letter sent on 9 May 2017, Belgium announced further reforms, in particular regarding pensions.

Overall, the analysis presented in this report includes the assessment of all the relevant factors and notably: (i) the previously unfavourable but improving macroeconomic conditions, which makes them less of a factor to explain Belgium's large gaps as regards compliance with the debt reduction benchmark; (ii) the fact that, based on the Commission forecast, the deviations from the required adjustment towards the MTO point to a risk of some deviation in 2016 and 2017 individually, but to a significant deviation in 2016 and 2017 together, which can still be corrected in 2017; and (iii) the implementation of growth-enhancing structural reforms in recent years, several of which are considered substantial and projected to help improve debt sustainability. In concluding, the current analysis suggests that the debt criterion as defined in the Treaty and in Regulation (EC) No 1467/1997 should be considered as currently complied with. At the same time, additional fiscal measures are to be taken in 2017 to ensure broad compliance with the adjustment path towards the MTO in 2016 and 2017 together.

(1) 1OJ L 209, 2.8.1997, p. 6. The report also takes into account the “Specifications on the implementation of the Stability and Growth Pact and guidelines on the format and content of stability and convergence programmes”, adopted by the Economic and Financial Committee on 5 July 2016, available at:
http://ec.europa.eu/economy_finance/economic_governance/sgp/legal_texts/index_en.htm .

(2) 2Regulation (EU) No 473/2013 of the European Parliament and of the Council on common provisions for monitoring and assessing draft budgetary plans and ensuring the correction of excessive deficit of the Member States in the euro area (OJ L 140, 27.5.2013, p. 11).

(3) According to Regulation (EC) No 479/2009, Member States have to report to the Commission, twice a year, their planned and actual government deficit and debt levels. The most recent notification of Belgium can be found at: http://ec.europa.eu/eurostat/web/government-finance-statistics/excessive-deficit-procedure/edp-notification-tables.

(4) Eurostat news release No 67/2017, http://ec.europa.eu/eurostat/documents/2995521/7997684/2-24042017-AP-EN.pdf/d83f50f3-ecab-457a-a46b-f58d3e42a030

(5) Initially, an annual structural adjustment of 0.9% of GDP over 2014-2016 would have ensured that – if followed – Belgium would comply with the debt reduction benchmark at the end of the transition period. Given the shortfall in 2014, the required annual structural adjustment over 2015-2016 rose to 1.4%. The additional shortfall in 2015 from this increased MLSA brought the required annual structural adjustment to comply with the debt reduction benchmark in 2016, the last year of the transition period, to 2.2%.

(6)  Compliance with the debt benchmark is assessed on the basis of three different configurations: the backward-looking, the forward-looking and the debt reduction benchmark adjusted for the impact of the cycle.

(7) Stability Programme Belgium 2017-2020, p. 18.

(8) E.g. the harmonisation of the valorisation of years of study in the three pension schemes is estimated to lower the projected increase in ageing costs by 0.1 pp. of GDP in 2060. See Federal Planning Bureau (2017), Une réforme de la régularisation des périodes d'études dans les régimes belges de pension – Estimation des effets budgétaires.

(9) High Council of Finance (2017), Avis 'Trajectoire budgétaire en préparation du Programme de Stabilité 2017-2020'. Based on data provided by the Federal Planning Bureau.

(10) When adjusting the Commission forecast for the sale of part of the BNP Paribas participation in May 2017, debt would fall from 105.9% of GDP in 2016 to 105.2% in 2017 and to 104.6% in 2018, provided proceeds are fully used to bring down debt.

(11) The federal debt represents 84% of the general government debt.

(12)  Belgian Debt Agency, Annual report 2016.

(13) Belgian Debt Agency, Borrowing requirements & Funding plan 2017.

(14) The proportion 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.

(15) Belgian Debt Agency, Borrowing requirements & Funding plan 2017.

(16) 2016 Debt Sustainability Monitor. Those projections start from the European Commission 2017 winter forecast, with the no-policy change assumption translated into a structural primary balance kept constant (excluding ageing costs) at the level of the last year of the forecast (2018). The baseline scenario is based on the following macroeconomic assumptions for the long term: potential GDP growth remains around 1.4%; inflation and the change in the GDP deflator stabilise at 2% in the medium term; long-term interest rates on new and rolled-over debt converge to 3% in real terms by 2026 and short-term rates to a value consistent with the long-term interest rate and historical (pre-crisis) euro area yield curve (see also European Commission, 2012). Projected ageing costs are based on the 2015 Ageing Report.

(17)  In its comments and remarks on the 2017 budget, the Belgian Court of Auditors made a number of reservations, in line with issues raised by the Commission in its draft budgetary plan evaluation, and stressed several risk factors surrounding the budget. A general remark concerned the lack of detail for several measures and projections. See Commentaar en opmerkingen bij de ontwerpen van staatsbegroting voor het begrotingsjaar 2017, Rekenhof, 2016.