EU - Main contents

Kerngegevens
Text
Referenced document
EU Monitor

1. Kerngegevens

Document date	07-02-2020
Publication date	08-02-2020
Reference	5817/20 ADD 1
From	Secretary-General of the European Commission, signed by Mr Jordi AYET PUIGARNAU, Director
External link	original article
Original document in PDF

2. Text

Council of the European Union

Brussels, 7 February 2020 (OR. en)

5817/20 ADD 1

ECOFIN 51 UEM 19

COVER NOTE

From: Secretary-General of the European Commission, signed by Mr Jordi AYET PUIGARNAU, Director

date of receipt: 6 February 2020

To: Mr Jeppe TRANHOLM-MIKKELSEN, Secretary-General of the Council of the European Union

No. Cion doc.: SWD(2020) 210 final

Subject: COMMISSION STAFF WORKING DOCUMENT Report on the application of Regulations (EU) No 1173/2011 i, 1174/2011, 1175/2011 i, 1176/2011, 1177/2011 i, 472/2013 i and 473/2013 and Council Directive 2011/85/EU Accompanying the document COMMUNICATION FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN CENTRAL BANK, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS Economic governance review Report on the application of Regulations (EU) No 1173/2011 i, 1174/2011, 1175/2011 i, 1176/2011, 1177/2011 i, 472/2013 i and 473/2013 and Council Directive 2011/85 i/EU

Delegations will find attached document SWD(2020) 210 final.

Encl.: SWD(2020) 210 final

EUROPEAN COMMISSION

Brussels, 5.2.2020 SWD(2020) 210 final

COMMISSION STAFF WORKING DOCUMENT

Report on the application of Regulations (EU) No 1173/2011 i, 1174/2011, 1175/2011 i, 1176/2011, 1177/2011 i, 472/2013 i and 473/2013 and Council Directive 2011/85 i/EU

Accompanying the document

COMMUNICATION FROM THE COMMISSION TO THE EUROPEAN

PARLIAMENT, THE COUNCIL, THE EUROPEAN CENTRAL BANK, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE

OF THE REGIONS

Economic governance review

Report on the application of Regulations (EU) No 1173/2011 i, 1174/2011, 1175/2011 i, 1176/2011, 1177/2011 i, 472/2013 i and 473/2013 and Council Directive 2011/85 i/EU

{COM(2020) 55 final i} - {SWD(2020) 211 final} COMMISSION STAFF WORKING DOCUMENT

Report on the application of Regulations (EU) No 1173/2011 i, 1174/2011, 1175/2011 i, 1176/2011, 1177/2011 i, 472/2013 i and 473/2013 and Council Directive 2011/85 i/EU

Accompanying the document

COMMUNICATION FROM THE COMMISSION TO THE EUROPEAN

PARLIAMENT, THE COUNCIL, THE EUROPEAN CENTRAL BANK, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE

OF THE REGIONS

Economic governance review

Report on the application of Regulations (EU) No 1173/2011 i, 1174/2011, 1175/2011 i, 1176/2011, 1177/2011 i, 472/2013 i and 473/2013 and Council Directive 2011/85 i/EU 1

1 Six pack: Regulation (EU) No 1173/2011 i of the European Parliament and of the Council of 16 November 2011

on the effective enforcement of budgetary surveillance in the euro area, OJ L 306, 23.11.2011, p. 1; Regulation (EU) No 1174/2011 i of the European Parliament and of the Council of 16 November 2011 on enforcement measures to correct excessive macroeconomic imbalances in the euro area, OJ L 306, 23.11.2011, p. 8; Regulation (EU) No 1175/2011 i of the European Parliament and of the Council of 16 November 2011 amending Council Regulation (EC) No 1466/97 i on the strengthening of the surveillance of budgetary positions and the surveillance and coordination of economic policies, OJ L 306, 23.11.2011, p. 12; Regulation (EU) 1176/2011 i of the European Parliament and of the Council of 16 November 2011 on the prevention and correction of macroeconomic imbalances, OJ L 306, 23.11.2011, p. 25; Regulation (EU) No 1177/2011 i of 8 November 2011 amending Regulation (EC) No 1467/97 i on speeding up and clarifying the implementation of the excessive deficit procedure, OJ L 306, 23.11.2011, p. 33; Directive 2011/85 i/EU of the Council of 8 November 2011 on the requirements for budgetary frameworks of the Member States, OJ L 306, 23.11.2011, p. 41; Two-pack: Regulation (EU) No 472/2013 i of the European Parliament and of the Council of 21 May 2013 on the strengthening of economic and budgetary surveillance of Member States in the euro area experiencing or threatened with serious difficulties with respect to their financial stability, OJ L 140, 27.5.2013, p. 1; Regulation (EU) No 473/2013 i of the European Parliament and of the Council of 21 May 2013 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.

Table of Contents

1.

Introduction ........................................................................................................................ 3

2.

Fiscal surveillance .............................................................................................................. 3

2.1 Objectives of the Stability and Growth Pact .................................................................... 3

2.2 Implementation of the Stability and Growth Pact ............................................................ 6

2.3 Developments at the level of national fiscal frameworks ................................................ 9

2.4 Effectiveness of the fiscal surveillance framework ........................................................ 11

2.5 Conclusions .................................................................................................................... 27

3.

Macroeconomic Imbalance Procedure ............................................................................. 29

3.1 Objectives of the Macroeconomic Imbalance Procedure ............................................... 29

3.2 Implementation of the MIP ............................................................................................ 30

3.2.1 The Alert Mechanism Report .................................................................................. 30

3.2.2 In-depth reviews and classification of imbalances .................................................. 33

3.2.3 Recommendation, monitoring, and enforcement .................................................... 37

3.3 Effectiveness of the MIP ................................................................................................ 41

3.3.1 Follow-up to policy recommendations .................................................................... 41

3.3.2 The development of a supportive policy environment ............................................ 43

3.3.3 Impact on economic outcomes ................................................................................ 46

3.4 Conclusions .................................................................................................................... 49

4.

Euro area Member States experiencing or threatened with serious difficulties with financial stability ..................................................................................................................................... 54

4.1 Objectives ....................................................................................................................... 54

4.2 Enhanced surveillance .................................................................................................... 54

4.3 Financial assistance and macroeconomic adjustment programmes ............................... 55

4.4 Post-programme surveillance ......................................................................................... 56

4.5 Conclusions .................................................................................................................... 56

5 Overall findings ................................................................................................................ 58

Annex 1: Assessing the impact of the SGP: empirical evidence ............................................. 60

Annex 2: Assessing the impact of the MIP on policy progress: an empirical assessment ....... 67

1.

I NTRODUCTION

This Staff Working Document provides evidence and analytical underpinning for the Commission Communication on the economic governance review. The review encompasses legislative packages known as the six-pack and two-pack, which were adopted with the aim of strengthening the EU’s economic governance system in response to the

economic and financial crisis. 2 The objective of the review, which is to be carried out every

five years as mandated in the seven Regulations, is to assess their effectiveness, the progress in ensuring closer coordination of economic policies and sustained convergence in the economic performance of the Member States in accordance with the Treaty of the Functioning

of the European Union, and where applicable 3 the contribution of the Regulations to the

achievement of the Union’s strategy for growth and jobs. All seven Regulations are equipped with a clause that the Commission shall submit to the European Parliament and to the Council a report on the application of the legislation, accompanied, where appropriate, by proposals for amendments.

In the following sections, the effectiveness of the individual surveillance strands is assessed against their key objectives. Section 2 outlines the main objectives of the EU fiscal framework, its evolution over time and the way it has been implemented. It also looks at the related developments within the national fiscal frameworks. It concludes by assessing the effectiveness of both the EU and national fiscal frameworks in strengthening public finances in various respects. Section 3 presents the main objectives of the Macroeconomic Imbalance Procedure as well as the evolution in its implementation, and discusses its effectiveness in fostering policy action and favouring the correction of imbalances. Section 4 evaluates Regulation No 1173/2011 i on the surveillance of Member States in the euro area experiencing or threatened with serious difficulties with respect to their financial stability. It looks at provisions regarding enhanced surveillance, and macroeconomic adjustment programmes as well as post-programme surveillance and the implementation thereof. Finally, section 5 provides the key conclusions.

2.

F ISCAL SURVEILLANCE

2.1 Objectives of the Stability and Growth Pact

The primary objective of the Pact is to ensure the sustainability of public debt. The Pact aims to prevent, and where necessary correct, excessive deficits so as to keep debt at sustainable levels. 4 Safe levels of debt allow so-called ‘automatic stabilisers’ to operate (e.g. higher expenditure on unemployment benefits and lower tax revenues during downturns) without leading to fiscal or financial market stress. Without prejudice to the objective of sustainability, the Pact is also meant to allow for fiscal-based macroeconomic stabilisation.

2 This Staff Working Document covers the seven Regulations included in the six and two-pack. With regards to Council Directive 2011/85 i/EU on requirements for budgetary frameworks of the Member States, which has a specific review clause, a dedicated review assesses its overall suitability, including the suitability of the statistical requirements for all sub-sectors of government, the design and effectiveness of numerical fiscal rules, and the general level of transparency of public finances in the Member States.

3 This is mandated in Regulation No 472/2013 i and Regulation No 473/2013 i.

4 A summary of the challenges associated with designing supranational fiscal rules can be found in Yared, P. (2019), ‘Rising Government Debt: Causes and Solutions for a Decades-Old Trend’, Journal of Economic Perspectives, Vol 33 No 2.

Differences in national preferences on the appropriate pace of debt reduction makes reconciling the sustainability and stabilisation objectives challenging.

The corrective arm of the Pact aims to correct excessive deficits when they occur. The Treaty defines deficit and debt thresholds, i.e. a deficit below 3% of GDP and a debt-to-GDP ratio of below 60% of GDP, unless the ratio is sufficiently diminishing towards that level at a satisfactory pace. Breaches of those thresholds are referred to as ‘gross policy errors’ and trigger an excessive deficit procedure (EDP), as set out in Article 126 TFEU. Member States in an EDP are required to correct their excessive deficit situation by a deadline set by the Council. If a Member State fails to put in practice the recommendations of the Council, the procedure may be stepped up. Member States concerned can be subject to the suspension of

payments or commitments under the European Structural and Investment Funds. 5 Euro-area

Member States can also be subject to financial sanctions under the Pact.

The preventive arm of the Pact aims to prevent the occurrence of excessive deficits and ensuring sound public finances. It aims to ensure that Member States build fiscal buffers during economic good times so that they have sufficient fiscal space for economic stabilisation during downturns. It requires Member States to reach a country-specific mediumterm budgetary objective (MTO), which is defined in structural terms, i.e. adjusted for the economic cycle. Meeting the MTO gives Member States scope to use fiscal policy for the purpose of macroeconomic stabilisation in bad economic times, i.e. by allowing automatic stabilisers to operate or by making policy choices that support the economy (e.g. increasing investment). The minimum level of a country’s MTO is determined based on the Member State’s debt level, the budgetary impact of population ageing, and cyclical volatility of its economy in order to ensure an appropriate safety margin with regard to the 3% of GDP deficit threshold. For euro-area Member States, the MTO cannot be lower than a structural deficit of

1% of GDP. 6 Member States that have not reached their MTO are required to adjust towards

that objective (by 0.5% of GDP per year as a benchmark).

The Pact has developed considerably since its creation, largely in response to weaknesses that became apparent during periods of crisis. The conceptual underpinning of the framework has been gradually elaborated and its scope extended with a view to making the rules ‘smarter’, i.e. better adapted to the economic situation. Based on the observation that Member States fail to make necessary fiscal adjustments during economic good times, and that debt levels tend to increase considerably during recessions, successive reforms have made the preventive arm of the Pact more central and increased the focus on debt developments. In particular, the economic and financial crisis shifted the focus to risks arising from unsustainable levels of public debt, and to possible debt crises, which could threaten the financial stability of the euro area as a whole. In a context of heightened market pressure and rapidly rising debt levels, there was a need to decisively strengthen safeguards for debt sustainability.

The 2011 six-pack reform reinforced the preventive arm of the Pact, in particular to ensure adequate efforts in good times. It introduced a greater emphasis on aggregate expenditure developments and revenue-increasing (or decreasing) policy measures, which are more directly under the control of governments than other fiscal indicators. Accordingly, a Member State’s adjustment towards its MTO is assessed annually on the basis of two

5 This is the so-called macro-economic conditionality of European Funds, which falls outside the scope of the Stability and Growth Pact. It is not a sanction, but instead aims to safeguard the funds belonging to the Union budget.

6 In addition, the intergovernmental Treaty on Stability, Coordination and Governance in EMU sets a lower limit of a structural deficit of 0.5% of GDP for its signatories. See section 2.3

measures: 1) the growth rate of primary expenditure developments net of discretionary revenue measures, relative to medium-term economic growth (the ‘expenditure benchmark’); and 2) the change in the cyclically-adjusted budget balance net of one-off measures (the ‘structural balance’). Member States that significantly deviate from their MTO or from the adjustment path towards it (i.e. by more than 0.5% of GDP), are required to correct the deviation within a given timeframe under the so-called ‘significant deviation procedure’. For euro-area Member States, insufficient correction (i.e. no ‘effective action’) can lead to the application of sanctions in the form of an interest-bearing deposit. Finally, reflecting the economic and financial crisis, a collective ‘escape clause’ was introduced, effectively allowing (but not prescribing) a suspension of the rules in case of ‘a severe economic downturn’ in the EU or the euro area as a whole.

In the corrective arm, the six-pack reform operationalised the debt criterion and strengthened the sanction mechanism. The six-pack introduced the debt reduction benchmark, making the Treaty’s debt criterion of sufficiently diminishing towards the 60% threshold operational. For euro-area Member States, sanctions under the excessive deficit procedure are now meant to kick in at an earlier stage and be more automatic, through the application of reverse qualified majority voting for Council decisions on Commission

proposals. 7

The 2013 two-pack reform strengthened budgetary coordination in the euro area in view of potential consequences of spillovers within the Economic and Monetary Union. In particular, it established an obligation for euro-area Member States to submit their draft budgetary plans to the Commission for an opinion before their adoption by national parliaments. Furthermore, it strengthened budgetary coordination through a discussion in the Eurogroup of those Commission opinions and of the budgetary situation and prospects in the euro area as a whole. It also reinforced the surveillance for euro-area Member States with excessive deficits.

After the entry into force of the six- and two-pack, various implementation decisions and approaches have been codified and endorsed by the Council, in particular the use of flexibility within the EU’s fiscal rules. The need to restore market confidence between 2010 and 2012 forced some Member States to implement large fiscal adjustments, sometimes going beyond the requirements of the Pact. Moreover, after the economic and financial crisis, the euro area experienced a second recession in 2012-2013. This was followed by a period of low growth relative to the pre-crisis period and low inflation, with monetary policy facing the zero lower bound and the European Central Bank introducing a number of non-standard monetary policy measures. In the context of a slow recovery from the economic and financial crisis, the Commission has placed an emphasis on making best use of the flexibility within the EU’s fiscal rules. That was operationalised by an interpretative Commission’s Communication on

Flexibility within the SGP in January 2015 8 and a Commonly Agreed Position on flexibility endorsed by the Council. 9 Namely, taking into account the need for the same highly-indebted Member States to implement reforms in labour and product markets, the Commission operationalised the so-called structural reform clause. The particular nature of investment has

7 Under the intergovernmental Treaty on Stability, Coordination and Governance in EMU, euro-area Member States committed to also use reverse qualified majority voting on other proposals or recommendations submitted by the Commission for euro-area Member States in breach of the deficit criterion in the framework of an excessive deficit procedure.

8 European Commission (2015), ‘Making the best use of the flexibility within the existing rules of the Stability and Growth Pact’, COM(2015)12 i of 13 January 2015.

9 Commonly agreed position on flexibility within the Stability and Growth Pact, endorsed by the Council on 12 February 2016.

also been taken into account, to a limited extent, by allowing temporary deviations from the fiscal requirements to cater for investment needs. Lastly, the Flexibility Communication and Commonly Agreed Position on flexibility introduced a matrix to determine the required adjustment under the preventive arm that takes into account cyclical conditions. Concretely, the required annual fiscal adjustment towards the MTO varies depending on the position in the economic cycle (i.e. higher in good times, lower in bad times) and the level of debt (below or above 60% of GDP).

Lastly, the Commission increasingly made use of the discretion available within the framework when applying the fiscal rules. It considered compliance with the preventive arm a key relevant factor when assessing compliance with the Treaty’s debt criterion. That discretion also included the use of the concept of ‘broad compliance’ with the preventive arm requirements as well as specific changes to the fiscal requirements, as recalled by the Court of Auditors. 10 The Commission has fostered a regular dialogue with Member States and the European Parliament and has consistently sought to clarify the rationale for its decisions in a

cross-country perspective. 11 The framework has nevertheless become more complex and

arguably less predictable. The increased use of discretion also reflected the fact that the sixand two-pack reforms increased the Commission’s responsibility for decisions, such as the imposition of sanctions, having far-reaching political implications in the Member States.

In parallel to the evolution of the Pact, EU requirements for national fiscal frameworks have been introduced and have become increasingly precise. There has been a growing recognition of the importance of fiscal arrangements at the national level as a means of ensuring compliance with EU fiscal rules. In that respect, national frameworks are perceived as serving two goals: first, to provide the necessary setting for the implementation of fiscal policy in compliance with EU rules; and second, as a means of strengthening national ‘ownership’ of EU rules. Member States agreed on a number of requirements for their fiscal frameworks with a view to improving their quality and effectiveness in the support of fiscal discipline in the EU context. Most of those requirements took the form of EU law, while others, more intrusive, were established in the so-called ‘Fiscal Compact’ as part of the intergovernmental Treaty on Stability, Coordination and Governance in the Economic and Monetary Union. 12

2.2 Implementation of the Stability and Growth Pact

There were no Member States in an EDP at the end of 2019. While there have frequently been breaches of the Pact’s deficit threshold of 3% of GDP since its inception, the economic and financial crisis resulted in an exceptional situation in which almost all Member States breached that threshold (see Graph 1). Due to the depth of the recession and bank recapitalisation needs in some Member States, the headline deficit at the aggregate EU level exceeded 6% of GDP in 2009-2010. As a consequence, 24 out of the then 27 Member States

entered the EDP for breaching the Treaty’s deficit criterion. 13 By 2018, all Member States had

10 European Court of Auditors, Is the main objective of the preventive arm of the Stability and Growth Pact delivered?, Special report No 18/2018;

11 Each year, the Directorate-General for Economic and Financial Affairs of the European Commission publishes a Vade Mecum on the Stability and Growth Pact that brings together all the relevant procedures and methodologies involved in the implementation of the EU's rules-based fiscal policy framework. For the 2019 edition, see: https://ec.europa.eu/info/publications/vade-mecum-stability-and-growth-pact-2019-edition_en 12 They included also substantive rules, on top of requirements for the design of policy frameworks

13 Finland was placed in EDP for a planned breach, although the deficit eventually stayed below 3%.

brought their headline deficits below 3% of GDP. 14 Since 2015, however, headline deficits

have mainly been reduced thanks to improving macroeconomic conditions, which have generated additional revenues and lowered unemployment expenditure. Some Member States have therefore been able to correct their excessive deficits without making any significant structural fiscal adjustments (the so-called ‘nominal strategy’, i.e. aiming at headline deficit targets). In addition, some Member States, including some of the largest, still have deficits that are close to the 3% of GDP threshold.

Graph 1: Member States’ headline deficits and EU output gap

Debt dynamics across Member States have been quite divergent since the economic and financial crisis. While the aggregate EU debt remained around 60% of GDP in the pre-crisis period, it increased to almost 90% between 2009 and 2014. Indeed, the crisis led to higher cyclical and structural deficits, pushing debt-to-GDP ratios substantially upwards. That was further amplified by the impact of low real and nominal economic growth (which delays the

reduction in the ratio for any given debt level in absolute terms) 15 and by direct intervention

by some Member States in the financial sector. Aggregate debt ratios started to decline in 2015, on the back of lower primary deficits/higher surpluses, higher economic growth and historically low interest rates on public debt. Overall, the Commission forecasts the debt-to GDP ratio to continue the declining path of recent years and to fall to around 79% in 2020, which is far lower than in the USA or Japan (around 114% and 237% respectively). However, debt dynamics across Member States overall have been quite divergent since the economic and financial crisis (see Graph 2). This reflects large differences in the pace of fiscal consolidation and the impact of the economic growth-interest rate differential, as well as in country-specific fiscal costs related to support measures for the banking sector. As a consequence, around half of the Member States now have debt levels (again) below 60% of GDP while some others have their debt around or (well) above 100% of GDP. The latter group of Member States account for half of the EU’s GDP.

14 With the exception of Cyprus, where there was a temporary peak in the deficit in 2018 (at 4.4% of GDP), due to the one-off support measures related to the sale of the Cyprus Cooperative Bank. In 2019, the budget balance is expected to have returned to a comfortable surplus of close to 4% of GDP.

15 In turn, growth dynamics, in both real and nominal terms, have been significantly affected by the unwinding of large macro-economic imbalances accumulated in the run-up to the crisis.

Graph 2: Debt dynamics (2018 debt levels versus pre-crisis low point and post-crisis peak)

180

160

140

120

100

EL IT PT CY BE FR ES UK HR AT SI HU IE DE FI NL SK PL MT SE LV RO DK LT CZ BG LU EE 2 8

trough 2018 peak E

Member States’ fiscal positions in relation to their MTOs also vary widely. All Member States have now moved to the preventive arm of the Pact and are thus required to maintain or make progress towards their MTOs. Around half of them have reached those objectives, with some accumulating structural surpluses (going beyond the requirements of the preventive arm) and a few others being close to their MTOs (see Graph 3). However, a group of Member States remains far away from their objectives and have not yet built sufficient fiscal buffers for the next downturn, notably with respect to the 3% of GDP headline deficit threshold. Most

of those Member States also have the highest debt ratios in the EU. 16

Graph 3: How far Member States were from their MTO in 2018?

5 Member States above or at MTO Member States close to MTO Member States far from MTO

-1

-2

-3

EL BG LU DE CY CZ HR DK NL SE MT AT LT IE PL FI PT LV SI SK EE RO BE UK FR HU IT ES

16 For a more detailed overview of recent fiscal developments in the EU, see European Commission (2019), ‘Fiscal policy’, Quarterly Report of the Euro Area, Volume 18 N. 2, Institutional Paper 119, November 2019. https://ec.europa.eu/info/publications/quarterly-report-euro-area-volume-18-no-2-2019_en

Expenditure dynamics now seem better controlled than before the crisis. The pre-crisis period saw government primary expenditure in most Member States grow faster than potential output. Since the crisis, the growth rate of primary expenditure has slowed down in relation to potential output, which itself was significantly affected by the crisis. Average expenditure growth since the crisis has actually been at or below the growth of potential output in nearly all Member States in spite of the concomitant slowdown in inflation (see Graph 4). The difference between the two periods would likely be even larger if discretionary revenue measures were also considered. Indeed, the pre-crisis period saw the majority of Member States cutting taxes, a trend that reversed in the aftermath of the economic and financial crisis (up to 2015).

Graph 4: Controlling primary expenditure dynamics

Pre-crisis periodPost-crisis period

2.3 Developments at the level of national fiscal frameworks

EU requirements for national budgetary frameworks have evolved rapidly in terms of scope and granularity in recent years. While EU policy-makers have paid increasingly more attention to national fiscal frameworks since the introduction of the euro, Directive 2011/85 i was the first EU legislative initiative in that area. The Directive introduced a set of binding minimum requirements for the main elements of budgetary frameworks, 17,18 although the requirements themselves remained fairly general. In particular, it requires Member States to adopt numerical fiscal rules conducive to the respect of the Maastricht criteria and defines general requirements for their design and specification, while leaving the choice of specific rules to the discretion of Member States. It was soon followed by further initiatives in the area of domestic budgetary frameworks, i.e. the Fiscal Compact and the two-pack. The Fiscal Compact, though not part of EU law, has had a big impact on national frameworks: it required

17 See also Review of the suitability of Council Directive 2011/85 i/EU on requirements for budgetary frameworks of the Member States, Commission Staff Working Document ( SWD(2020) 211). 18 i.e.: (i) systems of budgetary accounting and statistical reporting; (ii) rules and procedures governing the preparation of forecasts for budgetary planning; (iii) country-specific numerical fiscal rules; (iv) medium-term budgetary frameworks; and (v) transparency of public finances and mechanisms that regulate fiscal relationships between public authorities across sub-sectors of general government.

its euro area signatories 19 to introduce in national legislation (preferably at constitutional

level) a structural balanced budget rule – effectively reproducing the MTO - and to complement it with an automatic correction mechanism, as well as an independent fiscal institution to monitor the rule. Regulation No 473/2013 i, which concerns only euro-area Member States, introduced the criteria for establishing independent fiscal institutions into EU law and added requirements related to:

1)

independent macroeconomic forecasts being used for budgeting; and

2)

a common budgetary timeline across euro-area Member States.

As a result, the current economic governance system blends various requirements for domestic budgetary frameworks, with Directive 2011/85 i providing basic standards for all Member States, complemented by reinforcements in specific areas introduced by Regulation No 473/2013 i for euro-area Member States and by the Fiscal Compact for euro-area and voluntarily adhering non-euro-area Member States.

Since the adoption of the EU legal requirements and the Fiscal Compact, there has been significant development of national budgetary frameworks in Member States. The progress is most noticeable in Member States that had no or only rudimentary frameworks before the crisis. For those Member States, the EU requirements served as a basis for a modern fiscal framework.

Numerical fiscal rules are now at the centre of fiscal frameworks in all Member States and their quality has improved markedly. The number of fiscal rules has increased significantly in the Member States in recent years. However, rather than just the mere number of rules, which (if too high) can create challenges of its own, the greater quality of rules should be emphasised (see Graph 7). Overall, the Member States now have sets of fiscal rules that are more specific in terms of target and scope, generally have some consequences for non-compliance and are usually monitored on the basis of independent analysis. Where they contain escape clauses, these are limited in number and generally well-defined.

All Member States have a medium-term budgetary framework. Extending the horizon of fiscal policy beyond the annual budget brings benefits in terms of credibility, transparency and discipline. Recognising this, Directive 2011/85 i required Member States to establish credible and effective medium-term budgetary frameworks (MTBF). Currently, all Member States have a more or less developed MTBF based on multiannual objectives for the main budgetary aggregates, a noteworthy improvement over the period before the adoption of Directive 2011/85 i. However, as that Directive contains only general requirements on MTBFs, their design differs significantly across the Member States.

19 Outside the euro area, Bulgaria, Denmark and Romania have also submitted voluntarily to the same requirements.

Graph 5: Main design features of new or Graph 6: Number of independent fiscal reformed national rules institutions (IFIs) in the EU

1990-2011 New/reformed rules

0.8

0.6

0.4

0.2

Note: Each indicator ranges between 0 (non-existent) and 1 Source: Commission’s Fiscal Governance Database.

(very strong) and is averaged across all rules in force at all levels of government.

Source: Commission’s Fiscal Governance Database.

The transparency and independent monitoring of fiscal policy have improved significantly. Independent institutions have been set up (or reinforced) in virtually all Member States (see Graph 6), with a mandate to monitor public finances, in particular national fiscal rules, and to provide independent assessment. New streams of fiscal statistics have been created, most notably on Member States’ contingent liabilities, which has helped to raise awareness of specific aspects of public finances among policy-makers and the general public. The requirements of Regulation (EU) No 473/2013 i for the preparation or endorsement of macroeconomic forecasts by independent fiscal institutions appear to have led to slightly more prudent forecasting.

2.4 Effectiveness of the fiscal surveillance framework

Sustainability of public finances

The corrective arm of the Pact has been an effective tool for reducing and maintaining government deficits below the 3% of GDP threshold. As discussed above, no Member State is currently under the EDP, compared to 24 Member States in the wake of the economic and financial crisis. Furthermore, the biggest structural fiscal adjustments took place between 2011 and 2013, when most Member States were still under the EDP. 20 This suggests that the corrective arm of the Pact has proved effective in reducing excessive headline deficits (see box 1). However, as noted above, a number of Member States have corrected their excessive deficits thanks to better-than-initially-expected macroeconomic conditions rather than structural fiscal adjustments. 21

At the same time, government debt is now very high in some Member States, including some of the largest. While around half of Member States currently have debt ratios below the Treaty threshold of 60% of GDP, debt ratios increased substantially in some others following the economic and financial crisis, only stabilising in recent years. In particular, seven Member

20 This also partly reflected intense financial market pressure on some Member States during that period.

21 These Member States followed a so-called ‘nominal strategy’ whereby they met the headline deficit targets without implementing the structural adjustment targets of the EDP recommendations.

States (Greece, Italy, Portugal, Cyprus, Belgium, France and Spain) have debt ratios above or

close to 100% of GDP, as opposed to only two in the pre-crisis period (Greece and Italy). 22

There is also a clear inverse relationship between Member States’ debt levels and budget balances (see Graph 7). Member States that have met (or exceeded) their MTOs are also those with relatively low debt levels. This is the case for large Member States, such as Germany and the Netherlands, but also for Bulgaria, Czechia, Denmark, Luxembourg, Malta and Sweden. On the other hand, some of the most highly-indebted Member States (Belgium, Spain, France, Italy) still remain quite far from their MTOs and have made little progress towards them in recent years (see box 1).

Graph 7: Structural balance and debt ratios across Member States

(Size of bubbles is proportional to country shares in total EU-28 GDP)

6 Low debt / structural surplus High debt / structural surplus 1 8

) 5 EL

2 0 4 Fiscal space

P , availabe

G D 3 2

BG DE Significant pot

1 DK

CY MT adjustment

e (

% CZ

0 SE NL HR needs

la nc -1 LT SI SKFI AT PT IE

al ba -2 EE LV PL BE

tur -3 RO UK

ruc HU

St -4 Low debt / ES

-5 structural deficit High debt / structural deficit

0 20 40 60 80 100 120 140 160 180

Gross debt (%GDP, 2018)

On the basis of those observations, it could be argued that the reformed Pact has not been successful in bringing the level of debt down sufficiently in the most vulnerable Member States. Member States that were furthest away from their MTOs after the economic and financial crisis have not necessarily made more progress than those that were closer. This also suggests that the preventive arm has had little traction in those Member States (see box 1). As a consequence, some highly-indebted Member States remain far below their objectives, despite favourable economic conditions, and have thus not built sufficient buffers to provide themselves with a safety margin vis-à-vis the 3% of GDP deficit threshold in the case of deteriorating macroeconomic conditions.

Overall, the 3% of GDP reference value appears to have only acted as a target rather than as a ceiling for a number of highly-indebted Member States. This confirms the role of the 3% headline deficit ceiling as an ‘anchor for the laggards’: it acts as a safeguard for

22 It is sometimes argued that the fiscal consolidation following the economic crisis has been self-defeating in some high-debt Member States. However, the possible contractionary effects and possible induced increase in the debt-to-GDP ratio is in most cases short-lived and unwinding in the medium run. For further analysis: see European Commission (2012), ‘Consolidation and dynamics of the multipliers. Are there counter-intuitive effects?’ in ‘Report on Public Finances 2012’ European Economy 4. July 2012,

those being far away from their MTO and not progressing sufficiently toward it. 23 However,

with economic growth now clearly lower than before the crisis, maintaining headline deficits just below 3% of GDP may no longer be sufficient for some Member States to avert unsustainable debt dynamics, forcing those Member States to adjust in bad times in a procyclical manner.

At the same time, there is some evidence that compliance with the preventive arm has improved relative to the period before the reform of the Pact, especially for high debt Member States. In that respect, the operationalisation of the debt criterion is likely to have had a deterrent effect. The Commission has considered compliance with the preventive arm as a key relevant factor when assessing a possible opening of an EDP for breach of the debt criterion. That approach is likely to have had a positive impact on preventive arm compliance for those Member States.

A simulation exercise reveals that the use of the expenditure benchmark in the preventive arm contributes to more sustainable fiscal positions. Findings from counterfactual simulations show that public debt ratios would have been significantly lower than today, particularly in high-debt Member States, if the expenditure benchmark had been applied and complied with since 2000 (see Annex 1). 24 The simulations show that the positive effects from a credible and more front-loaded fiscal adjustment (higher primary budget balance and lower interest payments) outweigh the negative effects of temporary lower economic growth and inflation on the public debt ratio. Lower public debt ratios will reduce the need to consolidate in downturns and allow the automatic stabilisers to operate fully, which will help avoid future protracted losses in GDP. The findings also suggest that strict compliance with the expenditure benchmark, compared to the structural balance requirement, would have promoted a more growth-friendly path, as demonstrated by the lower average loss in output. The adjustment based on the expenditure benchmark was also found to be less procyclical than the one implied by the structural balance. The reason is that compliance with the expenditure benchmark would have required a larger fiscal adjustment in good times (as it is not impacted by revenue windfalls and interest windfalls). All in all, the findings confirm the relevance of the introduction of the expenditure benchmark in the fiscal framework in the context of the 2011 six-pack reform.

Overall, the experience so far raises the question of the extent to which the reformed Pact has made a material difference in the cases where fiscal discipline is most necessary. While the overall fiscal situation has improved on the back of the efforts done by many Member States and while an improvement in compliance is apparent compared to the pre-crisis period, the reinforced preventive arm does not seem to have had sufficient traction in several high-debt Member States. For those Member States with small fiscal imbalances, this has led to debates about the right balance between stabilisation, investment and consolidation needs. Indeed, with the development of the preventive arm, surveillance increasingly focused on promoting sound public finances, broadening the initial focus on correcting gross policy errors. From the point of economic externalities across Member States, the preventive arm may have unnecessarily focused on Member States with low public debt. Moreover, as highlighted in section 1, the changing economic context with a structural decline in potential growth and persistently low inflation and real interest rates adds new challenges to the EU’s rules-based fiscal framework. Meanwhile, the effectiveness of the 3% deficit ceiling points to the signalling value of very simple numerical rules. They are easy to

23 European Fiscal Board (2019), ‘Assessment of the EU fiscal rules with a focus on the six and two-pack legislation’.

24 European Commission (2019), ‘Performance of spending rules at the EU and Member States’ level’ in Report on Public Finances in EMU 2019, forthcoming.

grasp, not least for the general public. It should however be borne in mind that maintaining headline deficits just below 3% of GDP is not sufficient for high debt Member States to avert unsustainable debt dynamics, and forces those Member States to implement pro-cyclical fiscal consolidation in bad times.

Box 1: Performance of EU fiscal rules

The 3% of GDP headline deficit ceiling and fiscal outcomes in the EU

Empirical analysis suggests that the EU’s fiscal governance framework has contributed to

more prudent fiscal policies, although causality is difficult to establish. 25 Member States that

had large headline deficits just before the Pact was launched (in 1998) have reduced their deficits significantly, with the exception of the economic and financial crisis period. In graph A, the orange area shows the range of deficits for the quarter of Member States with the highest deficits (‘bad performers’) for each year since 1985. Before 1998, several Member States had deficits exceeding 5% of GDP. Deficits then decreased slowly until the start of the economic and financial crisis, with only three Member States having deficits exceeding 3% of GDP in 2007. In the aftermath of the crisis, Member States’ deficits rose significantly, with 24 out of the then 27 Member States entering the EDP. Since then, all Member States have corrected their excessive deficits. Overall, those developments suggest that the 3% of GDP deficit threshold has contributed to better fiscal outcomes, particularly in Member States that were characterised by high public deficits before the launch of the Pact. At the same time, the deficit threshold seems to have in some cases acted as a target rather than a ceiling. Indeed, several Member States that generally had a record of high deficits currently still have deficits close to 3% of GDP, even though macroeconomic conditions have remained favourable.

Graph A: Headline balances in the EU Member States (% of GDP)

Note: Headline balance figures from 1995 are based on ESA 2010 while previous figures are back-casted according to the observed change in the ratio as from the series based on ESA 1995. As a consequence of the variable composition of the groups of ‘bad performers’ and ‘good performers’, some of the fluctuations could be the result of composition effects (i.e. Member States shifting categories). Figures are based on the Commission’s spring 2018 forecast.

By contrast, there seems to be no clear-cut impact of the 3% deficit criterion on Member States that typically had headline surpluses or low deficits before the introduction of the Pact. The green area in graph A depicts, for each year, the range of budget balances for the quarter of Member States with the lowest deficits or highest surpluses in that year (‘good performers’). While the composition of the group has varied, there have always been Member

25 See for instance: European Commission (2019), ‘Fiscal outcomes in the EU in a rules-based framework – new evidence’ in Report on Public Finances in EMU 2018, Institutional Paper 095.

States recording (on average) surpluses since the launch of the Pact in 1998, with the exception of the years following the economic and financial crisis. This suggests that there has been no downward convergence of the good performers towards the 3% of GDP deficit criterion, as has recently been argued in policy papers. 26

Structural balances and sound budgetary positions

All Member States have now moved to the preventive arm of the Pact and are required to make progress towards sound budgetary positions, defined in the form of medium-term budgetary objectives (MTOs). Graph B shows developments in the structural balances of Member States, divided into three groups based on their debt ratio in 2011. The biggest structural adjustments took place between 2011 and 2013, when most Member States were still under the EDP. This partly reflected intense financial market pressure on some Member States during that period. Since 2014, the average fiscal effort of the Member States with the highest post-crisis debt levels (Italy, Portugal, Ireland, Belgium, France, Austria, the UK and Hungary) has slowed down. This is partly explained by the fact that improvements in headline balances in some Member States have been achieved by relying on better cyclical conditions, thus allowing those Member States in the corrective arm to achieve their required adjustments without making a structural fiscal effort. It also shows that the preventive arm has gained little traction in a number of Member States, despite the relatively high debt ratios of Member States in that group. On the other hand, Member States with lower debt have continued to adjust, even when a structurally balanced budget or surplus were reached. For both groups of Member States, however, one can detect an improvement in structural fiscal positions relative to the pre-crisis period.

Graph B: Structural balances in Member States (% of potential GDP)

1.0

0.0

-1.0

-2.0

-3.0

-4.0

-5.0

-6.0

MS with debt > 80% of GDP in 2011 (IT, PT, IE, BE, FR, AT, UK, HU)

MS with debt between 60 and 80% of GDP in 2011 (DE, MT, ES, CY, HR, NL)

MS with debt < 60% of GDP in 2011 (PL, FI, SI, DK, SK, LV, CZ, SE, LT, RO, LU, BG, EE)

Note: due to data availability, the figures up to 2009 refer to the cyclically-adjusted balance (i.e. including oneoffs).

Source:European Commission, based on the Commission’s autumn 2019 forecast

26 See for instance: Caselli, F. and Wingender, M. (2018), ‘Bunching at 3 Percent: The Maastricht Fiscal Criterion and Government Deficits’, IMF Working Paper 18/182.

Box 2: Experiences in selected Member States

It is possible for Member States under the corrective arm to apply a ‘nominal’ strategy, i.e. reaching intermediate headline targets while delivering very little or no fiscal effort, as shown by the examples of Spain and France. Both Member States broadly met their headline targets between 2016 and 2018 (Spain) and 2015 and 2017 (France) with fiscal efforts that were 0.5-1.0% of GDP smaller than those the Council had recommended. Such a strategy is possible when macroeconomic conditions turn out better than expected at the time the adjustment path was set, which is more likely to happen in excessive deficit procedures with a long adjustment period. In the case of Spain, the excessive deficit procedure in 2018 involved a less demanding requirement than those that would have been expected under the preventive arm.

Respecting the debt reduction benchmark during periods of relatively low growth and inflation has proved to be very demanding, as highlighted by the cases of Belgium and Italy. Between 2014, when it became subject to the transitional arrangements of the debt benchmark, and 2016, Belgium experienced a period of low inflation and modest real GDP growth, making it challenging to reduce its debt-to-GDP ratio. At the same time, negative inflation surprises contributed to upward revisions to required adjustments, with slippages accumulating over time. That led to the Commission giving priority to compliance with preventive arm requirements as a relevant factor in considering compliance with the debt criterion. Compliance with the preventive arm was also considered a key mitigating factor against opening debt-based excessive deficit procedures against Italy, despite non-compliance with the debt reduction benchmark. In the case of Finland, public debt hovered around the reference value of 60% of GDP between 2013 and 2017. Therefore, the Commission prepared seven consecutive reports for Finland for planned or actual breaches of the 60% threshold, even though public debt remained much lower than in many other Member States.

Various flexibility clauses and allowances have been put into operation in recent years under the preventive arm, with Italy as the main beneficiary. Over the period 2015-2018, Italy benefitted from cumulated allowances of around 1.8% of GDP, thus facilitating its broad compliance with the requirements of the preventive arm despite the absence of improvement in the structural balance over that period. The Commission Communication on the 2017 European Semester also introduced the concept of the ‘margin of discretion’, which allowed the Commission to assess Italy, together with Slovenia, against a lower requirement for 2018

than the one included in the matrix of requirements. 27

Under the preventive arm, the structural balance and expenditure benchmark can give differing indications of the fiscal effort, for example due to different underlying estimates of the growth potential of the economy. For example, both Portugal and Slovakia experienced strong downturns during the crisis period followed by strong recoveries. This led to marked falls in estimated annual potential growth rates (used in the calculation of the structural balance), followed by a rising trend. The medium-term growth rate that underlies the expenditure benchmark (i.e. an average of potential growth estimates over a period of tenyears) is less volatile. This means that, all else equal, requirements from the expenditure benchmark were less demanding than those of the structural balance in the acute phase of the crisis but, conversely, became more demanding in recent years. As a consequence, the use of the expenditure benchmark as an operational indicator for fiscal policy mitigates the risk of pro-cyclicality. Of course, other factors also contribute to differences between the two

27 European Commission (2017), ‘2017 European Semester: Country-specific recommendations’, COM(2017) 500 final i.

indicators, such as revenue windfalls/shortfalls and changes in interest payments, which both affect the structural balance but are excluded from the expenditure benchmark.

The cases of some small open economies highlight the problems associated with basing preventive arm requirements and assessments on unobservable variables. The output gaps of Slovenia, Slovakia and Latvia, for example, are subject to sizeable revisions. This is partly due to volatility in (EU-funded) public investment, which is characteristic of those Member States. The tools put in place by the Commission to deal with such uncertainty are not fully effective. For example, while the so-called ‘constrained judgement’ approach suggested a lowering of Slovenia’s preventive arm requirement for 2019 compared to the standard ‘matrix’ requirement, there exists no clearly defined methodology for setting alternative requirements. Moreover, while the Commission has placed an emphasis on the expenditure benchmark, the output gap is still a crucial factor when setting the requirement for Member States that are close to their MTOs, as their requirement is determined by the distance of their structural balance to this objective.

The case of Romania highlights the difficulty of enforcing a Significant Deviation Procedure. Romania has been subject to repeated Significant Deviation Procedures since 2017. Even so, its structural deficit increased from 0.2% GDP in 2015 to 2.7% GDP in 2018. This calls into question the effectiveness of the Significant Deviation Procedure when a Member State decides to ignore Council recommendations. It is worth recalling that the current rules do not provide for any legal sanctions in case of non-compliance by non-euroarea Member States. As a result of Romania’s structural deterioration, the Commission 2019 autumn forecast projects that the Member State will have breached the 3% of GDP headline deficit threshold in 2019 and in subsequent years.

Economic stabilisation at the level of individual Member States and the euro area

The pro-cyclicality of national fiscal policies remains an issue. Fiscal policy was already pro-cyclical in the mid-2000s, as fiscal policies were expansionary when economic times were good. National fiscal policy has remained largely pro-cyclical after the adoption of the six- and two-pack reforms. Most Member States implemented sizeable fiscal adjustments in the immediate post-crisis period, reflecting a lack of fiscal buffers at the onset of the economic and financial crisis, a need to correct excessive deficits and, in some cases, heightened market pressure. Those fiscal adjustments took place in a context of very low or even negative economic growth in some Member States and were, therefore, pro-cyclical. This likely contributed to the weak economic situation in that period, even if it aimed at bringing public finances on a sustainable path and at regaining market confidence. Fiscal consolidation came to a halt in 2014, at the moment when the output gap in the euro area was still negative. However, the aggregate fiscal stance in the euro area continued to be broadly neutral after 2016, even though the euro area’s output gap had turned positive in the meantime and the recovery had strengthened. In addition, national fiscal policies were not appropriately differentiated across Member States in light of their specific sustainability and stabilisation needs, and the fact that in many Member States the recovery was stronger than initially expected. Thus, despite the reinforced preventive arm, many Member States did not use the more benign economic times to build up counter-cyclical buffers.

Differing views among Member States about appropriate debt and deficit levels contributed to a pro-cyclical pattern of fiscal policy in good times. As discussed above, some Member States have remained just below the 3% of GDP deficit limit instead of moving towards their MTOs and remain with high debt levels, despite relatively favourable macroeconomic conditions. This raises the issue of implementation of the preventive arm, which aims to ensure that fiscal buffers are being built in good times. In contrast, other Member States have sought to exceed their MTOs and in spite of strong fiscal positions are only gradually implementing the Council's recommendations to use fiscal and structural policies to achieve a sustained upward trend in investment. That pattern qualifies the widespread view of the pro-cyclicality of the Pact.

Empirical evidence suggests that compliance with EU fiscal rules mitigates procyclicality of fiscal policy in the Member States (see Graph 8). 28 First, Member States that met the requirements of the preventive arm of the Pact had less pro-cyclical fiscal effort than others. Second, avoiding high headline deficits appears to reduce the pro-cyclicality of discretionary fiscal policy. Third, keeping public debt at a reasonable level mitigates the procyclical pattern of the fiscal effort. Conversely, having high deficit and debt levels tends to amplify pro-cyclicality. It suggests that the design of the Pact per se is addressing, at least partly, the issue of pro-cyclicality, which is a pervasive feature of fiscal policy (both within and outside the EU).

Graph 8: Cyclicality of the fiscal effort and compliance with EU rules

Note: For more information on the estimation technique used, see European Commission (2019), “Have EU fiscal rules mitigated pro-cyclicality?”, Report on Public Finances in EMU 2018, 121-130.

The EU’s ability to coordinate the aggregate fiscal stance in the euro area remains constrained. Regulation No 473/2013 i established a common budgetary timeline for the euro area, which provides for a check of the budgetary situation and prospects in the euro area on the basis Member States’ budgetary plans and their interactions across the euro area, before the budgets are adopted in national parliaments. While the legal framework allows the Commission to outline measures to reinforce the coordination of budgetary and macroeconomic policy at the euro area level, it does not allow it to enforce the appropriate fiscal stance for the euro area as a whole. While the Pact has established procedures for the correction of high public deficits in terms of nominal thresholds, it cannot compel fiscal policies in support of economic activity. In such a context and in the absence of a central fiscal capacity with stabilisation features, 29 the ability to steer the fiscal stance for the euro

28 European Commission (2019), ‘Fiscal outcomes in the EU in a rules-based framework – new evidence’ in Report on Public Finances in EMU 2018, Institutional Paper 095. 29 Such a fiscal stabilisation function was put forward in the 2015 Five Presidents’ Report on Completing the Economic and Monetary Union, but falls outside the scope of this review.

area as a whole remains constrained, limiting the scope for fiscal policy to contribute to macroeconomic stabilisation at the level of the euro area as a whole in the event of large shocks that are not policy-induced. This is partly the result of a lack of effort in good times, as a result of which the Pact required consolidation in a number of Member States when economic conditions became less benign.

While the issue could be considered secondary at the time when the EU fiscal rules were adopted, its salience has increased in recent years due to the changing economic circumstances. This is due to the growing perception of a risk that monetary policy may face constraints in fulfilling its mandate in an environment of unprecedented low inflation and interest rates, and hence in the event of a severe economic downturn the policy mix be unable

to fully support economic activity. 30 In the absence of a central stabilisation capacity, this is

liable to place strain on the rules-based fiscal framework. Moreover, a negative interest rate, or simply a growth rate differential, risks exacerbating the relative scarcity of safe assets (see box A1 in Annex 1), which would make it even more difficult for the central bank to achieve its inflation target. At the same time, as some recent experiences show, shifts in saving behaviour and preference/aversion for risk may result in sudden reversals of the interest rate/growth rate differential, and with it of the fiscal burden of debt, especially in high-debt Member States. Above all, even in the absence of changes in fundamentals, perceptions of the safety of different national issuers can change substantially, and put in danger the sustainability of their debt. In highly-indebted Member States, this could make the burden of debt servicing unsustainable. That risk is possibly more acute within a monetary union, where governments with very different debt levels share the same currency and the central bank cannot act as government lender of last resort. This confirms the need for rules imposing constraints on debt trajectories.

30 In that context, the European Central Bank has recently called for a more active role for fiscal policy. See for instance President Draghi’s press conference of 12 September 2019: https://www.ecb.europa.eu/press/pressconf/2019/html/ecb.is190912~658eb51d68.en.html

Quality of public finances, including the development of public investment

The quality of public finances is a complex and multi-faceted issue largely outside the reach of the fiscal rules and covered by other EU instruments, such as the countryspecific recommendations under the European Semester. In principle, the composition of public revenue and expenditure is a national prerogative, since the Pact primarily aims to prohibit policies endangering fiscal sustainability, while providing space for fiscal stabilisation. The European Semester, in particular the country-specific recommendations based on Article 121(2) TFEU, and specific EU initiatives have been the instruments to improve the quality of public finance while respecting national prerogatives. For instance, ageing populations are expected to confront most Member States with rising government expenditure over the next several decades. In light of that challenge, the Council addressed recommendations to those Member States to undertake fiscal structural reforms to improve fiscal sustainability and strengthen economic growth potential. Governance issues are also important for the quality of public finances: improving the functioning of national fiscal frameworks, implementing well-managed spending reviews and establishing transparent public procurement procedures can also strengthen the efficiency and effectiveness of public expenditure and the credibility of fiscal policy. A composition of public expenditure that prioritises productive spending helps strengthen potential growth in the longer term 31 and promotes a sustainable and inclusive economy. On the revenue side, shifting taxes away from labour, combating aggressive tax planning and addressing the race to the bottom in corporate taxation can make tax systems fairer and more efficient.

Regarding public investment, the EU has faced a widespread and persistent decline over the last decade (in some cases even below replacement levels keeping the stock of public capital stable). After the crisis, most Member States significantly cut public investment, probably at the expense of future growth. 32 Public investment in the EU declined from a peak of 3.7% of GDP in 2009 to 2.7% of GDP in 2017. If not reversed, the cut in public investment could result in a substantial reduction in the public capital stock. This in turn has a negative impact on potential growth and affects debt dynamics. It has also led to criticism of the appropriateness of fiscal consolidation. While the Pact is in principle neutral regarding the composition of public revenue and expenditure, it may affect public investment decisions in specific cases. It may be politically more expedient during periods of fiscal consolidation to raise certain taxes or to cut public investment, rather than to cut current expenditure.

Meanwhile, fiscal rules are conducive to macroeconomic stability and debt sustainability, which are necessary conditions for public and private investment. Fiscal rules tend to reduce policy volatility, and which makes them beneficial to investment and, more generally, to long-term growth. Moreover, when thoroughly adhered to over the business cycle, fiscal rules ensure that public debt remains sustainable, thus creating fiscal space that can be used for investment. 33 Sound national fiscal rules appear to reduce the

31 Ceparullo and Mourre (forthcoming), ‘How and How much? The growth-friendliness of public spending through the lens?’; European Commission (2016), ‘Insights on the quality of public expenditures’ in Report on Public Finances in EMU 2016, Institutional Paper 045.

32 See for example: European Commission (2019), ‘The 2019 Stability and Convergence programmes: an overview and assessment of the euro area fiscal stance’, Box 2.7. 33 Once achieved, a Member State’s MTO is primarily meant to ensure (i) the long-term sustainability of public finances and (ii) a safety margin towards breaching the 3% of GDP reference value of the Treaty. The definition of the minimum MTO under Regulation No 1466/97 i also refers to room for budgetary manoeuvre and in particular the need for public investment.

negative impact of public debt on public investment (see Annex 1), 34 and empirical research

does not point to the conclusion that fiscal rules have hampered public investment in Member States over the past decades.

The essential role of public investment to deliver public goods and to support sustainable public finances is well recognised in the EU’s fiscal framework. First, the Treaty requires the Commission to take into account government investment spending when assessing whether a Member State has an excessive deficit. Second, the Stability and Growth Pact, both in its title and content recognises the need to consider the overall quality of public finances in terms of the growth-friendliness of the taxation system and public expenditure. Third, the definition of the medium-term budgetary objective refers to room for budgetary

manoeuvre and in particular the need for public investment. 35 This has led to provisions in the

fiscal framework that seek to protect the level of public investment during downturns and to incentivise the implementation of structural reforms, which contribute to sustainable public finances, including by raising potential growth.

However, recent innovations in the Pact have not significantly improved the quality and composition of public finances. This primarily reflects deliberate policy choices in the Member States. In 2015, the Commission operationalised the so-called ‘investment clause’, granting a temporary deviation from the adjustment path under the preventive arm to cater for an increase in investment. 36 However, the investment clause can only be used if GDP growth is negative or if the output gap (which measures the amount of slack in the economy) is below -1.5% of potential GDP. The rationale is to safeguard investment during (very) bad economic times. Only Italy and Finland have so far benefited from the investment clause, both in 2016. However, in neither case was there a substantial positive impact on public investment. Similarly, the structural reform clause has had a rather mixed success in promoting reforms that improve the quality of public finances and/or increase potential growth, with only five of the eligible Member States having made use of the clause. 37

Fulfilling the ambition set by the European Green Deal to achieve a carbon-neutral continent by 2050, will require significant investment. The magnitude of the investment challenges requires mobilising both the public and private sector. Box 3 presents some considerations on green public investment in the context of the EU’s fiscal framework. As announced in the Commission’s Communication on the European Green Deal, the outcome of the debate on how to improve EU fiscal governance will form the basis for any possible future steps including how to treat green investments within the EU fiscal rules, while preserving

safeguards against risks to debt sustainability. 38

34 Mohl and Poissonnier (forthcoming), ‘Do fiscal rules hamper public investment?’; European Commission (2017), ‘Government investment in the EU’ in Report on Public Finances in EMU 2017, Institutional Paper 069. 35 Moreover, euro-area Member States that become subject to an EDP must submit an Economic Partnership Programme describing the policy measures and structural reforms needed to ensure an effective and lasting correction of the excessive deficit. The policy priorities in the programme must also be consistent with the Union’s strategy for growth and jobs.

36 European Commission (2015), ‘Making the best use of the flexibility within the existing rules of the Stability and Growth Pact’, COM(2015)12 i of 13 January 2015.

37 The structural reform clause was applied in the case of Italy in 2016, Latvia, Lithuania and Finland in 2017, and Belgium in 2019. On the assessment of the 2015 Flexibility Communication, see European Commission (2019), ‘Staff Working Document accompanying the Commission Communication on the review of the flexibility under the Stability and Growth Pact’, SWD(2018) 270 final.

38 European Commission (2019), ‘The European Green Deal’, COM(2019) 640 final i.

Box 3: Green investment in the EU’s fiscal framework

Climate change poses important challenges to public finances. Extreme weather events may raise the risks to the sustainability of public finances and the transition to a climate neutral economy will require substantial public and private investment. While climate adaptation investments should reduce the risks and costs of climate change in the longer run, they will also weigh on government finances in the short and medium run.

At present, the EU’s fiscal framework does not include specific provisions to deal with climate-related costs and risks. By promoting sound fiscal positions, the fiscal framework ensures that fiscal policy can fulfil its various roles and support emerging policy priorities, including climate change. Given the scale of the climate change challenge and the risks it poses to the long-term sustainability of public finances, a case can be made that specific provisions be incorporated into the EU’s fiscal framework. However, the public finance implications of climate change are multi-dimensional. At this stage there is no conceptual analytical framework to identify the main links involved and the possible trade-offs. For instance, as both measured fiscal sustainability risks and investment needs rise, trade-offs may arise between catering for increasing investment needs on the one hand and anticipating future costs by raising saving on the other.

There are several dimensions as regards financing green public investment needs. In particular, should those investments compete with current spending priorities and not affect the overall fiscal stance? Or could additional public investment spending come on top of existing commitments and needs and thus raise public debt? While the European Green Deal underlines the key role for the EU budget, it also acknowledges that national budgets will play a significant role in the transition.

There is literature on the use of fiscal rules as a means to preserve and incentivise public investment. Traditional public finance literature often refers to a ‘golden rule’ that allows for the financing of public investment through public debt. Advocates of such a rule argue that it is conducive for intergenerational fairness and economic growth. 39 Specifically, as long as public investment increases the public and/or social capital stock that they will inherit, future generations will benefit from it and so it is justifiable that they should assume the burden of the debt related to such investment. In fact, financing the public capital stock exclusively with taxation would lead to a disproportionate burden for the present generation, with the risk of insufficient provision of public investment. The ‘golden rule’ literature refers to net investment, i.e. gross fixed capital formation minus consumption of fixed capital, as that expenditure item does not affect the net debt of the public sector (i.e. gross debt minus assets).

Literature also provides more sceptical views on preferential treatment of public

investment within fiscal rules. 40 In particular, risks related to such a favourable treatment

have been highlighted, as such rules may distort the allocation of public goods, incentivise

39 See for instance: Balassone, F. and D. Franco (2000): Public investment, the Stability Pact and the ‘Golden Rule’, Fiscal Studies, 21(2), 207-229; Blanchard, O. (2019): Public debt and low interest rates, NBER Working Paper 25621; Buiter, W. (1984), Measuring aspects of fiscal and financial policy, NBER Working Paper 1332; European Fiscal Board (2019), Assessment of EU fiscal rules with a focus on the six and two-pack legislation, 11 September; Truger, A. (2016): The golden rule of public investment: A necessary and sufficient reform of the EU fiscal framework, IMF Working Paper, 168.

40 See for instance: Heinemann, F. (2006): Factor mobility, government debt and the decline in public investment, IEEP, 3, 11-26. Mehrotra, A. and T. Välilä (2006): Public investment in Europe: Evolution and determinants in perspective, Fiscal Studies, 27(4), 443-471. Minea, A. and P. Villieu (2009): Borrowing to finance public investment? The golden rule of public finance reconsidered in an endogeneous growth setting, Fiscal Studies, 30(1), 103-133.

creative accounting, increase the coverage of eligible expenditure over time, make fiscal surveillance more complex, etc. Some empirical work shows that, based on available information, there is only weak evidence that ‘golden rules’ support public investment and that investment is hampered by fiscal rules. 41 Other factors than fiscal rule constraints may lead to low public investment, such as insufficient administrative capacity, poor governance, weak monitoring institutions, etc. Political-economy arguments can also explain why public resources are devoted to spending items other than investment, as it is less visible for voters than for instance social spending.

Ownership and governance of EU fiscal rules

The accumulation of rules, indicators and implementation procedures over time has made the Pact increasingly complex and has harmed predictability. The development of ‘smarter’ and multiple rules has come at the cost of increased complexity. In the current framework, there are multiple rules (for the headline balance, the structural balance and, since the six-pack, for expenditure growth and debt reduction), associated with different indicators for measuring compliance and various clauses allowing for deviations from the required adjustments (see box 2). Procedures for implementing those rules (preventive and corrective arm, ex ante and ex post assessments) are equally complex. While that accumulation of rules and procedures was driven by a legitimate desire to make the framework more adaptable to changing economic conditions, it has led to a blurring of key requirements and objectives, such as avoiding gross policy errors.

The current framework relies heavily on variables that are not directly observable and that are frequently revised, which also hampers ownership. The output gap, which is used to estimate the structural balance, is a key variable underpinning the Pact. It enters the Commission’s assessments at different steps of the surveillance procedures, particularly in the preventive arm. Since potential growth and output gaps are unobservable, they need to be estimated from economic data. 42 This leads to unavoidable uncertainty and revisions of the estimates, which has allowed Member States to increasingly challenge the Commission’s fiscal policy guidance and compliance assessments, for example when it comes to estimating a Member State’s precise structural balance position compared to its MTO. The use of a wider set of macroeconomic indicators for determining the cyclical position of a Member State has had limited success in addressing Member States’ concerns. Overall, the wide use of

unobservable indicators has likely reduced ownership and political buy-in. 43 To some extent, this has been addressed by ‘freezing’ the required adjustment for the following year 44 and by increasing the focus on the expenditure benchmark, which provides more stable and

41 Mohl and Poissonnier (forthcoming), ‘Do fiscal rules hamper public investment?’; European Commission

(2017), ‘Government investment in the EU’ in Report on Public Finances in EMU 2017, Institutional Paper 069.

42 The structural balance also depends upon fiscal semi-elasticities, which are estimated from macroeconomic data and information on tax revenues; see Mourre, Poissonnier and Lausegger (2019), ‘The semi-elasticities underlying the cyclically-adjusted budget balance: an update and further analysis’, Commission ECFIN Discussion Paper 098.

43 However, it should be borne in mind that the main source of revisions are errors in forecasting, which are bound to affect the working of any forward-looking framework.

44 In order to provide ex ante guidance and to ensure predictability of the assessment’s outcome and certainty on what is expected from a Member State, the required adjustment for year t is frozen in the spring of year t-1, on the basis of the estimated output gap and Member State’s distance to its MTO at that time.

operational policy guidance by focussing on budgetary items directly under the control of

governments. 45

While the strengthening of national fiscal frameworks has improved ownership of fiscal discipline, discrepancies have emerged between EU and national fiscal rules, undermining the credibility of both and adding to complexity. The objective of stronger national frameworks was to underpin fiscal policies better at national level in line with EU obligations and to increase ownership of such policies among policy-makers and stakeholders. Consequently, the Directive 2011/85 i and Regulation No 473/2013 i created only broad requirements for national frameworks to accommodate national specificities. On the contrary, the Fiscal Compact went to impose a specific rule modelled on the preventive arm of the Pact (as re-defined by the six-pack). This means that most Member States now have in their national law – at a constitutional level or high in the national legal order – a structural balanced-budget rule that arguably does not take into account interpretative developments in the preventive arm agreed after 2012. This has had two consequences. First, in some Member States, it has arguably hampered rather than improved ownership of fiscal rules, including those at national level. Second, it has given rise to inconsistencies between EU and national rules and encouraged some Member States to arbitrage between them. That situation creates risks of putting in doubt the value of national frameworks, and with it that of the independent fiscal institutions (IFIs) that are required to assess compliance with national rules. Given that the strengthening of national frameworks is meant to improve compliance with EU rules, weakening the former will in turn reflect on the latter as well. Furthermore, differences of views between the independent fiscal institutions and the Commission may weaken the authority of both and give the impression that the national and the EU fiscal rules work at cross-purposes. This raises the question whether complementarity between the EU and national fiscal frameworks have been utilised in an optimal way, particularly now that the national frameworks are well developed.

The process for draft budgetary plans has proved useful for ex ante coordination and dialogue, but has also underlined the difficulty of influencing national fiscal policy in practice. Since Regulation No 473/2013 i, euro-area Member States have been required to submit their budget plans for the coming year to the Commission and the Eurogroup. They are subsequently assessed by the Commission. That process was designed to strengthen ex ante fiscal surveillance, i.e. allowing the Commission to give a view on the likely compliance of the draft annual budget with the EU’s fiscal rules (see Graph 9). This is meant to alleviate the burden of surveillance carried out ex post, which was perceived as more punitive and less efficient as it came too late in the process and offered less scope for timely corrective action. In practice, the process has turned out to be useful for setting up a dialogue between the EU institutions and euro-area Member States, as well as for fostering awareness of among national parliaments and the public of Member States’ obligations under the Pact. At the same time, the focus of the assessment on annual changes in the (unobservable) structural balance or annual expenditure growth imparted a certain narrowness to that dialogue. Only once has the Commission requested a Member State to submit a revised plan (Italy for its 2019 draft budgetary plan). The Commission has instead sent ‘follow-up letters’ in all rounds since 2014. The corrective action taken by Member States in response has been limited. Despite that seemingly limited impact, the mere existence of the process may have encouraged Member States to take the Pact’s requirements into account when preparing their draft budgets.

45 The European Fiscal Board has suggested to replace the debt reduction benchmark and the MTO by an expenditure rule (net of discretionary revenue measures) anchored on debt. Doing so would reduce the role of unobservable variables. See: European Fiscal Board (2019), ‘Assessment of the EU fiscal rules with a focus on the six and two-pack legislation’.

Graph 9: Overall compliance of the draft budgetary plans with the Stability and Growth Pact (number of euro-area Member States)

2014 2015 2016 2017 2018 2019 2020

particularly serious non-compliance at risk of non-compliance

broadly compliant compliant

The Commission and the Council have been reluctant to launch enforcement procedures against Member States and to impose financial sanctions. That reticence has also been stressed by the European Court of Auditors. The Significant Deviation Procedure was 46 introduced as part of the six-pack as a means to enforce the preventive arm of the Pact. However, it has so far only been applied in the clear-cut cases of Romania and Hungary, with limited effectiveness (see box 2). Similarly, the debt reduction benchmark was introduced in 2011 as an attempt to make the Treaty’s debt criterion operational. However, in a context of very low inflation, the benchmark has proved to be more demanding than foreseen and procyclical. Therefore, no EDPs have been opened so far on the basis of the debt criterion alone, in spite of successive breaches of the debt reduction benchmark, e.g. by Italy and Belgium. Indeed, in those highly indebted Member States, the debt reduction benchmark required particularly high fiscal efforts that could actually have aggravated debt dynamics. Therefore, the Commission’s assessment of compliance with the debt criterion appropriately took into account all relevant factors, in line with the provisions of the six-pack, including low inflation and weak growth, and in particular compliance with the requirements of the preventive arm (see box 2). Lastly, the introduction of swifter sanctions and of reverse qualified majority voting for Council decisions on Commission proposals have increased the role of the

Commission. 47 So far, the Commission has refrained from proposing financial sanctions. The

application of sanctions under the Pact is different compared to the field of statistics, where fines involve rather symbolic amounts but carry a credible reputational cost via adequate communication. Political incentives, which would also reinforce the disciplining effect of financial markets, are relatively underdeveloped under the Pact.

46 European Court of Auditors, Is the main objective of the preventive arm of the Stability and Growth Pact

delivered?, Special report No 18/2018; European Court of Auditors, Further improvements needed to ensure effective implementation of the excessive deficit procedure.

47 The European Fiscal Board has argued that the introduction of reverse qualified majority voting has blurred the distinction between the analytical and (growing) political role of the Commission. It recommends a better separation between the economic analysis performed within the Commission and the policy decisions taken on the basis of that analysis. In addition, it proposes to remove the reverse qualified majority voting principle. See: European Fiscal Board (2019), ‘Assessment of the EU fiscal rules with a focus on the six and two-pack legislation’.

The medium-term orientation of fiscal policy, including its multi-annual ‘track record’, is insufficiently taken into account. In principle, the Pact has a strong medium-term focus, for instance with a central role of Stability and Convergence Programmes, focus on a ‘closeto-balance or in surplus’ budgetary position in the medium term, and multi-annual targets in the EDP. In practice, however, the six- and two-packs have introduced a strong focus on an assessment of annual targets. As a result, the focus on medium-term performance (both backward- and forward-looking) has weakened in relative terms, especially given that the implementation of the debt reduction benchmark has so far relied on an assessment of compliance with the annual requirements of the preventive arm. The emphasis on annual fiscal adjustments has made it more difficult to differentiate between Member States that have markedly different overall fiscal positions and sustainability risks. It has also been argued that a more medium-term focus could help simplify the framework and be easier to reconcile with the stabilisation objective of fiscal policy. However, such an approach also has drawbacks. In particular, it could lead to procrastination or a lack of political ownership. From that perspective, past performance (not least in terms of public debt accumulation) is a key indicator.

National fiscal frameworks

There are indications that the strengthened national fiscal frameworks have contributed to the objectives of EU fiscal rules. While it is difficult to isolate the effects of national fiscal governance on budgetary outcomes in an unambiguous way, the fiscal frameworks have been developing in a manner that has been supportive of the objective of EU fiscal rules. The Commission analysis suggests that national fiscal rules, particularly when well designed, have contributed to sound budgetary policy in the Member States. Domestic rules with a strong legal basis, binding targets, independent monitoring, a well-defined correction mechanism and high media visibility appear to be effective in reducing budget deficits. A similar role, with potentially even larger effect, can be attributed to the Member States’ MTBFs developed by

the Member States. 48 At the same time, practices concerning the design and role of mediumterm

planning in national budgetary processes differ significantly across Member States, with significant impact on the effectiveness of MTBFs. The analysis also shows that national expenditure rules have had a cyclical stabilisation effect, in addition to the stability-increasing

effect of rules overall. 49 Furthermore, there are indications that the development of national

frameworks has strengthened national ownership of fiscal discipline and that stakeholders generally consider strong national frameworks as an effective and efficient tool to promote compliance with EU fiscal rules.

The strengthening of national fiscal frameworks has been a positive development that should be seen in a broader context. That strengthening has supported Member States’ domestic institutional frameworks, thus helping to improve the overall quality of public administrations. In particular, the establishment of independent fiscal institutions in all but one Member State should be viewed as a key institutional development. Indeed, while they differ in terms of scope, competences and experience, together they are playing an increasingly important role in fiscal discussions at national and EU levels. It is also worth noting that the EU’s efforts to create robust national fiscal frameworks have been part of a worldwide trend, with wide-ranging reforms in this area also taking place outside the EU.

48 European Commission (2019), “Has ownership of the EU’s fiscal rules been strengthened by national fiscal frameworks?” in Report on Public Finances in EMU 2018, Institutional Paper 095 49 European Commission (forthcoming), “Performance of spending rules at the EU and Member States’ level” in Report on Public Finances in EMU 2019, Institutional Paper forthcoming.

Nevertheless, there remains scope to further improve national fiscal frameworks. The state of development of fiscal frameworks differs considerably across Member States. This is linked, on the one hand, to different constitutional settings, preferences and experience among Member States and, on the other, to the rather general nature of EU requirements for most of the components of fiscal frameworks (which leaves ample room for a variety of approaches at country level). Likewise, the implementation of national provisions, in particular compliance with national fiscal rules, differs across Member States. In this context, the European Court of Auditors has recommended strengthening EU requirements for national fiscal frameworks and

improving their monitoring. 50,51 At the same time, as illustrated by experience of the Fiscal

Compact, it is important to bear in mind the risk of undermining national ownership and effectiveness if requirements are too specific or do not take national specificities sufficiently into account. Furthermore, while the design of fiscal frameworks matters for their effectiveness, steadfast implementation is key. The promotion of good budgetary practices, such as 'green budgeting', could also help to improve the quality of public finances.

The interaction between the national fiscal frameworks and the EU fiscal rules is key for the effectiveness of both. As the granularity of EU fiscal rules has increased in parallel with strengthened requirements for national frameworks, there has been a growing tension between ensuring national ownership and maintaining horizontal consistency. This has been more apparent in the preventive arm of the SGP, as the Fiscal Compact led most Member States to introduce in their national law substantive rules similar to those of the preventive arm. Better compliance with national fiscal rules and stronger national fiscal frameworks overall are a means of reducing the intensity of fiscal surveillance at EU level, strengthen national ownership of fiscal prudence and improve compliance with EU rules as well. Exchanges of best practices among Member States in terms of the design and implementation of their frameworks, and cooperation between the IFIs and the Commission on the design and implementation of the framework, help to improve the interaction of national frameworks with EU fiscal requirements.

2.5 Conclusions

The reformed EU fiscal framework has helped to make public finances in the EU more sound. Public deficits and debt levels have been falling, and all Member States have exited the excessive deficit procedure. 13 Member States have reached their medium-term budgetary objectives, providing them with fiscal buffers for the next downturn. The process for draft budgetary plans has proved useful, allowing for an early identification of fiscal risks and a degree of ex ante coordination. The reinforcement of national fiscal frameworks has contributed to better outcomes.

At the same time, the reformed Pact has not led to sufficient fiscal adjustment in highly indebted Member States in good times. Despite the return of economic growth, public debt levels remain high in several Member States, which are also far from their medium-term budgetary objectives. This implies that those Member States are more exposed to the risk of being forced to adjust in bad times in a pro-cyclical manner. Fiscal policy at Member State level has frequently been pro-cyclical, both in good and in bad times. In addition, the ability to deliver an appropriate fiscal stance for the euro area as a whole, with the appropriate

50 An attempt in that direction was made in December 2017, when the Commission presented a proposal for a

Directive incorporating the Fiscal Compact into EU law, which however has not received sufficient support from legislators.

51 European Court of Auditors, “EU requirements for national budgetary frameworks: need to further strengthen them and to better monitor their application”, Special Report No 22/2019.

differentiated fiscal effort among Member States, is hampered by the lack of prudent policies in good times. It remains inherently constrained as long as it rests exclusively on coordination of national fiscal policies in the absence of a central stabilisation capacity.

The assessment of fiscal performance above points to a number of shortcomings in the EU fiscal framework. The framework has become increasingly complex, reducing predictability and undermining political ownership among Member States. It consists of multiple rules, associated with different indicators for measuring compliance, and contains various clauses allowing for deviations from the requirements, each of them with their own eligibility criteria. It does not sufficiently differentiate between Member States that have markedly different fiscal positions and debt sustainability risks, diluting the focus away from the prevention and correction of gross policy errors. It has not fostered an improvement in the quality of public finances, nor has it been able to support investment despite the limited flexibility that was introduced to that effect, although it cannot be concluded that it has hampered investment. The interplay between EU and national fiscal frameworks could be improved to ensure that they work better in tandem.

An effective framework needs to ensure the sustainability of public debt where it is most necessary, while allowing for macro-economic stabilisation in both good and bad times. Fiscal policy guidance supports Member States in ensuring long-term sustainability of public finances and pursuing counter-cyclical fiscal policies to contribute to a better macroeconomic stabilisation in both good and bad times. Sufficient fiscal effort and debt reduction in good economic times helps to create the space to use fiscal policy in bad times. Medium-term fiscal policy planning and an appropriate policy anchor help in that regard. It needs to be discussed what the appropriate role of the EU surveillance framework is in helping to protect public investment in consolidation periods, promoting a growth-friendly composition of public finances and sustaining adequate levels of investment. In particular, it raises the question of the extent to which the fiscal framework could provide flexibility for the public investments needed to meet the broader ambition for a green and digital transformation of Europe, in line with the objective of the European Green Deal. Lastly, the deployment of fiscal surveillance instruments could be considered simultaneously with that of other instruments, such as the macro-economic imbalances procedure, to address low potential growth.

A simpler framework and implementation could help to increase ownership, improve communication, and reduce the political costs of enforcement and compliance. Whereas the current fiscal surveillance framework has included elements of flexibility and discretion through a complex set of provisions adopted against a background of lack of trust among key stakeholders, an effective application of economic judgement within a rules-based framework needs to be done in a manner which is objective and transparent. This includes, for example, a need to consider whether a clear focus on gross policy errors as set out in the Treaty, based on clearly defined objectives and operational policy targets, could contribute to an effective implementation of the fiscal framework. Indeed, surveillance should be commensurate to the gravity of the situation, with a stronger focus on the most pressing cases and less-intrusive procedures where overall risks are low. Therefore, it could also be considered whether the surveillance framework, in order to be effective, should focus more on ‘identifying gross errors’ 52 , i.e. on Member States whose policy strategy puts public debt on a potentially unsustainable trajectory. Lastly, it has to be considered whether a stronger role for national fiscal frameworks, in particular independent fiscal institutions, would contribute to better compliance with EU fiscal rules and improve ownership of the framework at the same time.

52 Cf. Article 126(2) TFEU.

3.

M ACROECONOMIC I MBALANCE P ROCEDURE

3.1 Objectives of the Macroeconomic Imbalance Procedure

The Macroeconomic Imbalance Procedure (MIP) was established after the economic and financial crisis with the aim of supporting macro-financial stability. The 2008-2009 crisis was accompanied by a general reappraisal of risks in financial markets and acted as a trigger for sudden stops and reversals in current account financing and for the bursting of asset bubbles. Macro-financial and macro-structural aspects driving the accumulation of both external (e.g., large current account imbalances) and internal imbalances (e.g., excess debt accumulation or the building up of housing bubbles) revealed themselves as key factors in triggering balance of payment crises and debt crises, which spilled over to other countries and in some cases required financial assistance to Member States. Therefore, the MIP was introduced with the aim to identify, prevent and address the emergence of potentially harmful macroeconomic imbalances that could adversely affect economic stability in a particular Member State, the euro area, or the Union as a whole. Regulation No 1176/2011 i therefore considers that, when assessing macroeconomic imbalances, account should be taken of their severity and their potential negative economic and financial spillover effects which aggravate the vulnerability of the Union economy and are a threat to the smooth functioning of the economic and monetary union. In particular, given vulnerabilities and the magnitude of the adjustment required, the need for policy action was mentioned to be particularly pressing in Member States showing persistently large current-account deficits and competitiveness losses. Furthermore, recital 17 to Regulation No 1176/2011 i notes that, in Member States that accumulate large current account surpluses, policies should aim to identify and implement measures that help strengthen their domestic demand and growth potential. With a view to provide an integrated response to economic and social challenges, the MIP was integrated into the European Semester.

The MIP legal framework consists of Regulation No 1176/2011 i on the prevention and correction of macroeconomic imbalances, and Regulation No 1174/2011 i on enforcement measures to correct excessive macroeconomic imbalances. Regulation No 1176/2011 i outlines the procedure, while Regulation No 1174/2011 i details an enforcement mechanism, including pecuniary sanctions for euro area Member States. The legal basis of Regulation No 1176/2011 i is the Treaty article dealing with economic policy coordination (Article 121 of the Treaty on the Functioning of the European Union), while the legal basis of Regulation No 1174/2011 i is Article 136 TFEU, which allows for economic policy measures specific to the euro area.

Regulation (EU) No 1176/2011 i sets a number of steps to identify imbalances in Member States, to recommend policy action, and to ensure adequate monitoring and enforcement. All Member States 53 are screened by means of a set of indicators known as the MIP scoreboard. The Commission’s economic reading of the scoreboard is complemented by additional relevant information and published in an annual Alert Mechanism Report. The Alert Mechanism Report indicates the Member States for which the Commission will carry out in-depth reviews on the basis of prima-facie evidence suggesting the possible presence of imbalances. Based on those in-depths reviews, the Commission adopts conclusions on the existence and severity of imbalances. A difference is made between “imbalances” and

53 According to Regulation (EU) No 472/2013 i on the strengthening of economic and budgetary surveillance of Member States in the euro area experiencing or threatened with serious difficulties with respect to their financial stability, MIP surveillance does not apply to euro area Member States subject to a macroeconomic adjustment programme for the duration of that programme.

“excessive imbalances” depending on the severity of the imbalances, both implying possible recommendations by the Council upon a Commission recommendation. Surveillance of Member States identified with imbalances takes place in the framework of the MIP preventive action. The identification of excessive imbalances implies a stronger surveillance process, possibly leading to corrective action under the MIP – the Excessive Imbalance Procedure (EIP) – which also can give rise to sanctions for the euro area Member States. The EIP has not been launched to date.

The legal framework has provided the Commission and the Council a degree of discretion. While being relatively detailed on the steps of the annual cycle, Regulation No 1176/2011 i does not specify which actions are to be taken in each contingency. This reflects the challenge of legislating on a broad, complex, evolving, and multi-faceted matter such as macroeconomic imbalances. Against this background, the Commission has exercised its discretion to apply the MIP in light of accumulated experience and in response to evolving economic conditions and thus the framework has developed over time. While the format of the surveillance process stabilised however over the first years of its application, the analytical framework has been gradually enriched. The gradual development of the approach to MIP surveillance is illustrated in the MIP Compendium. 54 In 2018, following its audit on the implementation of the MIP, the European Court of Auditors addressed recommendations to the Commission. The Commission accepted a number of those recommendations and proposed to take them into account in this review. 55

The remainder of this chapter is organised as follows. The next section reports on the actual application of the main steps of the procedure. Section 3.3 focuses on the actual effectiveness of the MIP in terms of direct impact on policies and in supporting the broader policy context, and looks into relevant economic developments. Section 3.4 outlines the main achievements and challenges with the current framework.

3.2 Implementation of the MIP

3.2.1 The Alert Mechanism Report

The Alert Mechanism Report is the initial step in the annual MIP cycle. The Alert Mechanism Report serves to select Member States for which prima-facie evidence indicates the possible presence of macroeconomic imbalances. This subset of Member States is thereafter subject to an in-depth review.

The analysis takes into account a scoreboard of indicators and other additional information. Regulation No 1176/2011 i provides that “the scoreboard shall be used as a tool to facilitate early identification and monitoring of imbalances” (Article 4(1)). Scoreboard indicators are assessed, inter-alia, against their respective thresholds. The regulation also specifies that, in the reading of the scoreboard, the Commission shall use economic judgement (Article 3(2)). The assessment also draws on other relevant economic and financial indicators not included in the scoreboard (Article 3(2)). In practice, the Alert Mechanism Report has taken the form of an annual Commission Communication to the European Parliament, the Council, the European Central Bank, and the European Economic and Social Committee.

54 European Commission (2016), “The Macroeconomic Imbalance Procedure. Rationale, Process, Application: a Compendium”, European Commission, Institutional Paper 39, November 2016. 55 European Court of Auditors (2018), "Audit of the Macroeconomic Imbalance Procedure", Special Report 03.

The scoreboard has a number of purposes. Although the MIP scoreboard is not intended to predict the occurrence of crises, it allows for tracking the evolution of indicators suggestive of the building up of risks. The Alert Mechanism Report also helps tracking the evolution of existing imbalances. The MIP came into force in a particular context, when most Member States were still running large current account imbalances and were accumulating large stocks of net foreign liabilities and of public and private debt. The Alert Mechanism Report has proved useful in tracking the unwinding of those imbalances in a comparative perspective. The scoreboard has also contributed to define the scope of MIP surveillance, by providing an anchor for its remit, and it has facilitated the communication on the surveillance of macroeconomic imbalances.

The MIP scoreboard currently includes 14 headline indicators and 28 auxiliary indicators. All indicators are based on ‘hard’ and published data to ensure the required high statistical quality. 56 The headline indicators have indicative thresholds to serve as alert levels in line with the requirements of Regulation No 1176/2011 i (Article 4(4)). 57 The headline scoreboard has been expanded to incorporate indicators on financial conditions and employment developments, in 2012 and 2015 respectively. As outlined in the 2016 Alert Mechanism Report, which first included additional employment indicators in the headline scoreboard, the inclusion of these variables into the scoreboard does not have legal implications nor changes the focus of the MIP and consequently cannot trigger further steps under the MIP. Auxiliary indicators have also been adjusted over time.

The scoreboard has been found to perform satisfactorily in capturing macroeconomic risks. The scoreboard has sometimes been criticised to be backward looking. Criticisms in this respect were notably raised by the European Court of Auditors. Data included in the scoreboard comes with lags because, in line with the principle enshrined in Regulation No 1176/2011 i, it should include only variables of high quality statistical data, not forecasts. The fact that some flow variables (e.g., current account balances, and unit labour costs growth) are averaged over past years introduces a backward-looking bias, but also helps avoiding that the scoreboard values for these volatile variables are fully driven by temporary developments. At the same time, analyses show that the scoreboard would have warned about the building of risks ahead of the economic and financial crisis. 58

The scoreboard has been fine-tuned with timely, targeted, and parsimonious revisions. Regulation No 1176/2011 i calls on the Commission to “assess on a regular basis the appropriateness of the scoreboard, including the composition of indicators, the thresholds set and the methodology used, and it shall adjust or modify them where necessary” (Article 4(7)). The Commission has revised the scoreboard to factor in improved availability of published

56 These indicators are mainly compiled by Eurostat from the data transmitted by Member States, following Union legislation. Some indicators are produced in the Member States by National Central Banks in their capacity as members of the European System of Central Banks (ESCB). Eurostat prepares a Statistical Annex to the Alert Mechanism Report scoreboard and auxiliary indicators. 57 The MIP Compendium (section 3.3.1.1.) elaborates on the indicators in the scoreboard, the rationale for their choice, their statistical definitions, data sources and the relevant thresholds and the criteria for their determination. See also European Commission, "Scoreboard for the Surveillance of Macroeconomic Imbalances" – European Economy, Occasional Paper 92, 2012; and "Scoreboard for the surveillance of macroeconomic imbalances: envisaged initial design", – SEC (2011) 1361 final, 8.11.2011. 58 E.g., C. Kamps, R. De Stefani, N. Leiner-Killinger, R. Rüffer and D. Sondermann (2014), "The identification of fiscal and macroeconomic imbalances unexploited synergies under the strengthened EU governance framework", ECB Occasional paper Series No 157, November 2014; B. Pierluigi, and D. Sondermann (2018), Macroeconomic imbalances in the euro area: where do we stand?", ECB Occasional paper Series No 211, June 2018.

statistics relevant for macroeconomic stability and to add variables that help qualifying imbalances and the adjustment process.

The Commission has fostered a regular dialogue on the scoreboard with the European Parliament and the Council. While Regulation No 1176/2011 i leaves revisions of the scoreboard to the discretion of the Commission, it calls for interaction with the European Parliament and the Council on possible adjustments to the scoreboard. The Commission has maintained a regular dialogue with Member States on the adequacy of the scoreboard and on possible improvements to it, mainly through the Economic Policy Committee (EPC) and its LIME working group. Likewise, the Commission has consulted the European Parliament on changes to the scoreboard. The Parliament and the Council have generally been supportive of the revisions.

The scoreboard reading has not been mechanical. This is in line with the spirit of the regulation. The overall number of indicators beyond the respective thresholds and the severity of those breaches have been taken into account as a starting point. However, that information has always been put in context and there has been no automaticity in the conclusions of the Alert Mechanism Reports. Likewise, a low number of indicators beyond the thresholds has not always precluded selection for an in-depth review if risks appeared sufficiently worrisome. The MIP Compendium reports a number of principles that have been followed in the economic reading of the scoreboard. 59

The Commission’s analysis has made use of information beyond the scoreboard indicators. That covers forecast figures, existing analyses, including analytical findings under

previous MIP cycles, including in-depth reviews, and other relevant research. 60 At the same

time, the country-specific analyses have accounted for relevant additional information, including policy settings or governance issues, which have macroeconomic relevance and thereby need to be taken into account when selecting a Member State for an in-depth review. In addition, the Alert Mechanism Report analyses have benefited from the use of a common set of analytical tools with country-specific benchmarks, which allow to gauge risks and to assess adjustment needs. These tools have been developed in cooperation with the EPC and its LIME working group and are described in section 3.2.2. 61

The Commission has followed the prudential practice of selecting for in-depth reviews the Member States that were already identified with imbalances in the previous annual cycle. That practice has allowed to continuously monitor how existing imbalances evolved, whether they were corrected or aggravated, with a view to timely adapt the categorisation of imbalances and the modulation of MIP surveillance. The approach to the Alert Mechanism Report has left room for appropriate economic judgment and the advisable flexibility. Such an approach has allowed considering also risks that were not fully or not yet reflected in the

scoreboard. 62

59 See, in particular, section 3.3.1.2 of the MIP Compendium.

60 This may include regular or ad-hoc reports and analyses by international organisations and institutions, such as ECB, ESRB, OECD, IMF, central banks, or academia.

61 Although the scoreboard has not been used mechanistically to identify imbalances, it turns out that the number of readings beyond the thresholds has tended to be higher in Member States with imbalances and especially excessive imbalances (see MIP Compendium, section 4.3.1).

62 For instance, the selection of Romania for an IDR as in the AMR 2019 was partly motivated by financial governance issues.

The selection of Member States for an in-depth review has been broadly consensual. The Council has always endorsed the selection of Member States for an in-depth review proposed by the Commission. 63

3.2.2 In-depth reviews and classification of imbalances

Identification of imbalances

Regulation No 1176/2011 i defines imbalances in a general and comprehensive manner, leaving ample room for economic interpretation. As compared with the rules-based surveillance framework of the Stability and Growth Pact, MIP surveillance is intentionally not driven by automatic triggers. Drivers of macro-financial instability can be diverse and multidimensional phenomena that need to be assessed taking into account also country-specific features, including those linked to the adjustment capacity of the economy. Therefore, the Commission carries out in-depth reviews that provide the basis for the identification of imbalances.

Regulation No 1176/2011 i distinguishes between “imbalances” and “excessive imbalances”. That distinction depends only on the severity of the imbalances. The latter implies a stronger activation of surveillance, possibly leading to the launch of corrective action, i.e. the EIP, which also can five rise to possible sanctions for euro area Member States. The severity of macroeconomic imbalances is determined not only by considerations relating to the Member State concerned, but also critically by the existence or risk of negative spillovers to other euro area Member States. In particular, the risks of large negative spillovers are considerably larger for current account deficits, and are especially worrying when combined with sustained losses of cost competitiveness and export market shares. At the same time, large and sustained current account surpluses, while not creating immediate risks to national economic stability, affect the smooth functioning of the euro area.

There have been calls for more transparency and predictability. Such calls should be seen in light of the substantial role for the Commission in exercising economic judgement in the implementation of the MIP. In its conclusions on the in-depth reviews, the Council has voiced the importance of transparency for what concerns the reasons underlying the identification and classification of macroeconomic imbalances and the activation of MIP surveillance,

notably the launch of the EIP once excessive imbalances are identified. 64 Policy makers and

experts have criticised that the scope of the MIP was sometimes unclear and, for instance, that the MIP should focus on key issues of macroeconomic and cross-border relevance. 65 The Five Presidents' Report recommends an effective and transparent implementation of the MIP and using the MIP at full potential, including the use of the EIP "as soon as excessive imbalances

63 See also the ECOFIN Council Conclusions on AMR 2019: https://data.consilium.europa.eu/doc/document/ST- 5603-2019-INIT/en/pdf

64 See, e.g., Council conclusions on the 2015 IDRs: https://www.consilium.europa.eu/en/press/press href="https://www.consilium.europa.eu/en/press/press-releases/2016/05/25/conclusions-in-depth-review-2015-csr/">releases/2016/05/25/conclusions-in-depth-review-2015-csr/

65 K. Efstathiou and G. B. Wolff (2018), "Is the European Semester effective and useful?", Bruegel Policy Contribution, Issue 09, June 2018. A. Bénassy-Quéré (2017), "Making the European semester more efficient", ECB Forum on Central Banking "Investment and growth in advanced economies: Conference proceedings".

are identified". 66 The Commission in its October 2015 Communication took steps to enhance the transparency of the MIP (see below). 67

The MIP Compendium reports principles followed by the Commission in the identification and assessment of imbalances.

• The Compendium explains MIP-relevant macroeconomic imbalances as unsustainable

trends (e.g., buoyant house price growth that could lead to the burst of a housing bubble with implications for the banking sector) or vulnerabilities (e.g., high private debt, which makes the economy more vulnerable to shocks) that could have harmful implications for macroeconomic stability for the country itself, the euro area or the

EU.

• In assessing imbalances, the Commission therefore takes account of potential spillovers

to other Member States, the euro area, or the EU as a whole.

• The Commission highlights the sources of imbalances and also adjustment issues, i.e.

factors that hamper a successful correction of existing imbalances. Adjustment issues also include challenges that have resulted from the unwinding of existing imbalances (e.g., the impact of demand compression following current account reversals on the

labour market, or the impact of deleveraging on investment).

The identification and assessment of imbalances has taken into account the gravity of the challenges, their evolution and policy responses. The Commission assessment has been based on three main criteria:

• The gravity of imbalances, i.e., the likely magnitude of the risks involved;

• Their evolution, to distinguish if the identified imbalances exhibit any tendency to

further aggravate or to correct;

• The policy response, with a view to assess to what extent already enacted or planned

policies are adequately tackling the identified issues.

Hence, in its assessment of imbalances the Commission has taken a forward-looking approach, factoring in also the extent to which policies contribute to the correction of risks.

Challenges linked to low productivity or potential growth have been considered among sources of imbalances when relevant for macroeconomic stability. Over time, the Commission has clarified the principles for deciding which issues are dealt with under MIP surveillance and why. These clarifications are relevant especially for those issues that are already dealt with in other surveillance streams. Notably, this concerns government debt, which is the object of surveillance under the Stability and Growth Pact, and economic growth, being the subject of the Commission’s growth strategy organised under the European Semester. The MIP Compendium has clarified that MIP surveillance is not aimed at directly enhancing the growth performance of Member States. In fact, there are instances where buoyant growth comes together with harmful trends (e.g., unsustainably widening external imbalances). However, in some cases, low potential GDP growth can pose macroeconomic risks if left unaddressed: this is especially the case for Member States burdened by high debt levels, as an insufficient growth potential also implies lack of the resources needed to finance

66 Juncker, J.-C. in close cooperation with D. Tusk, J. Dijsselbloem, M. Draghi and M. Schulz (2015), "Completing Europe's Economic and Monetary Union".

67 Communication from the Commission to the European Parliament, the Council and the European Central Bank "On Steps towards completing Economic and Monetary Union", COM(2015) 600 final i. https://ec.europa.eu/transparency/regdoc/rep/1/2015/EN/1-2015-600-EN-F1-1.PDF

high stocks of debt going forward. Government debt is object of MIP surveillance to the extent that it contributes to overall macroeconomic imbalances and has potentially harmful implications for macroeconomic stability. 68

The Commission’s identification and assessment of imbalances have concerned the overall situation of the economy. The assessment has not referred to the identified issues separately, but to the overall combination of factors contributing to risks to macroeconomic stability. Thus, Commission conclusions on imbalances have concerned whether imbalances exist in a certain Member State and whether they are excessive or not.

Commission analysis in the in-depth reviews

The in-depth reviews identify the sources of imbalances that need to be monitored. While the assessment of the existence and severity of imbalances concern the overall economy, the main sources of imbalances have been spelled out. The introduction of the MIP Assessment Matrix in 2016 has made that identification more transparent and came in the spirit of enhanced transparency announced in the October 2015 Commission Communication (see next section). 69

The Commission has developed a broad spectrum of analytical tools in its MIP analysis. Those tools help qualifying selected developments and assessing their drivers, going beyond the one-size-fits-all approach of the scoreboard thresholds and taking into account country specificities. The tools are mainly used to assess the evolution of a number of variables, including their sustainability, also with a view to inferring about underlying vulnerabilities and adjustment needs. Areas covered currently include:

• External sustainability: cyclically-adjusted current account balances, current account

balances based on the economy’s fundamentals ("norms"), and balances taking into account the implications for the net international investment positions (NIIP); NIIP levels predicted on the basis of fundamentals and levels suggested by prudential

concerns;

• Private debt (households and non-financial corporations): levels predicted on the basis

of fundamentals and levels suggested by prudential concerns and subsequent implications for estimated deleveraging needs;

• Housing prices: benchmarks are used to infer about possible over/undervaluation of

housing;

• Spill-overs, with general equilibrium modelling approaches being followed for the five largest euro area Member States. 70

The use of a common set of analytical tools allows a deeper analysis while maintaining consistency across Member States. Before being applied, the tools are disclosed to and discussed in Council Committees mainly through the LIME working group of the EPC. The tools follow state-of-the-art methodologies and are regularly updated.

68 This was corroborated by the Council on its conclusions on the original design of the scoreboard: https://register.consilium.europa.eu/doc/srv?l=EN&f=ST%2015781%202011%20REV%202 69 See footnote67.

70 Government debt sustainability analysis developed in the context of the surveillance related to the Stability and Growth Pact is also used when relevant for the MIP analysis.

Classification of imbalances

The Commission streamlined the categorisation of imbalances in 2016. Until 2015, up to six categories of imbalances were used not only to reflect the three-category framework stipulated by Regulation No 1176/2011 i ("no imbalances", "imbalances", "excessive imbalances"), but also to better reflect different needs of policy action and monitoring. Amidst efforts to enhance transparency, clarity, and stability, in October 2015, the Commission committed to stabilise the categorisation of imbalances. 71 In spring 2016, the MIP categories were streamlined to four: "no imbalances", "imbalances", "excessive imbalances", and "excessive imbalances with corrective action" (i.e., the opening of the EIP), which have been stable since then (see MIP Compendium, section 4.3.3).

The format of the in-depth reviews has changed over time. Until 2014, the in-depth reviews were self-standing documents published as Commission Staff Working Documents. While that allowed for a fully-fledged assessment, it also partially overlapped with other documents in the European Semester. Therefore, in 2015, they became part of the Commission's Country Reports, which are also Staff Working Documents, with the aim of further integrating surveillance and ensuring consistency of diagnoses. As of 2016, to safeguard the MIP and to identify the in-depth review in the Country Reports, the main elements of the analysis of the in-depth reviews have been summarised in a separate chapter of the Country Report. Also as of 2016, the findings and rationale have been summarised in a synthetic table to improve transparency. The so-called MIP Assessment Matrix lists the sources of imbalances and adjustment issues, reports the main findings regarding the gravity of identified challenges, the evolution of risks, and the relevant policy responses taken by the authorities and remaining policy gaps. This concise repository of information has enhanced transparency and communication across countries and over time.

The conclusions of the in-depth reviews have been reported in a Commission Communication. The in-depth reviews have presented the technical assessment, while the conclusions on the existence and severity of imbalances have been reported in an accompanying Commission Communication. That practice ensures consistency on the classification across Member States and allows taking better account of systemic aspects, including spill-overs and the overall context.

Findings from the in-depth reviews

The number of Member States identified with imbalances or excessive imbalances has evolved as result of the changing severity of risks. That in turn has been owed to the evolution of the economic situation or the policy action taken by the Member States concerned to tackle those risks.

• On the one hand, the reduction of macroeconomic risks led more Member States into

the MIP because some of them transitioned into MIP surveillance once their macroeconomic adjustment programmes ended between 2013 and 2016 (Ireland, Portugal, Romania, Cyprus) and were identified as experiencing imbalances (or excessive imbalances). The same applied to Greece in 2019. After its accession to the

Union in 2013, Croatia was also identified with excessive imbalances.

71 See footnote 67, and also Communication from the Commission to the European Parliament, the Council, the European Central Bank and the Eurogroup, "2016 European Semester: An assessment of progress on structural reforms, prevention and correction of macroeconomic imbalances, and results of in-depth reviews under Regulation (EU) No 1176/2011 i", Strasbourg, 8.3.2016 COM(2016) 95 final i, http://ec.europa.eu/europe2020/pdf/csr2016/cr2016_comm_en.pdf .

• On the other hand, notably after the economic recovery became more widespread,

reduced macroeconomic risks coincided with the exit or de-escalation of Member States already under the MIP. This was the case in 2016 for four Member States that were found to be no longer experiencing imbalances (Belgium, Hungary, Romania,

and the United Kingdom).

• A number of Member States exited or were de-escalated under the MIP surveillance

also on the basis of policies they took that improved the forward-looking assessment of risks.

• As some imbalances tend to be persistent phenomena, a number of Member States

have continued to be marked by imbalances or excessive imbalances.

Graph 10: Macroeconomic Imbalances

2013 2014 2015 2016 2017 2018 2019

Excessive Imbalances Imbalances IDR but no imbalances No IDR MIP not applicable

Note: Categories used in 2012 are not directly comparable to those used in subsequent years (see MIP Compendium, section 4.3.3) and thereby they are not included in this figure. Source: European Commission

There has been broad consensus on the findings from the in-depth reviews. As with the Alert Mechanism Report, the in-depth reviews are discussed by Member States at the Council preparatory committees, including the EPC and its LIME working group. In addition, the proposed categorisation of imbalances is subject to a dialogue with Member States, with the ECOFIN Council eventually adopting conclusions. While borderlines cases are discussed, the Council has regularly endorsed the Commission categorisation of imbalances. Nonetheless, some doubts on the Commission decisions regarding the classification of imbalances have been expressed publicly. For instance, the European Court of Auditors in its report on the MIP finds that the evidence contained in the 2016 in-depth review for France does not obviously signal an aggravation of risks requiring the escalation from imbalances to excessive imbalances as decided by the Commission in 2016.

3.2.3 Recommendation, monitoring, and enforcement

Policy recommendations

Member States experiencing imbalances or excessive imbalances have received more policy recommendations than other Member States. In particular, this holds for Member States identified with excessive imbalances. That difference became even stronger after the

streamlining of the country-specific recommendations (CSRs) in 2015 (see Graph 11). 72 The

higher number of policy recommendations for those Member States is justified by the additional challenges they face.

Graph 11: Average number of CSR subparts per Member State per MIP status

Non-MIP

25 MIP imbalances

a rt 20 MIP excessive imbalances

b p

s u

f C SR 15

b er 10

N u

2013-2014 2015-2019

Source: European Commission

The MIP-relevant recommendations cover the most pressing challenges. The relevant policy challenges in Member States are identified on the basis of the in-depth review analysis. Those recommendations have reflected the policy efforts needed to address imbalances and foster macroeconomic stability. Compared with non-MIP Member States, MIP Member States received more recommendations on macro-financial and macro-structural areas, in particular in areas relating to the financial sector (see also MIP Compendium, Graph 4.5). Recommendations that are relevant from an MIP perspective have been indicated as such in the relevant Commission and Council recommendations. The information on the MIP- relevance of the CSRs has been included in the recitals in the preamble of the Commission recommendation for a Council recommendation and of the Council recommendations by listing which CSRs are issued under Article 6 of Regulation No 1176/2011 i. Since a CSR can incorporate recommendations in different policy fields and as a CSR is "tagged" as MIP relevant in its entirety, it could be the case that MIP-relevant CSRs also include some policy measures not directly aimed at addressing macroeconomic imbalances.

For Member States experiencing imbalances or excessive imbalances, most of the CSRs have been linked to those imbalances. CSRs under the MIP have always been a major share of the total number of CSRs, especially for Member States with excessive imbalances (see Graph 12, left panel).² Almost all recommendations relating to the financial sector have been indicated as MIP relevant (see Graph 12, right panel). In other policy areas less directly linked to macroeconomic stability (e.g., those concerning social issues), a relatively lower share of CSRs were considered MIP-relevant. The share of MIP-relevant CSRs for Member States with imbalances increased with the streamlining of CSRs around 2015, and it receded somewhat in 2019 reflecting an increasing number of CSRs subparts for all Member States.

72 To allow a finer degree of detail, data on policy recommendations presented here refers always to CSR subparts unless stated otherwise.

The fact that a majority of CSRs were considered as MIP-relevant for Member States under MIP surveillance is linked to two main reasons. First, CSRs have aimed not only at preventing and correcting imbalances but also at addressing adjustment issues (e.g., addressing the labour market implications of the unwinding of imbalances). Second, some MIP-relevant challenges require policy action on a wide range of policy fields (e.g., restoring competitiveness may require action on a number of fronts including taxation, product and labour markets).

The MIP has usefully complemented other EU surveillance processes. The fact that government debt was in some cases considered a source of imbalances did not imply duplications in surveillance. Recommendations aimed at reducing high public debts in the context of SGP surveillance were simply considered to be relevant also from an MIP standpoint for those Member States identified with imbalances that also originate from public debt. A devoted recital in the Commission Communication accompanying its recommendations to the Council for the country-specific recommendations made clear which fiscal recommendations originating from the SGP were also relevant from the viewpoint of the MIP. Conversely, Member States under MIP surveillance received also recommendations focusing on longer-term potential growth and productivity issues for high debt countries, which usefully complemented the SGP by providing a surveillance framework to address public finance challenges arising from an insufficient growth potential. In other cases, recommendations to address imbalances in the context of the MIP called for fiscal expansions, notably with a view to reducing large and persistent current account surpluses, a type of recommendation that could not originate from the SGP, a procedure that is aimed at ensuring fiscal sustainability. Yet the MIP has sometimes been criticised for lacking focus. In particular, it has been criticised that the broad scope of the MIP does not help transparency

and may have reduced the effectiveness of the MIP in triggering policy responses. 73

Graph 12: Share of MIP-relevant subparts in country-specific recommendations for Member States experiencing imbalances or excessive imbalances

(a) per MIP category (b) per policy area (average 2013-2019)

100 2013 2014 2015 2016 2017 2018 2019

100 90

90 80

80 70

70 60 60 50

40 50

30 40

30 10

20 0

10 Financial Public Public Structural Labour Social

sector admin & finances & policies market inclusion 0 business taxation and

Imbalances Excessive imbalances environm. education

Source: European Commission

73 See, e.g., Bénassy-Quéré (2017).

Monitoring risks and policy reaction

A process of “specific monitoring” has been developed to strengthen the monitoring of policy progress for Member States under MIP surveillance. Specific monitoring comprises fact-finding visits to the Member States concerned, reports by the Commission, and discussions of the reports in the EPC and the Economic and Financial Committee (EFC). For the Member States subject to MIP surveillance, specific monitoring complements the monitoring of policy progress under the European Semester with an additional check on the

state of play. 74 Specific monitoring was introduced after the identification of excessive

imbalances in Spain and Slovenia in 2013. In 2014, it was expanded to all Member States with excessive imbalances and to selected euro area Member States with imbalances of systemic relevance. Since 2016, in light of the streamlining of MIP categories, specific monitoring has covered all Member States found to be experiencing imbalances or excessive imbalances. Its frequency changed from two to one reports per annual cycle in 2015/2016.

The ECOFIN Council adopts conclusions on the annual specific monitoring exercise and where applicable the monitoring is integrated with post-programme surveillance. Since 2016, the Council has adopted horizontal conclusions on specific monitoring, a practice that is in line with the October 2015 Communication, which invited the Council to be more involved in the specific monitoring. Those conclusions have been integrated with the ones on the Alert Mechanism Report. Specific monitoring reports have been made public. For Member States in post-programme surveillance, the two reports have been bundled. 75

Overall, specific monitoring has contributed to an enhanced policy dialogue. It has also allowed an up-to-date and informed peer review discussion in the Council Committees, focused on the implementation of policy measures to address identified imbalances.

Follow-up to the identification of excessive imbalances

The EIP has so far not been launched. For Member States identified with excessive imbalances, the Commission, rather than immediately recommending the Council to launch the EIP, aimed first at raising policy commitments in line with the challenges identified and at strengthening monitoring. For such cases, the Commission proposed more numerous, more detailed, and sometimes time-bound, CSRs. In addition, the Commission raised its vigilance, notably in the context of the specific monitoring (see above), by monitoring policy commitments, including those in the National Reform Programmes, and their subsequent implementation, while making clear that launching the EIP remained an option.

MIP surveillance has had some effectiveness in Member States with excessive imbalances. Member States identified with excessive imbalances were on average characterised by more recommendations as shown above, and a higher degree of policy responsiveness. This is found also in econometric work that permit to control for other factors that could have contributed to comply with country-specific recommendations (see Annex 2). 76 A possible interpretation is that the mere possibility of launching the EIP could have had an impact on policy progress. Nonetheless, more recent evidence relating to policy progress

74 Within the European Semester, a regular monitoring process applies to all CSRs, irrespective of whether countries are under MIP surveillance or if the CSRs are flagged as MIP-relevant or not, and the results are published in detail with the annual Country Reports.

75 See, e.g., the January 2019 ECOFIN Council Conclusions on the specific monitoring, which are bundled with those on the AMR: https://data.consilium.europa.eu/doc/document/ST-5603-2019-INIT/en/pdf .

76 See also MIP Compendium and J.C. Bricongne and A. Turrini (2017), "The EU Macroeconomic Imbalance Procedure: Some impact and no sanctions", VOXEU column, 22 June 2017, for similar results limited to the first years of MIP implementation.

according to MIP status reveal a somehow reduced policy response in Member States with excessive imbalances.

The lack of full enforcement of the MIP has been a subject of debate. The Council in its conclusions on in-depth reviews has remarked on several occasions that “the MIP procedure

should be used to its full potential, with the corrective arm applied where appropriate”. 77 The

Five Presidents’ Report stated that the EIP “should be triggered as soon as excessive imbalances are identified and be used to monitor reform implementation.” In its October 2015 Communication, the Commission expressed the view that the EIP could be “opened in case of insufficient commitment to reforms and lack of effective progress in implementation, and will be used in case of severe macroeconomic imbalances (…), like those that led to the crises.” The ECB has repeatedly claimed that the use of the EIP “could increase the procedure’s

effectiveness and credibility.” 78 The ECA in its audit of the MIP, expressed the view that “the

systematic non-activation of the EIP has reduced the credibility and effectiveness of the MIP”, yet also pointing out to somewhat contrasting views by other stakeholders that “using the EIP in circumstances which were not clear-cut would also affect its credibility, and that it should be reserved for use only in very obvious circumstances”.

Regulation No 1176/2011 i does not provide full guidance on the way the EIP should be implemented. In particular, it does not specify the time frame of the EIP, the scope and time frame of the corrective action plan, the assessment of effective action in the execution of the corrective action plan, or the criteria for abrogating the procedure or for the application of fines. This lack of detail coupled with the potentially intrusive nature of the corrective action plan, as recommendations could concern a wide range of policy fields and policy action can be prescriptive and are time-bound in the corrective action plan, may have contributed to nonuse of the EIP so far.

3.3 Effectiveness of the MIP

3.3.1 Follow-up to policy recommendations

Member States experiencing imbalances or excessive imbalances have implemented a higher share of their CSRs. Graph 13 reports that Member States under MIP surveillance have on average recorded more progress compared with other Member States. Thus, Member States under MIP surveillance in general have adopted a higher number of recommended measures both because they received more recommendations and because more recommendations were followed up by policy action. Statistical analyses suggest a stronger

follow-up to policy recommendations under the MIP. 79 Analysis in the MIP Compendium

finds that MIP surveillance was associated with a higher record of progress in addressing CSRs, controlling for other factors. More specifically, recommendations issued over the years 2013 and 2014 were followed-up to a greater extent by policy action in Member States with stronger activation of MIP surveillance (as measured by the categorisation of imbalances), after controlling for policy field, growth, interest rate spreads, and election variables.

77 See, e.g. the Council Conclusions on the 2016 and 2017 IDRs.

78 ECB, Economic Bulletin, Issue 2/2019, Box 7. Similar views have been expressed in several other instances, for example, Economic Bulletin, Issue 2/2018 and Economic Bulletin, Issue 2/2017.

79 See e.g. S. Darvas and A. Leandro (2015), "Economic policy coordination in the euro area under the European Semester", European Parliament, Directorate-General for Internal Policies, Study provided at the request of the Economic and Monetary Affairs Committee. The authors find that over the 2012-2014 period, the highest rates of implementation were observed for recommendations under the Stability and Growth Pact, followed by MIP- related recommendations, and then other country-specific recommendations.

The degree of CSR implementation has nevertheless declined recently. For CSRs issued in 2016 and 2017, the evidence suggests that the degree of policy progress recorded in Member States under MIP surveillance fell short of that of other Member States, while for 2018 Member States subject to MIP still exhibit stronger policy progress in light of a big drop in CSR compliance in the rest of the EU (Graph 13). Consistently, recent statistical evidence finds that CSR implementation for Member States with excessive imbalances was not significantly higher compared with that of other Member States, controlling for other

factors. 80

Graph 13: CSR progress indicator, averages according to MIP category 81

(a) MIP vs non-MIP (b) Excessive imbalances, imbalances, no imbalances

50 50 Non-MIP

Non-MIP MIP MIP imbalances

45 45 MIP excessive imbalances

re re

s co s co

a n ce 40 40

p li a

n ce p li

o m c 35 o

ge c 35

ra ge ra

A ve 30 A

ve 30

25 25 2013 2014 2015 2016 2017 2018 2013 2014 2015 2016 2017 2018

Year of issuance Year of issuance

Note: the MIP status is the one at the time of CSR issuance and CSR progress in year t refers to CSRs issued in year t and assessed in year t+1; the data reported concern progress at CSR subpart level.

Source: European Commission

A number of factors could have played a role in reducing policy progress in Member States experiencing imbalances or excessive imbalances:

• The improving economic situation and reduced market pressure reduced the sense of

urgency to address persisting imbalances;

• Considerable reform activity took place in earlier years in a number of Member States,

thereby reducing the perceived need of carrying out additional marginal reforms;

80 K. Efstathiou, and G. Wolff (2019), "What drives national implementation of EU policy recommendations?" Bruegel Working Paper, Issue 04, April 2019.

81 The CSR progress indicator is calculated as the simple average of progress for all CSRs. It is based on five possible categories: no progress; limited progress; some progress; substantial progress; full progress, to which values of 0, 25, 50, 75, and 100 are assigned respectively following the methodology first used in S. Deroose and J. Griesse (2014), "Implementing economic reforms – are EU Member States responding to European Semester recommendations?", ECFIN Economic Brief, Issue 37. Thus, an overall progress indicator of 100 indicates full progress for all CSRs subparts, and a reading of 0 arises in case of no progress in any CSR subpart. The progress indicator at CSR subpart level is available starting only from the CSRs issued in 2013 as no assessment at subpart level was carried out for the CSRs issued in 2012.

• In light of past reform record, reform fatigue may have played a role, implying a

reduced “political capital” left available for additional reforms;

• Statistical results could be affected by “reverse causation”, because strong policy

responses were observed in Member States that exited the MIP (or that improved their categorisation). This implied that policy-active Member States moved away from the set of so-called MIP countries and entered that of non-MIP countries, thus contributing to the statistical findings that the policy progress in MIP countries fell over time (in

Graph 13 the set of countries change over time).

The evidence in Annex 2 confirms that Member States under MIP surveillance recorded a significantly higher degree of policy progress controlling for other factors. As compared with the MIP Compendium, the analysis includes more recent years, and controls for additional factors, notably the past stock of recommendations received and past CSR progress, thus capturing reduced need of marginal reforms and reform fatigue respectively. The inclusion of those variables tend to increase the estimated advantage of MIP Member States in terms of policy progress. The estimates also show that the positive impact of MIP surveillance is estimated higher once the role of reverse causation is taken into account.

3.3.2 The development of a supportive policy environment

The impact of the MIP should also be judged on the basis of its contribution to a supportive policy environment, strengthening awareness of challenges and the policymaking framework. Box 4 reviews several country-specific examples for the impact of MIP surveillance on the policy environment. MIP surveillance has contributed to a supportive policy environment in a number of respects:

• Raising awareness about the issues around macroeconomic imbalances;

• Providing a framework for prioritising the national policy agenda with a view to

ensuring macroeconomic stability;

• Strengthening policy dialogues on macro-stability issues between national authorities

and EU institutions and among Member States.

MIP surveillance has contributed to raise awareness about the relevance of macroeconomic imbalances and related challenges. The MIP has put macroeconomic imbalances on policy agendas at national level and helped creating awareness and mobilising

political capital for subsequent policy action. 82

The MIP has helped focusing and prioritising the domestic policy debates. For instance, in a survey of EPC members conducted by the European Court of Auditors in the context of their audit of the MIP, nearly three quarters of the respondents found that the MIP had some or considerable impact on domestic policy debate, and a similar share was of the view that the MIP is somewhat or highly effective. Almost two thirds of the respondents to that survey agreed that the MIP has made the Commission’s views more important for policy making at

domestic level. 83 At the same time, the MIP sets a framework that allows policy makers to

82 See, e.g., D. J. Bokhorst (2019), "Governing Imbalances in the Economic and Monetary Union: A Political Economy Analysis of the Macroeconomic Imbalance Procedure", Doctoral dissertation, University of Amsterdam, Department of Political Science; Pierluigi and Sondermann (2018), op. cit.

83 ECA (2018): answers to questions 36, 40 and 44 of the survey to EPC members.

disclose their policy commitments in a structured way, and where commitments are regularly monitored.

The MIP has provided a comprehensive framework for dialogue. The MIP has set a framework that allows policy makers to disclose their policy intentions in a structured way. All steps of the annual cycle of the MIP implementation (Alert Mechanism Report, in-depth reviews, country-specific recommendations, policy follow-up including monitoring) bring the main messages to the attention of the Member States. Technical discussions take place in committees, notably the EPC and its LIME working group. The ECOFIN Council adopts and publishes conclusions in the MIP context twice a year, namely a conclusion on the Alert Mechanism Report and the specific monitoring outcomes at the start of the calendar year, and another one on the in-depth reviews in spring. Overall, the MIP has enshrined macroeconomic imbalances on the Council agenda on a structured and regular basis.

Overall, the MIP was generally associated with more recommendations and more policy responses, and contributed to a supportive policy environment. But its political traction was in some cases insufficient. The complexity of MIP issues, the broad scope of the procedure, and the lack of simple numerical decision rules have implied that MIP objectives and recommendations are less visible and known to the large public, e.g. as compared to the Stability and Growth Pact. For those reasons, in the current set-up of the surveillance process it has been challenging to leverage political traction of the MIP recommendations.

Box 4: Learning from country cases

This box presents a number of instances in which MIP surveillance contributed to the strengthening of the policy response to macroeconomic imbalances. The country cases selected are not exhaustive and aim simply to illustrate different ways in which the MIP contributed to a more supportive policy environment.

Spain and Slovenia were the first two cases identified with excessive imbalances and where specific monitoring was applied. In April 2013, both Member States had been found with excessive imbalances. The MIP provided the framework for them to engage in a structured dialogue with the Commission, the Council and the other Member States on reform plans, to disclose their National Reform Programmes, and to support enhanced reporting on actual progress. In both cases, the relatively strong progress with reforms contributed to fostering macroeconomic stability in high-risk settings without the need to resort to the EIP. In more detail:

• Spain agreed a financial assistance programme for the recapitalisation of its financial

institutions in July 2012 and which ended in early 2014. There was a visible acceleration of reform efforts in late 2013 and throughout 2014. Several important reforms were adopted to: strengthen public finance management, the local administration and the pension system; avoid the build-up of commercial arrears; adopt product market reforms aimed at addressing regulatory fragmentation in Spain's internal market; reform the corporate insolvency framework; establish the independent fiscal council. Reforms to network industries as well as reforms to firm’s access to finance were also on the list. In March 2014, Spain was de-escalated from excessive

imbalances to imbalances requiring decisive action.

• In 2013, Slovenia was struggling with excessive imbalances and hiking market

pressures on its sovereign debt. Against that background, considerable progress was made as of 2013 in repairing the banks' balance sheets. Efforts included asset quality

reviews, stress tests, recapitalisation of State-owned banks and transfer of nonperforming-loans to a 'bad bank', the Bank Asset Management Company; a new legislative framework for corporate restructuring was introduced in December 2013; also in 2013, the labour market reform addressed segmentation and introduced greater flexibility, and the pension reform improved the medium-term sustainability of the pension system. The Slovenian Sovereign Holding, responsible for the management and divestment of State assets, became fully operational, and a new corporate governance code for State-owned enterprises was adopted in December 2014. Slovenia moved from excessive imbalances to imbalances in February 2015 and eventually corrected the imbalances by March 2018.

In other situations, the MIP worked by combining calls on the need of addressing governance issues with a ready-set framework for swift policy response.

• In early 2015, Bulgaria was moved from imbalances to excessive imbalances

following the financial turbulence in the second half of 2014 that revealed major governance flaws and could have had significant implications for the financial sector and the overall macroeconomic stability. Such escalation in MIP surveillance was followed by CSRs to ensure the robustness of the banking sector and the revision of the supervision of the banking and the non-banking financial sectors. The MIP surveillance leveraged the recommendations for timely completion of asset quality reviews and stress tests of the banking sector, a portfolio screening for the pension funds and insurance sectors, and the take up of actions found needed. All that happened in a context of a still negative, even if improving, external position, corporate overleveraging and weak labour market adjustment. In addition to calls for action, the MIP crucially provided the setting for the close and integrated monitoring of policy action and relevant developments. In light of considerable progress in tackling the issues identified and monitored under MIP surveillance, and positive developments on the front of bank balance sheets, non-performing loans, and other relevant macroeconomic variables, Bulgaria imbalances were no longer identified as excessive

in 2018.

• Romania was found to be with imbalances in March 2019. That decision was also

taken based on concerns around recent legislative initiatives creating risks to the financial sector and private investment. The Romanian authorities have since reversed part of those initiatives even though risks are still building up. Here too the MIP can

provide a framework to closely and timely monitor those developments.

There were also cases where the MIP worked by raising awareness and giving prominence to certain issues and help mainstreaming and entrenching them on the policy debate.

• For instance, in Sweden, the in-depth review gave augmented prominence to housing

issues since the very beginning of MIP surveillance in 2012. More generally, the MIP helped in keeping the topic high on the policy debate.

• Similarly, the in-depth review called for attention to the large current account surplus

of the Netherlands. In addition, the MIP helped keeping the issue of mortgage interest deductibility on personal income taxes on the policy agenda despite reluctance

from the national authorities to act on it.

• Likewise, the MIP has managed to cast Germany’s high current account surplus on

the policy debate. It raised awareness and shifted the debate with the current account surplus not being regarded only as a re-assurance of strong competitiveness, and led to an engagement on a debate with the Commission and the other Member States (see

also Bokhorst, 2019).

• The MIP has fed the policy debate in Croatia and, in light of potentially stronger

enforcement associated with the EIP. Moreover, the policy response right after being classified as excessive imbalances triggered visible preparation of policy plans, as shown in the subsequent National Reform Programmes, even if subsequent action was

somewhat left wanting.

3.3.3 Impact on economic outcomes

A thorough assessment of the MIP surveillance on economic outcomes is beyond the scope of the present review. The MIP can affect relevant economic variables not only via its impact on policies carried out by Member States under surveillance but also via the impact that the presence of a structured surveillance framework could have on market sentiment. A thorough assessment would require a counterfactual that permits to check countries’ performance in absence of the MIP. 84 However, the application of econometric techniques usually employed to construct such counterfactual would hardly permit robust conclusions in the case of the MIP for two main reasons. First, the experience with the implementation of the framework is rather limited, implying that the data do not contain sufficient observations to assess the effectiveness of the MIP in the correction of slow-moving stock variables, or the prevention of the accumulation of new imbalances. Second, and moreover, MIP objectives are multifaceted and country specific, implying that a test of MIP effectiveness based on the impact of the MIP on a single or few economic variables cannot lead to general conclusions. The remainder of this section will present nonetheless prima-facie evidence aimed at corroborating a number of tentative findings on the impact of MIP surveillance.

Graph 14: Sovereign 10-year bond spread against Germany

o in p 3

n ta

ce 2 p er

2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Identified with excessive imbalances for at least 3 years

Identified with imbalances for at least 3 years

Other EU countries

Note: A Member State is classified as identified with excessive imbalances or imbalances if it has been in that category for at least 3 years (not necessarily consecutively) since MIP inception. Member States never subject to IDR, or subject to an IDR but with no imbalances found, or with imbalances (or excessive imbalances) but for

84 An approach sometimes used to establish a counterfactual against which to compare the effects of economic surveillance frameworks is by means of a difference-in-difference regression approach (aimed at assessing whether the presence of MIP surveillance made a difference for the countries concerned as opposed to those countries not under the MIP, controlling for other factors). An alternative is the use of “synthetic counterfactual” analysis (where the performance of each MIP country under MIP surveillance is confronted with that of a synthetic comparator combining the relevant characteristics that best resemble the country under investigation, except being under MIP surveillance).

less than 3 years, are included in the group “Others” for the sake of comparison. Actual composition of the Member State groups is as follows: “Identified with excessive imbalances”: BG, FR, HR, IT, CY, PT; “Identified with imbalances”: BE, IE, ES, HU, NL, SI, FI, SE, UK”. Averages are unweighted.

Source: Eurostat

Member States experiencing imbalances, and especially those found to have excessive imbalances, recorded, on average a bigger drop in risk premia as compared with other Member States. Over the post-crisis period, when the MIP was introduced, macro-financial risks as reflected by markets were considerably higher in Member States that were subject to MIP surveillance. Over time, notably as a result of enhanced governance for euro area Member States and the establishments of firewalls addressing default risk and contagion (European Stability Mechanism; Banking Union, with the Single Supervisory Mechanism and the Single Resolution Mechanism; and measures by the European Central Bank, such as Quantitative Easing and announcement of the Outright Market Transactions programme), risk premia fell in across the EU, with more sizeable reductions on average in cases identified with excessive imbalances (Graph 14).

Economic activity in Member States that have undergone MIP surveillance has strongly accelerated after years of weaker economic performance. In particular, Member States marked by excessive imbalances have underperformed relative to their EU peers in the post crisis period, with weaker GDP growth and higher unemployment. However, the economic recovery appears to have benefited especially a number of Member States under MIP surveillance as growth resumed more markedly therein, especially those that were more affected by adjustment to large current account deficit reversals. In addition, unemployment exhibited a bigger drop as compared with non-MIP Member States.

The evidence suggests that MIP surveillance has helped to address the risks stemming from macroeconomic imbalances, which in turn improved economic growth. The reduction in risk premia in Member States subject to MIP contributed to a stronger growth performance by improving prospects for public finances and the banking sector, thereby

creating the conditions for resumed domestic demand, notably investment. 85 Resumed growth,

in turn, had a positive feedback loop on perceived risks. The fact that the reduction in risk was stronger in Member States under MIP cannot be taken as evidence. Other factors undoubtedly played a role, including market reactions and policies not affected by MIP surveillance. The evidence is nonetheless consistent with increased policy activism due to the MIP or improved market sentiment associated with the presence of a structured surveillance framework set by the MIP.

Sources of imbalances have been receding in Member States. Sources of imbalances identified in the earlier years of MIP implementation were often related to large current account deficits, past competitiveness losses, inflated house prices, high debt levels often coupled with undercapitalised banking systems. The available evidence suggests that in most cases, Member States under MIP surveillance have recorded improvements in variables linked to the identified imbalances. Graph 15 and Graph 16 report consistent evidence for external imbalances (current account balances and NIIPs). Member States under MIP surveillance with imbalances identified in areas relating to excessively large current account deficits were reporting much larger deficits compared with other Member States but recorded faster

85 On the role of macroeconomic imbalances in affecting patterns of growth and convergence in the euro area see, e.g., M. Buti and A. Turrini (2015), "Three waves of convergence. Can Eurozone countries start growing together again?" VOXEU column, 17 April 2015; and L. Coutinho and A. Turrini (2019), "Convergence and Macroeconomic Imbalances", Quarterly Report on the Euro Area, European Economy, Institutional Paper 111, July 2019.

improvements during the years under MIP surveillance. Competitiveness also improved faster

in Member States with the need to rebalance after a history of large current account deficits. 86

Similar considerations apply to stock variables like the NIIP and private debt, although improvements were visible only recently and stock imbalances remain important.

Graph 15: Current Accounts: deficits and Graph 16: NIIP surpluses as source of imbalances

10 60

5 20

P 0

f G

D 0

-20

% o f G

-5 % o -40

-60

-10 -80

-100

-15 -120 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018

Under MIP for at least 3 years, CA deficit a source of imbalances Under MIP for at least 3 years, CA deficit a source of imbalances

Under MIP for at least 3 years, CA surplus a source of imbalances Under MIP for at least 3 years, CA surplus a source of imbalances

Other EU countries Other EU countries

Note: "CA deficit a source of imbalances" ("CA surplus a source of imbalances") includes those Member States that have had current account deficits (surplus) identified as a source of imbalance at least once (as presented in the MIP matrix or chapeau communication); and, being classified with imbalances or excessive imbalances for at least 3 years. Actual composition of the groups is as follows: "CA deficit a source of imbalances": BG, CY, ES, HR, HU, IE, PT; "CA surplus a source of imbalances": DE, NL, SE. Averages are unweighted.

Source: AMECO

Despite the positive findings on the effectiveness of MIP surveillance in addressing imbalances and contributing to risk reduction, a number of challenges stand out:

• Stock variables like high public and external debt adjust slowly. Imbalances

associated with flow variables, like current account balances or house price growth, adjusted swiftly in light of market reactions in a context of compressed demand dynamics and, later on, on the back of the improving economic environment. More recently, rebalancing has been slowing, also as cost competitiveness developments became less supportive. Stock variables like the NIIP or private debt, which are measured as a share of GDP, adjusted much more slowly because the unwinding of imbalances was associated with a detrimental “denominator effect”, on account of the reduced or even falling nominal GDP growth. The economic recovery, which became gradually more widespread and consolidated, eventually helped to reduce stock imbalances, but vulnerabilities linked inter alia to high debt, non-performing loans, or external liabilities still remain high in some Member States, requiring continued

prudent external positions.

86 See, e.g., European Commission, Alert Mechanism Report, issues 2015-2019. Regarding imbalances associated with past housing booms, downward house prices corrections were in most cases recorded in the countries concerned already before MIP surveillance started being implemented (e.g., Spain).

• Large and persistent current account surpluses in some Member States adjust

slowly. Because of the simultaneous progressive reduction of large current account deficits and the persistency of large current account surpluses in euro area Member States, in light of a major deleveraging processes involving various sectors of the economy, including in Member States with little or no deleveraging needs, and strong export performance, the overall current account position of the euro area has moved from broadly balanced to a substantial surplus. The progressive rise in the euro area surplus reveals persistent weak demand dynamics in the post-crisis period that underpinned the low inflation environment in the euro area. Rebalancing at the euro area level of both current account deficits and surpluses matter for the MIP surveillance as that rebalancing would help fostering nominal growth, which in turns helps narrowing imbalances. Systemic issues linked to the persistency of large surpluses and their impact on the euro area were not at the centre of the policy discussion when the MIP entered into force. 87 Progress on this front will be a key test for the effectiveness of the MIP. In this respect, it is important to highlight that while the correction of large current account deficits is often the result of market dynamics (notably, a sudden stop in external financing), the adjustment of surpluses is generally more gradual. In this context, supportive policies sustained over time could make a difference in fostering a rebalancing of both current account deficits and surpluses

across the euro area and the EU.

• The effectiveness of the MIP in preventing the accumulation of new imbalances is

still untested. The economic recovery is bringing new possible challenges in a number of Member States, some of them not directly under MIP surveillance, linked notably to fast-growing unit labour costs and deteriorating competitiveness, acceleration in house prices in a context of resuming household indebtedness, and current account balances returning into negative territory. Detecting those risk and eliciting appropriate policy responses will be a key test on the preventive capacity of

the MIP.

3.4 Conclusions

Overall, the MIP helped broadening economic surveillance. But a number of challenges stand out. The procedure has sometimes been judged as lacking transparency. In particular, the scope of the MIP has sometimes been seen as too broad and its boundaries as blurred. The MIP contributed to a supportive environment to address macro-stability challenges, but it is debatable whether the full potential of the MIP in triggering policy action has been exploited, especially due to the limited political traction that the procedure has generated. Regarding economic outcomes, while flow imbalances linked to the pre-crisis booms have been corrected in a majority of Member States, the correction of stock imbalances such as high public, private and external debt has been slower and is still incomplete. Regarding external rebalancing, the correction of large and persistent current account surpluses has been slow. The effectiveness of the MIP in preventing the accumulation of new imbalances is largely untested. The rest of this section reviews the achievements to date, and related challenges ahead, along three central questions.

87 Recital 17 to Regulation No 1176/2011 i acknowledges the lower degree of risk associated with large current account surpluses as compared with deficits, a point made also in a Commission declaration joined to the same Regulation.

Widening the scope of the macroeconomic surveillance

The MIP broadened surveillance, complementing other procedures, and enhancing the analytical and statistical basis to detect imbalances. The reason underpinning the introduction of the MIP was to give EU macroeconomic surveillance more effective instruments to prevent and correct challenges to macroeconomic stability of various nature, thereby complementing existing surveillance procedures. The MIP framework for analysis was enriched and expanded in cooperation with Member States. Progress was made on towards enhancing the standards and methods used for the compilation of data used in MIP analysis as well as strengthening the statistical quality assurance framework. A Memorandum of Understanding was concluded in 2016 between Eurostat and the European Central Bank/Directorate General Statistics on the quality assurance of statistics underlying the

MIP. 88 This review of economic governance provides an opportunity to reflect further on how

the quality of MIP statistics is best assured.

The scope of MIP surveillance has remained anchored to macroeconomic stability. Gradually, it has been taking into account also broader implications relating to adjustment, notably those affecting employment and social developments. The crisis highlighted the need for an enhanced framework to monitor and correct challenges arising from issues such as current account imbalances, private debt, or housing bubbles, as all these factors compounded stability risks originating from high government deficits and debt levels. In this respect, the MIP can act as a useful complement to the Stability and Growth Pact by prioritising policies not dealt with directly by the SGP, but which can be of relevance for achieving sound public finances. In particular, by covering longer-term potential growth and productivity issues in high-debt countries, the MIP can help addressing public finance challenges arising from an insufficient growth potential. The existing synergies between the MIP and the SGP may have not been fully exploited effectively.

European-level considerations and systemic aspects were taken into account in applying MIP surveillance. Decisions under MIP surveillance were driven mostly by the assessment at Member State level. However, issues pertaining to the level of the EU and euro area as well as systemic and international aspects were also taken into account. Thus, decisions also reflected the overall economic situation and outlook of the euro area, and the presence of interdependencies and spill-overs with possible negative implications for the rest of the euro area or the EU. Following the Five Presidents' report, the euro area dimension of imbalances has been assessed on a regular basis in recent Alert Mechanism Reports and reflected in country-specific and euro area recommendations.

The MIP has gradually taken on board social aspects linked to adjustment issues. Social and employment considerations linked to adjustment to macroeconomic imbalances have been gradually incorporated in MIP surveillance and recommendations were increasingly geared to cater for the trade-offs between ensuring macroeconomic adjustment and addressing social distress. The need to take into account also trade-offs between macroeconomic adjustment and environmental and climate challenges may arise in the future, while keeping the macro-stability focus of the MIP.

Policy responses and correction of imbalances

The MIP contributed to a supportive policy environment. The MIP contributed to raise awareness of the relevance of the challenges, mobilise political capital and focus policy

88 As an outcome, a joint ESS-ESCB annual quality reports provides an assessment of the quality of these statistics.

action, and deepen the dialogue between EU institutions and national policy authorities. In a number of cases, the MIP provided a framework for policy efforts to shore up macroeconomic stability in Member States severely threatened by the risk of debt or financial crises, without the need to resort to the more intrusive EIP instrument. More generally, the MIP was effective in raising prominence of risks and challenges and feeding the national debate on MIP-related challenges and policy responses.

MIP surveillance was associated with stronger policy progress. Evidence suggests that Member States under MIP surveillance received more recommendations and exhibited more policy progress as compared with the other Member States. That effect was particularly strong in the first years of MIP implementation and in a context where market-based discipline proved relevant and awareness of the need to address macroeconomic imbalances was widespread. However, policy progress declined more recently, and is now comparable to that of other recommendations under the European Semester, reflecting lower perceived urgency of reform and reduced political capital.

MIP surveillance was associated with a correction of most flow imbalances and a gradual reduction of stock imbalances. Large current account deficits were reduced or turned into surpluses, competitiveness developments started being supportive of rebalancing, deleveraging in the private sector took place in Member States with large debt, inflated house prices undergone a downward correction. Stock variables (e.g., private debt, NIIP) underwent a more gradual correction, and embarked on a downward trend only once nominal GDP growth resumed, requiring continued prudent external positions going forward. The evidence is consistent with a possible role played by the MIP in contributing to reduced risks and thereby a fast recovery in output and employment in Member States under MIP surveillance, although causation is difficult to establish.

The MIP has been punching below its weight when it comes to the effectiveness in triggering policy action. Despite MIP surveillance being associated with somehow stronger policy responses and a more supportive policy environment, the MIP might have not managed to get sufficient political traction to mobilise the political capital needed to carry out major reforms also in the absence of market pressure. This may point to weaknesses in terms of communication and focus. It has also been noted that the MIP was so far not used at full potential, as the EIP has not yet been activated. This may be related to the lack of clarity on the implementation of the EIP and on how corrective action under the MIP would be applied.

The jury is still out on the effectiveness of the MIP in preventing the accumulation of new imbalances. When the MIP was introduced, imbalances were already present in many Member States. However, the effectiveness of the MIP in preventing the accumulation of new imbalances is still untested. The economic expansion of recent years has brought new possible challenges in a number of Member States, some of them not directly under MIP surveillance. Those challenges are linked notably to fast-growing unit labour costs and deteriorating competitiveness, acceleration in house prices in a context of resuming household indebtedness, current account balances returning into negative territory. Detecting those risks and eliciting appropriate policy responses will be a key test on the preventive role of the MIP.

External deficits have seen a stronger correction that external surpluses. The larger risks associated with external deficits has entailed a stronger focus of the MIP on those Member States, especially in the first years of application. Those deficits have been gradually reduced, while prudence remains warranted in view of the remaining stock imbalances. However, large and persistent current account surpluses in a number of Member States have not yet been corrected. In this respect, it is important to highlight that while the correction of large current account deficits is often the result of market dynamics linked to a sudden stop in external financing, this is not the case for surpluses, whose adjustment is generally more gradual.

Altogether, the progressive rise in the euro area surplus reveals persistent weak demand dynamics in the post-crisis period that underpinned the low inflation environment in the euro area. Therefore, the rebalancing not only of current account deficits but also that of surpluses would help boost growth in the Member States concerned, and would have positive spillovers to the rest of the euro-area. For instance, already in the conclusions to its audit, the European Court of Auditors pointed out that the MIP was not encouraging symmetric rebalancing within the euro area and recommended the Commission through the MIP to "give systematic consideration to policies with cross-country impacts that can enhance symmetric rebalancing

within the euro area." 89

Ownership and governance

The Commission has used economic judgement in the application of the MIP. MIP surveillance does not lend itself to a mechanistic use of numerical rules. Issues dealt with in the MIP context are multifaceted and interrelated. Moreover, the nature of imbalances may change across Member States and over time. To address the many macroeconomic challenges that are the object of MIP surveillance, a wide range of policy instruments is needed, which sometimes exert only an indirect and possibly weak effect on economic outcomes. As opposed to fiscal surveillance, use of numerical rules would not be appropriate to MIP surveillance. Hence, the identification and assessment of imbalances has required economic judgement in the first place in line with definition of imbalances in Regulation No 1176/2011 i:

• Economic judgement has also been used in the surveillance and follow up to the

identification of imbalances. The assessment and monitoring framework has evolved in light of the accumulated experience, making use of the discretion allowed by the

legal framework.

• While the EIP has never been launched, where Member States were identified with

excessive imbalances, recommendations in the context of the MIP were more prescriptive compared to those for Member States not directly under MIP, and

monitoring of policy response to recommendations was stepped up.

The use of economic judgement by the Commission was balanced by enhanced transparency and accountability. The natural counterpart to discretion on the part of the Commission is transparency and accountability towards Member States and main stakeholders, with a view to ensure equal treatment and predictability. This balance is anchored in the Regulation No 1176/2011 i, which endows the Commission with discretion as concerns assessments and decisions in the MIP context, but under close scrutiny from the Council, which takes conclusions on Commission decisions. While the Council has generally endorsed Commission decisions, a request of increased transparency and predictability was voiced by Member States and EU institutions (notably in the Council conclusions on in-depth reviews and the Five Presidents' report). The Commission has taken steps to enhance transparency regarding the criteria underpinning assessments (made public in the MIP Compendium), the reasons underlying specific decisions (clarified in the in-depth reviews by means of MIP Assessment Matrices), and the implications of MIP decisions in terms of recommendations and activation of surveillance. At the same time, the Commission simplified the framework by streamlining and stabilising the classification of imbalances. In a nutshell, consistency and predictability of the framework was enhanced by increased transparency and the built up of a stock of precedents under the MIP, which made it possible to judge new cases in light of past decisions with analogous challenges and under similar conditions.

89 ECA(2018) op. cit., recommendation 5.

There have been calls to further enhance transparency and predictability. Despite steps being taken towards a more transparent and stable MIP implementation, some policy makers and experts see scope for further improvement. The European Court of Auditors has recommended clarifying the link between MIP recommendations and recommendations under other surveillance streams (MIP vs. European Semester, link to SGP recommendations, link with recommendations to the euro area). In this respect, the Commission has already taken first steps: in line with that recommendation by the European Court of Auditors, the 2019 country-specific recommendations include recitals that explain the links between SGP and MIP-related recommendations. A finer mapping between MIP challenges and recommendations would also help focusing the procedure, giving priority to critical recommendations to address imbalances and favour adjustment. Some authors have argued

that a more focused MIP could increase its effectiveness. 90 Calls for transparency and

disclosure have also concerned the Commission’s reasoning for not launching the EIP, as echoed in the European Court of Auditors recommendations. 91 Increased transparency could also concern aspects linked to the implementation of the EIP not detailed in the Regulation No 1176/2011 i (see section 3.2.3)

90 E.g., Bénassy-Quéré (2017), op. cit.; Efsthatiou and Wolff (2018), op. cit.

91 See also the Council conclusions on the 2019 IDRs: https://data.consilium.europa.eu/doc/document/ST-9473- 2019-INIT/en/pdf

4.

E URO AREA M EMBER S TATES EXPERIENCING OR THREATENED WITH
SERIOUS DIFFICULTIES WITH FINANCIAL STABILITY

4.1 Objectives

Regulation No 472/2013 i provides a framework under which the Commission and the Council can exercise a degree of strengthened surveillance complementary to other existing multilateral surveillance processes. It sets out rules for enhanced surveillance, macroeconomic adjustment programmes and post-programme surveillance. In particular, it sets the framework for macroeconomic adjustment programmes for euro area Member States in case financial assistance is inter alia provided by the European Stability Mechanism (ESM), which is an intergovernmental institution outside the EU legal framework.

Under Regulation No 472/2013 i, the Commission and the Council can require that a euro area Member State takes further measures to address the specific risks emanating from that Member State to preserve financial stability in the euro area. Those measures aim to rapidly re-establish a sound economic and financial situation and, if relevant, restore the euro area Member State’s capacity to finance itself fully on the financial markets.

The changes agreed in the context of ESM Treaty reform do not require a change to Regulation No 472/2013 i. The Regulation was also used as the basis for agreeing the Memorandum of Cooperation between the Commission and the ESM on their future working relations. 92

4.2 Enhanced surveillance

Enhanced surveillance aims to ensure that a euro area Member State takes corrective measures to address the sources of its structural imbalances with a view to ensure its swift return to a normal situation and to protect other euro area Member States from potential spill-over effects. As such, the decision by the Commission to subject a euro area Member State to enhanced surveillance will usually be based on the findings of the normal European Semester cycle of economic and fiscal policy coordination, notably the Alert Mechanism Report and where available the latest in-depth review.

Enhanced surveillance has so far never been applied ex-ante. It has been applied, expost, in the case of Greece. Indeed, when Regulation No 472/2013 i was introduced, no further cases warranted applying that process. Moreover, at that stage, the other euro area Member States in difficulty were already under a macroeconomic adjustment programme. Cyprus was the only Member State whose economic and budgetary surveillance took place within the European Semester process before requesting a macroeconomic adjustment programme. However, the macroeconomic imbalances were so high that it was only a matter of months before Cyprus had to request financial assistance. Greece started its third macroeconomic adjustment programme in 2015 but had been subject to two other programmes prior to then.

92 The Memorandum of Cooperation will be concluded between the Commission and the ESM once the amended Treaty of the ESM has been ratified. It will be based on the Memorandum of Understanding concluded in April 2018 and the Joint Position of the Commission and the ESM in November 2018, https://ec.europa.eu/info/business-economy-euro/economic-and-fiscal-policy-coordination/eu-financial href="https://ec.europa.eu/info/business-economy-euro/economic-and-fiscal-policy-coordination/eu-financial-assistance/loan-programmes/european-stability-mechanism-esm_en">assistance/loan-programmes/european-stability-mechanism-esm_en .

The regime foreseen by Regulation No 472/2013 i relative to enhanced surveillance was applied in the case of Greece, but given that the procedure only started at end of August 2018, it is too early to draw definitive conclusions. The decision to subject Greece to enhanced surveillance following the end of its financial assistance programme in August 2018 was taken because Greece continued to face risks with respect to its financial stability, which, if they materialise, could have adverse spill-over effects on other euro area Member States, in particular via confidence effects that could increase refinancing costs for their banks and sovereigns.

4.3 Financial assistance and macroeconomic adjustment programmes

For macroeconomic adjustment programmes, the aim of Regulation No 472/2013 i was to codify the process for designing a programme and properly involving the necessary stakeholders sufficiently early in the process. In particular, Regulation No 472/2013 i provides for specific requirements for the Commission for euro area Member States under a macroeconomic adjustment programme, namely: (i) the preparation of the programme and the information to relevant stakeholders on progress made; (ii) ensuring consistency between the Memorandum of Understanding signed with the euro area Member State and the macroeconomic adjustment programme prepared; (iii) monitoring the progress made in implementing the programme through regular review missions and the involvement of social partners and relevant civil society organisations during these missions; (iv) the granting of technical assistance where the requirement is that it helps with the implementation of programme conditionality.

The provisions of Regulation No 472/2013 i governing the process for framing macroeconomic adjustment programmes have been respected. Information has been provided more regularly to the European Parliament on the progress made with the implementation of programmes in the case of Greece and Cyprus than Portugal. This was due to the fact that the programme for Portugal began prior to the adoption of Regulation No 472/2013 i.

The framework put in place by Regulation No 472/2013 i has been effective in dealing with the problems that led euro area Member States to request macroeconomic adjustment programmes. All the euro area Member States that received financial assistance have been able to return to markets at reasonable financing rates; external and fiscal balances have been largely resolved; and the stability of the financial sector has been restored. Regulation No 472/2013 i also helped to preserve the financial stability of the euro area as a whole by putting those procedures in place.

Regulation No 472/2013 i has provided the right degree of scrutiny in the economic situation of the vulnerable euro area Member States. It has also allowed a proper management of the procedures such that the imbalances in the euro area Member States were resolved in an orderly fashion and without creating major disruption. At the same time, it appears that it has not always been easy to sustain national ownership of the macroeconomic adjustment programmes particularly at the time of programme exit or programme discontinuation when incentives for authorities to backtrack on certain reforms agreed in the programme increased.

In addition, Regulation No 472/2013 i has contributed to a closer coordination of economic policies and a sustained convergence among euro area Member States. This has been achieved by ensuring consistency between the normal surveillance cycle of the European Semester and the strengthened surveillance of a macroeconomic adjustment programme. Vulnerabilities and imbalances identified in the normal surveillance cycle were explicitly addressed in macroeconomic adjustment programmes. Outstanding adjustment needs after the end of a programme were subsequently addressed by the country-specific recommendations in the context of the European Semester.

4.4 Post-programme surveillance

According to Regulation No 472/2013 i, once a euro area Member State is no longer subject to a macroeconomic adjustment programme, it will be subject to postprogramme surveillance until at least 75% of the financial assistance received is paid back. The objective of post-programme surveillance is to ensure a Member State’s capacity to repay the financial assistance granted. At the same time, post-programme surveillance aims to ensure a continuous assessment of a Member State’s economic, fiscal and financial situation and identify any risks to its medium-term viability allowing for the continuation of any remaining adjustment.

The provisions governing post-programme surveillance have been implemented in full accordance with the requirements of Regulation No 472/2013 i. The framework put in place by Regulation No 472/2013 i for post-programme surveillance foresees specific requirements in terms of: (i) the information to be provided by the euro area Member State; (ii) the monitoring process by the Commission in liaison with the ECB; (iii) the information provided to the European Parliament, the EFC and the national Parliament of the Member State concerned. The assessment of remaining adjustment needs to be undertaken by the euro area Member State exiting a macroeconomic adjustment programme has been consistent with the findings of the Country Reports published as part of the European Semester cycle. These assessments are also in line with the in-depth reviews conducted for those Member States that are still found to have imbalances in the framework of the Macroeconomic Imbalance Procedure. The requirements of Regulation No 472/2013 i have been respected in terms of the additional data to be provided by the Member State concerned and the regular information provided by the Commission to the European Parliament, EFC and parliaments of the Member State under post-programme surveillance. So far, no corrective measures have been deemed necessary in any of the euro area Member States, given the evolution in their adjustment process. There has also been no specific requests from the European Parliament or parliaments for exchanges of views on the progress in post-programme surveillance in those euro area Member States, in addition to the regular review mission reports.

4.5 Conclusions

All of the criteria 93 that provided the basis for the evaluation have been broadly met. At the same time, experience has shown that the implementation of the Regulation No 472/2013 i could be improved, particularly as regards certain aspects of the surveillance processes as well as in relation to economic adjustment programmes.

When to place a Member State in enhanced surveillance

Regulation No 472/2013 i provides concrete elements of when to place a euro area Member State under enhanced surveillance in some particular cases, but leaves

93 These criteria are: (i) the effectiveness of the Regulation; (ii) progress in ensuring closer coordination of economic policies and sustained convergence of economic performance of the Member States in accordance with the TFEU; and (iii) the contribution of this Regulation to the achievement of the Union’s strategy for growth and jobs.

discretion to the Commission in others 94 . The original purpose of the procedure was to be a first step in the strengthened surveillance spectrum to lead a euro area Member State to implement measures aimed at solving accumulated imbalances and possibly avoid the next step in terms of strengthened surveillance, which is a macroeconomic adjustment programme. Enhanced surveillance also serves to avoid situations where a euro area Member State delays the request for a financial assistance programme and avoids taking measures to address its imbalances until the adjustment costs become very high. The enhanced surveillance procedure has only been applied once (in the case of Greece and only after the end of a financial assistance programme rather than prior to a programme). In practice, however, Regulation No 472/2013 i allows for a certain degree of flexibility when deciding to place a vulnerable euro area Member State from normal surveillance into enhanced surveillance.

Increased national ownership

As regards economic adjustment programmes, the appropriate involvement of stakeholders in the Member State is crucial to ensure national ownership, particularly in the transition from a programme back into normal surveillance. Regulation No 472/2013 i establishes specific requirements as to the involvement of social partners and civil society when preparing a macroeconomic adjustment programme, with a view to building consensus over its content. That requirement was broadly met in all the financial assistance programmes. However, both in the case of Cyprus and Portugal, the programmes ended without the completion of the last review and the disbursement of the last tranche of financial assistance. Overall, those experiences suggest that national ownership of the reforms implemented in the programmes tends to decrease at the end.

Intensity of post-programme surveillance

The experience of the recent years has led to a de facto modulation of the intensity of the surveillance based on the needs of the euro area Member State in question. This was done through the length of the post-programme surveillance missions to the Member States which varied from four working days to seven working days, as well as focusing the scope of surveillance. Regulation No 472/2013 i allows for such flexibility as it requires that the Commission conducts regular review missions with a view to identifying factors of importance to the repayment capacity of the Member State concerned and every six months communicates its assessment to the competent committee of the European Parliament, the EFC and the parliament of the Member State concerned.

94 Regulation No 472/2013 i provides that in case a euro area Member State requests precautionary financial assistance from the ESM with the exception of a Precautionary Conditioned Credit Line that is not drawn, the EFSM, the EFSF, one or several other Member States or third countries or another relevant international financial institution such as the IMF, a Member State should be placed in enhanced surveillance. If the euro area Member State does not request precautionary financial assistance, then it may be placed in enhanced surveillance if it is experiencing or threatened with serious difficulties with respect to its financial stability which are likely to have adverse effects on other euro area Member States. This is a discretionary decision of the Commission based on its expert judgement of when a euro area Member State would find itself in such a situation.

5 O VERALL FINDINGS

This review reveals strengths as well as possible areas for improvement. It confirms that the six-pack and two-pack reform, together with the roll out of the European Semester, have strengthened the framework for economic surveillance in the EU and euro area and guided Member States in achieving their economic and fiscal policy objectives. They have also led to a broader and more integrated approach to surveillance that better assures the overall consistency of policy advice within the European Semester. The establishment of a common budgetary timeline and the policy guidance issued on the basis of Member States’ draft budgetary plans has led to closer coordination within the euro area. The measures introduced by the six-pack and two-pack reforms thus have contributed to ensuring closer coordination of economic policies.

The implementation of the recommended policies by Member States has contributed, among other factors, to the gradual strengthening of EU economies and strong job creation recorded in recent years. Furthermore, by supporting the correction of existing macroeconomic imbalances, the prevention of the build-up of new imbalances and the reduction of government debt levels, the surveillance framework has helped to create the conditions for sustainable growth, strengthened resilience, reduced vulnerabilities and lowered the risk of potentially harmful spillovers. The six-pack and two-pack reforms have thus contributed to achieving the Union’s strategy for growth and jobs.

The strengthened surveillance framework has also fostered convergence in the economic performance of Member States, with an overall return to economic growth and declining unemployment rates in all Member States, reduced macroeconomic imbalances, and falling public deficits and debt levels, with all Member States having exited the excessive deficit procedure. Furthermore, the framework put in place to support euro-area Member States experiencing, or threatened with, difficulties with respect to financial stability contributed to reducing the temporary divergence of the Member States concerned from the rest of the euro area and laid the ground for renewed convergence.

The recovery since the financial crisis has been long by historical standards, but there has also been a decline in trend growth, accompanied by persistently low inflation, while public debt levels remain high in some Member States. Those Member States are also far from their medium-term budgetary objectives while their reform efforts are waning. Some Member States’ economies remain vulnerable to an economic slowdown with risks of spillovers that would affect the functioning of the euro area as a whole.

The fiscal stance at Member State-level has frequently been pro-cyclical. Moreover, the composition of public finances has not become more growth-friendly and national governments have revealed their preference for increasing current expenditure rather than protecting investment. In the event of large shocks that are not policy-induced, the ability to steer the fiscal stance for the euro area as a whole, with the appropriate differentiated fiscal effort among Member States, is hampered by the lack of prudent policies in good times and remains constrained as long as it rests exclusively on coordination of national fiscal policies in the absence of a central stabilisation capacity. The current framework and its implementation thus has not fostered macroeconomic stabilisation.

Furthermore, the current surveillance framework and its implementation did not ensure a sufficient differentiation between Member States that have markedly different fiscal positions, sustainability risks or other vulnerabilities. The interplay between Union fiscal rules and national fiscal frameworks is another area of improvement.

The fiscal framework (which includes the secondary legislation and other documents that provide more details and transparency on how surveillance is carried out in practice) has grown excessively complex. This complexity results from the framework pursuing multiple objectives and the need to cater for a wide variety of evolving circumstances, including by the use of flexibility, in a context of divergences of views among Member States. It is reflected in a very detailed codification, encompassing several operational indicators of which a number are non-observable and frequently revised, as well as a variety of escape clauses. As a result, the fiscal rules have become less transparent, hampering predictability, communication and political buy-in.

Other imbalances accumulated during the economic crisis are receding, but in many instances only slowly. The MIP has been more successful in reducing current account deficits than it has been in reducing persistent and large current account surpluses. While this may be explained by the more pressing needs to correct large current account deficits given the concerns about sustainability, persistent current account surpluses can also affect the smooth functioning of the euro area. The traction of policy recommendations is suboptimal and has been declining over time as the momentum for reform has faded. Moreover, while the interaction between the specific surveillance strands has been adequate, there is further scope to make them work better together. Important links between individual surveillance instruments are not sufficiently taken into account, in particular where public debt sustainability issues are intrinsically linked to wider macroeconomic imbalances or low potential growth. 95

Finally, thought should be given to the need for the surveillance framework to help tackling today and tomorrow’s pressing economic, demographic and environmental challenges. Currently, one of the key economic challenges is that growth may remain subdued for a longer period unless policy action is taken to boost potential growth. With an already highly accommodative monetary policy, it is important to consider whether and how Member States’ fiscal and structural policies could contribute to the policy mix in the euro area and to raising potential growth, and to reflect on the potential role of the EU economic governance framework therein. Furthermore, it should be considered to what extent the framework can support economic, environmental and social policy needs related to the transition towards a climate-neutral, resource efficient and digital European economy, complementing the key role of the regulatory environment and structural reforms.

95 In addition, the surveillance of sovereign-bank spillovers has not played a central role within the framework, although their relevance for macroeconomic and financial stability remains.

A NNEX 1: A SSESSING THE IMPACT OF THE SGP: EMPIRICAL EVIDENCE

A.

Relationship between fiscal rules and sustainability

There is ample of empirical evidence showing that Member States with sound fiscal rules

have, on average, lower debt ratios compared to countries without rules. 96 Still, causality is

difficult to establish for endogeneity reasons. Having or adopting a fiscal rule indeed depends on a range of factors that can correlate with fiscal performance. For instance, countries with fiscal rules may have a clear preference for a prudent conduct of fiscal policy whether or not a rule is in place. 97 In addition, evidence on the performance of spending rules used at the EU and Member States’ level was very scarce so far.

A recent counterfactual simulation shows that public debt ratios would have been significantly lower today, in particular in high-debt Member States, if the expenditure

benchmark had been applied and complied with since 2000 (Graph A1.1, green line). 98 Strict

compliance with the expenditure benchmark compared with the structural balance adjustment would have required a larger fiscal adjustment in good times, which would have resulted in higher fiscal buffers at the start of recession. As a consequence, the expenditure benchmarkbased adjustment would have led to a more growth-friendly adjustment, given the lower average loss in output.

Graph A1.1: Debt developments in case of strict implementation of the preventive arm (EU 28, 1999-2019)

Note: The black line refers to the baseline in line with the Commission Spring Forecast 2019. The impact of a strict compliance with the preventive arm are shown using two counterfactual scenarios, namely compliance with the expenditure benchmark (green line C EB ) and the structural balance requirement (red line C SB ).

B.

Relationship between fiscal policy and the economic cycle

Existing evidence points to a rather pro-cyclical pattern of discretionary fiscal policy in the EU. Studies typically find that total fiscal policy (i.e. discretionary fiscal policy and automatic

96 See IMF (2009), Heinemann et al. (2018), Tapsoba (2012), Debrun et al. (2008), Caselli et al. (2018).

97 Poterba (1996).

98 European Commission (forthcoming), ‘Performance of spending rules at the EU and national level’, Report on Public Finances in EMU 2019.

stabilisers) tend to be a-cyclical or countercyclical, 99 while discretionary fiscal policy appears to be pro-cyclical (Table 1.1). 100 Pro-cyclicality happens in particular in good economic times. 101 In Europe, most Member States implemented sizeable fiscal adjustments in the

immediate post-crisis period, reflecting a lack of fiscal buffers at the onset of the economic and financial crisis, a need to correct excessive deficits and, in some cases, market pressure. The role of the reinforced EU fiscal rules on cyclicality has only scarcely been investigated.

Table A1.1.:Empirical findings on cyclicality

Dependent variable: Dependent variable:

∆ Structural prim. balance EB-based fiscal effort

(1)

(2) (3) (4) (1) (2) (3) (4)

Fiscal effort (t-1) 0.128* 0.08 0.079 0.074 0.288** 0.307** 0.309** 0.261*

(respectively ∆SPB and EB) (1.758) (1.226) (1.158) (1.135) (1.978) (2.357) (2.355) (1.890)

∆ Output gap (t) -0.321*** -0.370*** -0.371*** -0.369*** 0.754*** 0.892*** 0.869*** 0.791**

(-3.756) (-5.190) (-5.093) (-4.730) (2.765) (2.908) (2.847) (2.166)

Public debt (t-1) 0.006*** 0.011*** 0.011*** 0.011*** -0.019** -0.036*** -0.036*** -0.045***

(3.529) (3.804) (3.209) (2.897) (-2.485) (-2.874) (-2.754) (-3.530)

Current account (t-1) 0.108*** 0.114*** 0.112*** -0.198 -0.198 -0.087*

(3.315) (3.508) (3.487) (-1.265) (-1.252) (-1.973)

Age dependency ratio (t-1) -0.074*** -0.076** -0.103** 0.244* 0.249* 0.211**

(-3.332) (-2.440) (-2.584) (1.664) (1.702) (2.139)

Election year (t-1) -0.003** -0.003** 0.011** 0.014***

(-2.106) (-1.974) (2.436) (3.388)

Crisis dummy 2008-09 -1.584** 1.396*

(-2.102) (1.948)

# observations 437 427 427 427 347 340 340 340

# countries 28 28 28 28 27 27 27 27

R-squared

Wald test time/country dummies 0.00 0.00 0.00 0.00 0 0 0 0

AR(1) (p-value) 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

AR(2) (p-value) 0.31 0.22 0.28 0.29 0.84 0.84 0.83 0.90

Hansen (p-value) 0.29 0.83 0.78 0.77 0.52 0.58 0.57 0.68

# instruments 25 29 30 30 22 26 27 28

Note: The sample includes the 28 Member States covering the period 2000-18. All estimations are based on real time data from the COM autumn forecast vintages. Regressions are estimated based on the two-step system GMM estimator following Blundell and Bond (1998), controlling for endogeneity of the lagged dependent variable, output gap and current account. ***, ** and * denote respectively statistical significance at 1, 5 and 10%.

Source: European Commission (2019), ’Have EU fiscal rules mitigated pro-cyclicality?’, Report on Public Finances in EMU 2018, 121-130.

New evidence by the Commission shows that that compliance with EU fiscal rules reduces

the pro-cyclicality of the fiscal effort. 102 The analysis assesses pro-cyclicality for Member

States that conducted fiscal policy in line with the preventive and corrective arm of the SGP.

99 Fatas, and Mihov 2009, Benetrix and Lane 2013, Afonso and Hauptmeier 2009, Baldi and Staehr 2015.

Woo 2009, Checherita-Westphal and Žďárek (2017), Baldi and Staehr (2016); The findings by Eyraud et al.

(2017) indicate acylical fiscal policy based on Member States plans, but pro-cyclical fiscal policy based on realtime and ex-post data.

101 European Commission (2018), ‘Fiscal outcomes in the EU in a rules-based framework – new evidence’,

Report on Public Finances in EMU 2018, 105-156, Eyraud, Gaspar and Poghosyan (2017), Fiscal politics in the Euro area. IMF Working Paper 17/18.

102 European Commission (2018), ‘Fiscal outcomes in the EU in a rules-based framework – new evidence’,

Report on Public Finances in EMU 2018, 105-156,

For that purpose, real-time data is used from past Commission forecast vintages for a sample of Member States since 2000. The results show that Member States who complied with the expenditure benchmark and/or the structural balance requirement exhibit on average a lower pro-cyclical fiscal effort. In addition, Member States with public debt ratios below 60% of GDP have on average a smaller pro-cyclical fiscal effort than Member States with a debt ratio above 60% of GDP. Conversely, pro-cyclicality appears higher for Member States with debt ratios exceeding 60% of GDP.

As regards national fiscal rules, fiscal policy is found to be least pro-cyclical when

expenditure rules operate in combination with budget balance rules. 103 Well-functioning fiscal

frameworks typically consist of several fiscal rules targeting different budgetary aggregates and with different time horizons, thus reinforcing each other. Results suggest that the combination of budget balance rules and expenditure rules provides for the least pro-cyclical fiscal policy, namely an a-cyclical fiscal policy (Graph A1.2). In the absence of both expenditure and budget balance rules, fiscal policy would have been more pro-cyclical, while pro-cyclical bias declines only marginally if expenditure rules operate in the absence of budget balance rules.

Graph A1.2. Pro-cyclical bias for different combinations of fiscal rules

Source: Stability and Convergence Programmes, AMECO spring vintage and Commission’s Fiscal Governance Database 2016 vintage.

Note: The graph shows the pro-cyclical bias, i.e. by how much expenditure increases/decreases to a 1pp of GDP unexpected revenue shortfall/windfall in different combination of rules. ER stands for expenditure rule while BBR stands for budget balance rule. Four combinations of rules are shown: no expenditure or budget balance rules, only expenditure rules, only budget balance rules and both expenditure and budget balance rules. The bars indicate the 95% confidence interval.

C.

Relationship between fiscal rules and public investment

There is a vivid debate on the relationship between fiscal rules and public investment. Evidence from the early years of Economic and Monetary Union finds no clear impact of EU fiscal rules on public investment. 104 However, the evidence can be challenged by the short sample period. Recent studies point to a more nuanced picture. Member States’ investment

103 European Commission (forthcoming), ‘Performance of spending rules at the EU and national level’, Report on Public Finances in EMU 2019.

104 Galí and Perotti 2003, Turrini 2004, Perée and Välilä 2005, Heinemann 2006, Mehrotra and Välilä 2006.

decisions appear to be more constrained by high debt than by the SGP. 105 Similarly, fiscal rules and governance quality appear to reduce the negative impact of public debt on public

investment. 106

Commission analysis provides new evidence on the relationship between fiscal rules and public investment. The key objective of that analysis was to run a positive analysis supported

by a very large number of robustness checks. 107 Overall, the study does not find evidence that

fiscal rules hamper public investment. This finding holds irrespective of the analysed duration of the impact (short vs. long term), the type of investment (total investment, investment in infrastructure, redistribution), the level of government (general, national and sub-national government level) and the type of fiscal rules (balanced budget, expenditure, debt and revenue rules) (Graph A1.3).

Graph A1.3: Summary: significance levels of regressions (in percentage points)

Note: The chart summarises the main findings of the panel regressions on the impact of fiscal rules on public investment for a sample of EU Member States over the time period 1990-2018. It shows the percentage share fiscal rules turned out to be significant and non-significant based on more than 10000 robustness tests. Reading example: Taking the full sample shows that 87 percent of the regressions suggested that fiscal rules have no statistically significant impact on public investment. Only in 13 percent of the regressions fiscal rules turned out to be significant, with an almost equal split on the direction of fiscal rules on public investment (6% positive impact of fiscal rules on public investment, 7% negative and significant impact).

Box A1: Possible implications for the policy mix of current trends in debt reduction

across the euro area

Fiscal discipline is a crucial component of the economic and monetary union framework, as responsible fiscal policies are key for the macroeconomic stability and long-term growth potential of its Member States. In particular, in a monetary union fiscal policy might be relied

105 Bacchiocchi et al. 2011.

106 European Commission 2017, Government investment in the EU : the role of institutional factors, Report on Public Finances in Emu 2017, 133-186.

107 Overall, the study is based on more than 25000 regressions.

on to address country-specific shocks in the absence of national monetary and exchange rate policies. From a monetary policy perspective, excessive indebtedness carries the risk of fiscal dominance, i.e. a situation in which the central bank is restricted in the pursuit of its price stability objective by the potential impact of a policy tightening on the sovereign’s solvency. In light of this, the Treaty on the Functioning of the European Union includes strong clauses

on central bank independence and the prohibition of monetary financing. 108 Furthermore, a

Member State’s over-indebtedness might increase its banks’ financing costs via the sovereignbank nexus and ultimately distort the transmission of the ECB’s single monetary policy across Member States. That being said, a sustainable volume of government debt is necessary to ensure the smooth implementation of monetary policy, which partly relies on the presence of a sufficient amount of safe assets.

Government debt currently with the highest rating in the euro area is set to rapidly decrease in the next decade. After a period of steep increase following the Great Financial Crisis, government debt-to-GDP ratios have gradually decreased in most Member States since 2014, albeit at different pace across Member States (see Table A). In 2019, 9 out of 19 euro-area Member States would have government debt ratios below the Treaty reference value of 60% of GDP. Assuming full implementation of Member States’ fiscal plans, that overall downtrend trend should continue in the next decade. Debt projections for countries such as Germany and the Netherlands in particular show a reduction over that horizon (see Graph A), notably on the back of a continuing favourable differential between interest and growth rates. In those two Member States, government debt-to-GDP ratios are projected to fall to around 45% and 40% of GDP respectively by 2029. In that context, the amount of government debt

rated AAA, which has already declined markedly, 109 is expected to continue to do so over the

coming years (see Graph B). At the same time, in some high-debt Member States without AAA rating (such as IT and FR), debt burdens are set to decrease considerably less in the next decade, remaining at values (well) above 90% of GDP.

Table A and Graph A: Government debt-to-GDP ratio, historical and projected values

in the euro area (Stability and Convergence Programmes’ scenario)

% of GDP 2007 2014 2019 2029 Var. Var.

2014/19 2019/29 120 Projections

BE 87.0 107.5 100.6 84.7 -6.9 -15.9 DE 63.7 75.3 58.6 45.2 -16.6 -13.5

EE 3.7 10.5 8.2 15.8 -2.3 7.6 100

IE 23.9 104.1 61.1 34.5 -43.0 -26.6

ES 35.6 100.4 95.8 85.4 -4.5 -10.4 80

FR 64.5 94.9 98.9 95.7 4.0 -3.2

IT 99.8 131.8 132.6 126.1 0.8 -6.5

CY 54.0 108.0 95.7 56.9 -12.3 -38.8 60

LV 8.0 40.9 37.4 27.6 -3.5 -9.8

LT 15.9 40.5 37.0 30.1 -3.5 -6.9 40

LU 7.7 22.7 20.2 7.1 -2.5 -13.1 MT 62.3 63.4 42.7 15.0 -20.6 -27.7

NL 43.0 67.9 49.1 39.8 -18.8 -9.3 20

AT 65.0 84.0 69.6 50.4 -14.4 -19.2

PT 68.4 130.6 118.6 86.9 -12.0 -31.7 0 SI 22.8 80.4 65.4 39.9 -15.0 -25.5 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 SK 30.1 53.5 47.5 34.3 -6.0 -13.2 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20

FI 34.0 60.2 58.1 55.4 -2.1 -2.7

EA-19 65.0 94.4 85.5 73.7 -8.9 -11.9 DE NL Other EA

Source: European Commission based on projections according to the ‘Stability and Convergence Programmes’ scenario (see 2018 Fiscal Sustainability Report for more explanations on the methodology)

108 In particular Articles 123, 124 and 130 TFEU. Of course, fiscal dominance can arise from governments’ overindebtedness irrespective of monetary financing.

109 First, as a result of several downgrades following the crisis, then, as a result of the reduction of debt to GDP ratios in sovereigns rated AAA.

Graph B: Government debt in the euro area, per credit rating (Stability and

Convergence Programmes’ scenario)

% of euro area GDP

100% Projections

90% 80% 70% 60% 50% 40%

30% 20% 10% 0% 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q 07 08 09 10 12 13 14 15 17 18 19 20 22 23 24 25 27 28 29 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20

AAA Total EA

Note: for the projections, countries considered rated AAA are those rated at this grade end 2019 (i.e. DE, LU and NL).

Source: Bloomberg, European Commission

While declining supply of safe assets 110 currently does not seem to pose a problem for monetary policy implementation and transmission in the euro area, a hypothetical severe contraction could have negative consequences. For example, the ECB’s Extended Asset Purchase Programme so far relied to a large part on sovereign debt purchases, which were conducted according to the ECB’s capital key and were subject to an issuer limit of 33% applied to all consolidated Eurosystem holdings. A significant reduction in outstanding sovereign debt of individual Member States would thus restrict the scope of any future purchase programme 111 with similar parameters from the onset. While the ECB could react to a severe reduction in outstanding sovereign debt by selected Member States in a possible future asset purchase programme by focusing on other asset classes such as corporate bonds, this could come at the cost of higher Eurosystem risk-taking and market distortions in those relatively less deep markets. Compared with the euro area sovereign bond markets, the pool of outstanding euro area corporate bonds is significantly smaller in volume, overall riskier and

stronger focused on short- to medium-term maturities. 112

A second concern relates to the role of safe sovereign bonds as collateral in central bank liquidity-providing operations and secured money market transactions. While central government securities made up around half of the universe of assets eligible as collateral in Eurosystem credit operations in 2019Q3, their share in actually pledged collateral was only

110 A safe asset can be understood as any asset with low default risk or risk of loss of value, independent of cyclical conditions.

111 The ECB has repeatedly indicated that it considers asset purchases a permanent feature of its monetary policy toolbox going forward.

112 As of early January 2020, investment-grade EMU non-financial corporate bonds were estimated to amount to just 25% of the outstanding amount of investment-grade EMU sovereign bonds. Among the respective investment-grade bonds, only 8.4% were rated AA or higher for non-financial corporate issuers (sovereign bonds: 57%), while 59% were rated BBB (sovereign bonds: 25.7%). The average remaining maturity of investment-grade corporate bonds was 5.6 years, compared to 8.3 years for investment-grade rated euro area sovereign bonds (Source: Bloomberg).

14% and has been decreasing since the start of the Public Sector Purchase Programme. By contrast, government bonds play a dominant role in repo markets, 113 and consequently, a severe reduction in outstanding safe assets (triple-A in this case) could impact repo market pricing and impair repo market functioning. Already following the launch of the PSPP, tensions occurred in collateral markets as reflected in both widening repo premia on special collateral (SC) markets – in particular for German government bonds – and also in a marked divergence between core and peripheral euro area countries’ national General Collateral (GC) rates 114 . Although the ECB’s securities lending programme 115 has alleviated those concerns to some extent and has led to a decrease in repo premia, 116 this suggests some dispersion of

money market funding conditions across the euro area and thus could signal a potentially uneven transmission of monetary policy. One explanation for diverging rates on national GC repo baskets that has been brought forward is that at the current juncture, country-specific GC rates are not completely cash-driven when all bonds from a given country have value as special collateral. 117 In other words, in an environment of safe asset scarcity, increased bond specialness 118 might spill over to GC repo funding. Therefore, a further significant reduction of outstanding safe government bonds could resuscitate tensions in SC repo markets and reinforce existing divergences in GC repo spreads across Member States, which might ultimately manifest itself in steady differences in lending and borrowing conditions at the later stages of the monetary policy transmission mechanism. Safe asset scarcity and its effects on repo markets could gain particular relevance in situations of heightened uncertainty and market stress, with potential financial stability implications. What is more, financial stability could be adversely affected in the long run with financial markets relying on other assets to play the role of the safe asset.

113 According to the International Capital Market Association’s (ICMA) June 2019 European repo market survey, EU central government bonds accounted for two-thirds of pledged collateral in European repo transactions, with German central government bonds accounting for 15%. 114 Besides euro area-wide General Collateral (e.g. GC pooling) baskets, several national GC baskets exist, i.e. baskets containing only sovereign bonds from the respective country.

115 Under the securities lending programme, the Eurosystem makes securities purchased under the PSPP available for lending. The programme was enhanced in December 2016 by allowing Eurosystem central banks to also accept cash as collateral in the lending process.

116 See Brand, C., Ferrante, L. and A. Hubert (2019). “From cash- to securities-driven euro area repo markets: the role of financial stress and safe asset scarcity”. ECB Working paper No 2232, January 2019. 117 See, e.g. Arrata, W. Nguyen, Rahmouni-Rousseau, I. and M. Vari (2018). “The Scarcity Effect of Quantitative Easing on Repo Rates: Evidence from the Euro Area”. IMF Working paper 18/258. 118 The International Capital Market Association defines specialness as the case when the repo rate quoted on a particular security falls at least about 10 basis points below the GC repo rate for that currency.

A NNEX 2: A SSESSING THE IMPACT OF THE MIP ON POLICY PROGRESS : AN

EMPIRICAL ASSESSMENT

Assessing to what extent MIP surveillance strengthens policy progress requires controlling for other factors that may play a role in driving reform outcomes. Empirical work was performed to this purpose in the MIP Compendium. Those results showed that, controlling for other

factors, the MIP did have a positive impact on policy progress. 119 The analysis below updates

and extends that earlier analysis to cover the CSRs issued over the period 2013-2018 and to take into account additional factors that may have had an impact on policy progress. 120

In Table A2.1, columns (3) and (4) display results corresponding to the specification found in the MIP Compendium estimated to a sample extended until CSRs issued in year 2018. Some additional explanatory variables are tested in columns (1) and (2) of Table A2.1: with a view to capture potential reform fatigue, past progress and the number of sub-CSRs cumulated until the year before have been added as regressors. More precisely, for each year, Member States are classified by quartiles, depending on the sum of progress over all their sub-CSRs and the number of sub-CSRs, higher quartiles corresponding to more progress and more sub-CSRs respectively. This treatment helps comparability over time, enabling to control for the streamlining of the number of CSRs and sub-CSRs around 2015, which appears in the second part of the period now under review.

A number of findings emerge from the analysis.

• In all cases, over the whole period, the MIP imbalance category has a significant and

positive impact on progress, at the 5% significance level.

• Recommendations that a priori are hard-to-comply with, have a negative and significant

impact on the overall compliance rate at the 1% level (columns (1 ) to (4 ) ).

• GDP growth normally has a positive sign, implying that growth helps reform compliance

by raising the available political capital. The interest rate spread variable has also normally a positive effect, linked to market pressure. Neither variable however reach

statistical significance.

• Parliamentary elections, notably if to come by the subsequent year, have a strong and

significant negative impact as reforms are perceived to be potentially costly in electoral terms.

• The cumulated number of past sub-CSRs plays a negative and significant role at the 1%

level, implying that CSR progress tend to be lower the higher the cumulated number of previous recommendations, consistently with the idea of “reform fatigue”.

• At the same time, it appears that reform progress can unfold only gradually, as reform

progress depends positively past progress. This is confirmed by the coefficient of the variable "cumulated progress until previous year" which is significant at the 1% level and

positive.

119 The control variables were as follows: a dummy indicating whether the recommendation is “hard-to-comply” with because politically costly, GDP growth, the sovereign spread with respect to the benchmark German 10- year interest rate on government bonds, dummies indicating legislative elections at most one year before or by next year, and dummies indicating the main policy area of the sub-CSR (public finances and taxation; financial sector; labour market, education & social policies; structural policies; public administration and business environment). The MIP imbalances category used in the MIP Compendium were more articulated than those used in the present exercise.

120 Progress is assessed at sub-CSR level and on the year following the issuance of the CSR. The size of the sample is defined by the year of issuance of CSRs, namely to all CSRs issues between 2013 and 2018, both dates included.

• Once reform fatigue effects are taken into account, the impact of MIP surveillance on

policy progress appears stronger. This is consistent with a stronger reform fatigue in MIP countries, which needs to be taken into account to fully appreciate the impact of MIP

surveillance on CSR compliance.

Table A2.1: MIP surveillance and CSR progress, data at detailed recommendation level, all Member States (except those under adjustment programme)

(1)

(2) (3) (4)

Dependent variable: CSR compliance Ordered OLS Ordered OLS score PROBIT PROBIT

MIP imbalance category 0.222*** 0.170** 0.184*** 0.133**

(0.0762) (0.0672) (0.0673) (0.0616)

GDP growth -0.0133 -0.0452 0.00525 -0.0230

(0.0300) (0.0335) (0.0322) (0.0351)

Cumulated progress until previous year 0.0842*** 0.0733***

(0.0213) (0.0216)

Cumulated number of sub-CSRs until -0.0670*** -0.0629*** previous year

(0.0166) (0.0180)

Hard-to-comply with recommendation, -0.232*** -0.320*** -0.211*** -0.296*** dummy

(0.0705) (0.0885) (0.0673) (0.0857)

Interest rate spread in previous year 0.0158 0.0220 -0.0148 -0.00901

(0.0309) (0.0344) (0.0350) (0.0388)

Elections in previous year -0.0583 -0.0892 -0.125 -0.148

(0.0838) (0.0886) (0.0881) (0.0932)

Elections by the following year -0.187** -0.192** -0.189* -0.201**

(0.0920) (0.0926) (0.101) (0.101)

Constant 3.507*** 3.558***

(0.273) (0.271)

Observations 1,548 1,548 1,580 1,580

R-squared 0.061 0.043

Year F.E Yes Yes Yes Yes

Control for policy area Yes Yes Yes Yes

Clustering at country-year level Yes Yes Yes Yes

Note:*, ** and ***: coefficients significant at the 10%, 5% and 1% level respectively. Student-t are reported in parenthesis. Standard errors are robust with-respect to clustering at the country-year level. Dependent variable: CSR compliance score defined at disaggregated policy field. 0=no progress; 25=limited progress; 50=some progress; 75=substantial progress; 100=full achievement. The scores are transformed as log(1+score) to smooth the effect of the retained scaling.

MIP categories: no imbalances=0; imbalances=1; excessive imbalances=2. The MIP categorisation refers to the preceding year. The correspondence between the classifications in MIP categories used in 2013 and 2014 is the one displayed in table 4.3 of the MIP Compendium, with the two categories qualified as being excessive imbalances in 2013 corresponding to excessive imbalances category of 2014. The "hard-to-comply with recommendations" variable is a dummy variable taking value 1 for sub-CSRs belonging to the fields listed in the note to Table A.3.1. in the MIP Compendium. Interest rate spreads are expressed as the difference of the 10-year government bond yield with the one of Germany.

"Elections in previous year" and "Elections in following year" are dummies equal to one if legislative elections have taken place at most one year before the related year or if these elections are to take place in at most by one year, respectively.

The variables on cumulated progress and number of sub-CSRs are based on classifications by quartiles, depending on the sum of progress over all sub-CSRs and the number of sub-CSRs, higher quartiles corresponding respectively to more progress and more sub-CSRs.

Source: European Commission CeSAR database, DG ECFIN AMECO database, Eurostat, World Bank political database.

Additional regressions are displayed in Table A2.2, which distinguishes by sub-periods. Moreover, additional specifications attempt to address reverse causation, i.e. the fact that policy progress affects the extent of MIP surveillance. To control for this issue, instrumentation is performed in columns (3) and (4) of Table A2.2, over the whole period and the last two years respectively. Instruments correspond to the synthetic indicators of economic conditions in stocks and flows, used in Annex 2 of the MIP Compendium, taking into account sources of imbalances identified in in-depth reviews.

Results in Table A2.2 show that the additional impact of MIP surveillance on policy progress seems to vanish in the sub-period 2016-2018 (column (2)). However, this appears to be partly related to a reverse causation effect, namely, to the fact that policy progress also reduces the likelihood for a country to remain under MIP surveillance. Once this effect is taken into account by means of instrumental variables estimation, the magnitude of the impact of MIP surveillance rises over the whole period and also for the 2016-2018 sub-period, although not reaching statistical significance.

Table A2: MIP surveillance and CSR progress, data at detailed recommendation level, all Member States by sub-periods and taking into account reverse causation (except those under adjustment programme)

(1)

(2) (3) (4)

Dependent variable: CSR compliance Ordered Ordered IV Regress IV Regress score PROBIT PROBIT [2013:2018] [2016:2018] [2013:2015] [2016:2018]

MIP imbalance category 0.252*** 0.0895 0.331*** 0.196

(0.0934) (0.114) (0.0862) (0.216)

GDP growth 0.00724 -0.0714 -0.0179 -0.0555

(0.0337) (0.0582) (0.0329) (0.0602)

Cumulated progress until previous year 0.101*** 0.0660** 0.0719*** 0.0662**

(0.0280) (0.0306) (0.0223) (0.0260)

Cumulated number of sub-CSRs until -0.0538** -0.0709*** -0.0717*** -0.0793*** previous year

(0.0233) (0.0273) (0.0199) (0.0304) Hard-to-comply with recommendation, -0.179** -0.337*** -0.329*** -0.452*** dummy

(0.0870) (0.124) (0.0874) (0.137)

Interest rate spread in previous year 0.0112 0.0310 0.0156 0.0382

(0.0346) (0.0616) (0.0380) (0.0603)

Elections in previous year -0.0998 0.116 -0.104 -0.0320

(0.114) (0.136) (0.0898) (0.133)

Elections by following year -0.218* -0.00900 -0.238** -0.0884

(0.126) (0.163) (0.0990) (0.166)

Constant 3.469*** 3.087***

(0.274) (0.381)

Observations 1,029 500 1,548 500

Year F.E Yes Yes Yes Yes

Control for policy area Yes Yes Yes Yes

Clustering at country-year level Yes Yes Yes Yes

Sample countries All All All All

R-squared 0.056 0.097

Fisher coefficient 62.17 14.47

Note:*, ** and ***: coefficients significant at the 10%, 5% and 1% level respectively. Student-t are reported in parenthesis.

Standard errors are robust with-respect to clustering at the country-year level.

For the definition of variables see notes to Table A1.

Source: European Commission, Eurostat, World Bank political database, national sources, own calculations.

original article: 'COMMISSION STAFF WORKING DOCUMENT Report on the ap...'

3. Referenced document

7 Feb
'20

COMMUNICATION FROM THE COMMISSION Economic governance review Report on the application of Regulations (EU) No 1173/2011, 1174/2011, 1175/2011, 1176/2011, 1177/2011, 472/2013 and 473/2013 and on the suitability of Council Directive 2011/85/EU
COVER NOTE

Secretary-General of the European Commission
5817/20

4. EU Monitor

The EU Monitor enables its users to keep track of the European process of lawmaking, focusing on the relevant dossiers. It automatically signals developments in your chosen topics of interest. Apologies to unregistered users, we can no longer add new users.This service will discontinue in the near future.