New governance and the displacement of Social Europe: the case of the European Semester

Abstract

Has the European Semester led to a displacement of Social Europe, or to the development of social policy through fiscal processes and actors? – Potential for Semester to increase soft law’s binding effects or ‘socialise’ EU policy-making – Positive effects severely limited by the Semester’s overall goals: fiscal stabilisation and the creation of increasingly uniform economic policies – Dilemma for Social Europe: how can an autonomous EU social policy be (re) established without risking marginalisation?

Introduction

Policy coordination was once the future of social Europe. For one commentator, Martin Rhodes, the 2000 Lisbon Strategy, which had instruments of open coordination at its heart, had the potential to become Europe’s ‘Maastricht for Welfare’: a turning point in the historic imbalance in the Treaties between market and non-market objectives. ¹ Much of that hope rested on overcoming competence ‘firewalls’ established by the Treaty. By acting in areas of limited EU competence, open coordination was also seen as an instrument able to build more autonomous EU social policy, i.e. one elaborated and defended on its own terms, without the constant need to ‘piggy-back’ on market integration.

What happened to this optimism? As open coordination developed – as will be discussed below – the terms of this debate changed significantly. To be effective, or to concentrate political attention, social policy coordination has often been held in uneasy tension with other coordination processes operating in the economic field. The zenith of this process has been the European Semester: the EU’s new over-arching cycle of policy coordination. ²

The central purpose of this article will be to examine how the Semester can be situated in the debate over ‘displacement’ of social policy. As the article’s third section will explore, some argue that the embedding of social policy coordination within the Semester offers distinct advantages for EU social actors. Some scholars have argued that, owing to the advent of more coercive regulatory instruments and novel institutions such as the Social Protection Committee, the Semester is undergoing a process of gradual ‘socialisation’. ³ Such a thesis casts a positive spin on the embedding of social policy coordination in the broader coordination framework the European Semester establishes.

As the article’s fourth section will argue, the claim of socialisation should be treated with caution. While social actors may indeed be fighting valiantly for a more socially conscious Semester, they do so in an institutional and policy-making structure that tends to displace social policy, or reduce it to its fiscal impacts. The evidence from the Semester’s earlier years suggests that the Semester continues rather than challenges the gradual erosion of the autonomy of social policy coordination observed since the Lisbon Strategy’s inception (while adding a number of new problems such as an increasingly harmonised and diversity-averse form of policy coordination). In this sense, the article explores both of the varieties of displacement discussed in the introduction to this special issue – displacement in the sense of ‘moving elsewhere’ and displacement in the sense of being threatened or hollowed out.

As the article will conclude, the lessons of the Semester place EU social policy at an uneasy fork in the road. A return to autonomy carries the hope of rescuing Social Europe from the unfriendly discourse into which it must now fit; in doing so, however, such a return also presents the risk of marginalisation and political irrelevance. Absent a more radical overhaul of Europe’s Economic and Monetary Union, EU social policy coordination seems to have entered a blind alley, without easy escape routes.

Displacement in new governance: a short history

When considering ‘displacement’ through the lens of new governance and the European Semester, it is worth recalling that the displacement of social Europe has been a more or less constant feature of the new governance debate from its inception. ⁴ What we now understand as ‘new governance’ was a response to a number of questions facing the EU governance of social policy in the late 1990s and early 2000s. One of these questions was constitutional, i.e. at what level should social policy be governed? Whereas some saw social policy as primarily a national competence to be ‘ring-fenced’ from EU intervention, many centre-left governments of the late 90s – conscious of the social and economic inter-dependencies of the Eurozone – saw considerable added value in creating common social policies at the EU level. In this sense, methods like the Open Method of Coordination were a ‘constitutional compromise’, i.e. a way of creating common EU social policies but through a flexible and decentralised process where diversity in national processes of social delivery could be managed and maintained. ⁵

There was, however, also a second, more substantive question at the heart of ‘new governance’. This concerned the nature of EU social intervention (a question mark implicit in EU social policy in a larger sense). Should EU social policy be an autonomous policy field – one that attempts to shield individuals from market forces and to compensate economic globalisation’s ‘losers’ – or should it instead be seen as ancillary and complementary to the EU’s other policy goals, i.e. of re-enforcing open borders and economic competitiveness? In crude terms, should EU social policy primarily condition the market (aiming at a purely social set of goals) or instead aim to facilitate market integration (i.e. to reconcile integration’s social and economic objectives)? ⁶

This substantive issue was not only a source of disagreement between EU states but an issue that the Open Method of Coordination, and the policies pursued through it, attempted to fudge. The answer of the EU’s 2000 ‘Lisbon Strategy’ to the dilemma was to envisage the EU’s political future as three mutually supporting ‘pillars’ of economic competitiveness, better employment and greater social cohesion. ⁷ Lisbon was based on what Maurizio Ferrara has described as the ‘virtuous circle’ of integration. ⁸ Integration in the 21^st century would be a ‘win-win’ for both sides of the political divide. Free and competitive markets would create the sustainable tax base able to support generous welfare policies; meanwhile, smart investment and ‘flexi-curity’ would create the nimble, educated and confident workforce able to sustain future economic growth. Social and economic goals were thus continually touted in this period as mutually enabling, with one supporting the other, and no need at any point for choices, trade-offs or ideological conflict between these two sets of goals. ⁹

The idea of the ‘virtuous circle’ spilled over into institutional design, where a constant question has concerned the extent to which social policy should have a separate coordination process or be coupled to other forms of policy coordination. The original Lisbon strategy fudged this question too. The three pillars were each to have their own autonomous yet mutually reinforcing policy-making processes. The showcase for this development was the Social Open Method of Coordination: a candidate for the most altered and tinkered-with EU policy process in history. ¹⁰ The Social Open Method of Coordination began in 2001 as an open coordination process for social inclusion, with Member States establishing National Action Plans for social inclusion based on common EU goals, and reviewed via a two-year cycle (without an explicit Treaty basis). ¹¹ This was to be the heart of Lisbon’s social pillar next to separate coordination processes for fiscal policy (the so-called ‘Broad Economic Policy Guidelines’) and employment (the European Employment Strategy). ¹²

The story of the Social Open Method of Coordination from the mid-2000s onwards is in many ways a story of the erosion of Lisbon’s original ‘separate but equal’ message. Following the midterm review of the Lisbon Strategy in 2004, social inclusion was excluded from the core of the Lisbon framework. Lisbon’s review highlighted the vast gap between the strategy’s ambitions and its implementation. Lisbon, its critics argued, was about ‘everything and thus nothing’, leading to poor national ownership of its goals, and weak implementation. ¹³ The answer, the review argued, was to focus the strategy on the core priority of ‘jobs and growth’ (not social cohesion). As a result, the Broad Economic Policy Guidelines and European Employment Strategy were merged into a single set of ‘integrated guidelines for jobs and growth’. A revamped Social Open Method of Coordination was to continue but would report separately from the integrated guidelines, with the OMC SPSI (for social protection and social inclusion) ‘feeding in and feeding out’ to the central jobs and growth cycle. ¹⁴ The virtuous circle was to continue, albeit a circle within which social policy was increasingly marginalised.

The process of marginalisation was completed with the renewal of Lisbon in 2010. ¹⁵ At the policy level, the new ‘Lisbon 2020’ strategy seemed to return social inclusion to its central place, naming ‘inclusive growth’ as one of the strategy’s central goals and placing for the first time an overall poverty target (of lifting 20 million Europeans out of poverty) on Lisbon’s agenda. ¹⁶ In reality, the visibility of social goals in the strategy was vastly diminished. Social inclusion was incorporated as one of ten ‘integrated guidelines’ while the separate Open Method of Coordination for social inclusion, and its associated obligation for Member States to develop explicit social inclusion strategies, was abolished altogether. When EU social actors decided to resuscitate a separate Social Open Method of Coordination in 2012 (realising the paucity of attention social goals received in the larger integrated guidelines ¹⁷ ), it was in a diminished form, with a number of Member States simply refusing to engage in social reporting and with an unclear relationship to the broader Lisbon framework. ¹⁸ While some separate explicitly ‘social’ institutions remain – such as the Social Protection Committee, which monitors national social performance – the idea of an autonomous social pillar for the Lisbon strategy seemed in full retreat.

As the next section will discuss, post-2010, social policy coordination is largely conducted under the rubric of Lisbon’s other goals – a new triumvirate of growth, jobs and (increasingly, post-Euro crisis) financial stability. In short, while ‘displacement’ (in the sense of coordinating social policy largely through the lens of other policy goals) was one among many options for the future of EU social policy coordination in the late 1990s, displacement has become more and more prominent as the European Semester has evolved. This historical picture shifts our attention: the relevant question for Social Europe when approaching the European Semester no longer seems to be whether displacement is taking place, but how to normatively assess it.

From new governance to the European Semester: the optimistic case for displacement

If a complete outsider analysed the websites and policy documents of the EU in 2017, he/she would probably conclude that the Lisbon Strategy is dead. It lives on, however, in the European Semester, the EU’s central cycle for economic policy coordination. The shift from the Lisbon Strategy to the framework of the European Semester is central to unpacking whether and how Social Europe may be displaced via EU policy coordination. While Lisbon was ostensibly an overarching strategy for advancing and reconciling different EU objectives, the Semester has a very clear link to fiscal policy. ¹⁹ It was envisaged as a measure to buttress and strengthen the Eurozone economy in particular and to recognise the need for heightened EU supervision of domestic budgets. As a result, its surrounding legal framework, i.e. the ‘six pack’ of measures adopted in 2011 and the later ‘two-pack’ Regulations (472 and 473/2013), is deeply embedded in the Stability and Growth Pact with each of the key stages of the policy cycle – i.e. the initial publication of an Annual Growth Survey by the Commission, the issuing of Country-Specific Recommendations and the cycle of budgetary surveillance – carrying an explicit economic objective. Social policy surveillance is to take place but within a policy cycle whose overarching goal is the need, under Art. 121 and 136 of the Treaty, to coordinate economic policies (and thereby ensure fiscal stability).

Where does this leave Social Europe? Let us start with the optimistic case first. There are a number of features of the Semester system that provide renewed leverage for EU social policy or could even be seen as compensating for historic weaknesses in social policy coordination. One of these concerns is the effectiveness of ‘autonomous’ social policy coordination. Many have questioned whether the ‘constitutional compromise’ mentioned above really worked. While there are many empirical difficulties in establishing whether vague ‘guidelines’ issued at the EU-level are complied with nationally ²⁰ , early evaluations of the social Open Method of Coordination were often critical of its ability to drive domestic policy change. ²¹ Many Member States responded to EU guidelines and reporting requirements merely by ‘listing’ domestic reforms that were already a part of government plans. The temptation – faced with a voluntary policy process – was to comply selectively with EU guidelines, if at all, i.e. to implement those suggestions close to national preferences and reject those with real economic or political costs. ²²

Compared with this voluntarism, the Semester offers a number of advantages. In so far as the Semester process includes social policy recommendations, those recommendations benefit from the tools and incentives the Semester provides. ²³ This starts with the relatively mundane; the social Open Method of Communication never contained for example recommendations that were ‘country-specific’. More generally, there is no specific legal basis for poverty and social inclusion recommendations in the Treaty. While national performance was benchmarked and compared, country specific recommendations offer a tool to highlight particular deficiencies in national social performance. To give some examples from the most recent 2017 round, these recommendations criticise Austria for the attainment gap in education between those with versus those without a migrant background; similarly, Bulgaria is asked to improve the ‘coverage and adequacy of the national minimum income’ and to address problems of Roma exclusion. ²⁴

Travelling up the compliance ladder, the Semester has its own budgetary tools. As the EU’s 2014-2020 Structural Funding Regulation makes clear, EU social and structural funding can be aligned to Semester priorities providing positive incentives and, potentially, vital state capacity in the form of social spending and infrastructure for national governments to fulfil common EU social goals. ²⁵ New initiatives discussed elsewhere in this issue – such as the ‘Pillar of Social Rights’ – promise to strengthen the ‘compliance carrot’ of EU funding still further. ²⁶

Finally, while non-compliance with social guidelines under the OMC SPSI had few negative effects bar a frosty reception at the Social Protection Committee, consistent failures to comply with Semester recommendations could potentially play a role in informing the ‘corrective’ arm of the Stability and Growth Pact, leading to financial sanction. ²⁷ In this sense, even in circumstances where such penalties are not levied, social policy governance under the Semester could benefit from the ‘shadow of hierarchy’ provided by processes like the Macroeconomic Imbalance Procedure and Excessive Deficit Procedure. ²⁸

The Commission’s 2015 Communication on how it plans to apply the Stability and Growth Pact makes this clearer. ²⁹ The purpose of the Communication is to lay out in greater detail the circumstances under which states will be more closely monitored via the Pact’s preventive and corrective arms, and ultimately what conditions will be taken into account when considering sanctions (a step famously not yet taken). The Communication makes clear that implementation of country specific recommendations will be taken into account in considering whether Excessive Deficit Procedures will be halted or advanced. More specifically, the Communication states that Member States may avoid closer surveillance under the preventive ³⁰ and corrective ³¹ arms where they can show that failure to meet budgetary objectives is owing to ‘major structural reforms’ recommended under the European Semester process that are likely to have positive long-term budgetary effects. While a soft law document, the Communication indicates the ‘hard edges’ the broader framework of economic governance could give the Semester process.

A second reason for optimism regarding the Semester’s ‘displacement’ of social Europe comes from recent academic work. Jonathan Zeitlin and Bart Vanhercke, for example, have argued that – while the Semester emerged from the shadow of financial crisis – it has been significantly ‘socialised’ over time. ³² Zeitlin and Vanhercke’s argument contains two main components. First, they argue for changes in the Semester’s policy orientation. Key framing documents, such as the Annual Growth Survey and country specific recommendations, increasingly include social objectives. In the case of the latter, for example, while the initial Semester ignored social policy, by 2014 12 country specific recommendations related to poverty and social inclusion. ³³

Secondly, they point to governance changes. We may have shifted from a Semester primarily seen by social actors as something to be resisted to one where social actors seek opportunities to influence and shift fiscal policy in a more socially ameliorative direction. To give some examples, from 2013, the Irish Presidency began a practice of consulting the Social Protection Committee and EMCO (the employment committee) on social policy country specific recommendations. This process of inter-committee consultation allowed the amendment of some 10 recommendations between Commission Recommendation and Council adoption in 2013, with a number of these amendments seeking to provide Member States with more leeway in adjusting to fiscal consolidation demands. ³⁴ These efforts may be aided by other developments – the Five Presidents Report tasked the Social Protection Committee and EMCO with establishing five baseline employment and poverty indicators in 2013, providing new possibilities to compare performance between states (or even ‘name and shame’ them) via the Semester on social as well as economic grounds. These were realised in 2017 via the establishment of a ‘social scoreboard’ as part of the Pillar of Social Rights proposal. ³⁵ For Zeitlin and Vanhercke:

We interpret these developments as not only a response by the Commission and other EU institutions to rising social and political discontent with the consequences of post-crisis austerity policies among European citizens but also as a product of reflexive learning and creative adaptation on the part of social and employment actors to the new institutional conditions of the European Semester. We consider such reflexive learning as another form of ‘socialization’. ³⁶

If they are right, perhaps displacement under the Semester is not to be feared. The thesis of socialisation could be seen as displacements’ exact opposite, replacing the coordination of social policy through economic lenses with the spread of social goals and instruments across the entire gamut of EU policy coordination. Perhaps, if these scholars are right, the Semester can be ‘tamed’ or socialised, acting as a new (and more effective) forum through which EU social policy can be developed.

Assessing the ‘Social’ Semester empirically: displaced or ignored?

We should not jump to this conclusion too quickly. Examining in more detail the social dimension of the Semester in the first five years of its life, to what extent do the arguments raised above really convince? Certainly, there is little doubt that the language and terminology of social policy is far more present in the Semester than ever before. This is unsurprising given the Commission’s 2013 efforts to strengthen the social dimension of economic and monetary Union. ³⁷

It is important, however, to distinguish between poverty or other social goals being ‘mentioned’ and being given a useful, autonomous meaning. Often poverty in particular is reduced under the Semester process to other goals and priorities. The Annual Growth Survey – the document that kicks off the policy cycle by analysing and setting out the Semester’s main policy objectives – is an acute example of this. As Daly and Copeland point out, poverty is often discussed in the Annual Growth Survey yet normally through the lens of employment: vulnerable people are to be protected but via ‘the strengthening of the link between social assistance and labour market activation policies’. ³⁸

What, however, if employment does not, of itself, address the risk of poverty. The rate and prevalence of ‘in-work poverty’ i.e. those working poor who are below the poverty rate in spite of employment, reached 9.5% of the working population in 2014. ³⁹ Or what if people are unable to work because of illness or disability, with a significant proportion of this group facing a high poverty risk? There is thus a sense in which EU action in the poverty field is targeted either at only a section of those facing poverty or at what 19^th century romanticists once referred to as the ‘deserving poor’. ⁴⁰ When examining poverty-related country specific recommendations, their focus is often on specific discriminated against or vulnerable groups such as children or the Roma, rather than being more general and cross-cutting in orientation. ⁴¹ General social transfers – despite acting as one of the most effective routes in minimising material poverty ⁴² – are often conceptualised by the Annual Growth Survey as a route to dependency and a possible inhibition to labour market participation. As stated by the 2016 Annual Growth Survey:

More effective social protection systems are needed to confront poverty and social inclusion, while preserving sustainable public finances and incentives to work. Any such development will have to continue to ensure that the design of in-work benefits, unemployment benefit and minimum income schemes constitutes an incentive to enter the job market. ⁴³

This emphasis on a particular ‘growth friendly’ reading of social policy can also be seen in other aspects of the country specific recommendations process. There are numerous examples of country specific recommendations which encourage states to reduce spending to gain competitiveness but also seek to exempt certain public programmes from spending cuts. Those exempted, however, are typically areas of government spending seen as conducive to growth and competitiveness, i.e. education, research and innovation, rather than those related to poverty or deprivation. ⁴⁴ This is also evident in the country specific recommendations from 2016 – while national efforts in social policy are to be directed at ‘modernisation’ or ‘targeted’ efforts, areas like R&D or infrastructure are discussed in terms of spending: ‘where available, for example in Germany, Member States should use their fiscal space to increase public investment in areas conducive to growth, such as infrastructure, education and research’. ⁴⁵

To make this point more concretely, it may be helpful to examine the issue of ‘displacement v socialisation’ by examining specific aspects of social policy in more detail. One example is health – an area historically cordoned-off from EU intervention. ⁴⁶ As argued by Bart Vanhercke and Rita Baeten, the Euro crisis provided a unique window of opportunity for EU health intervention. ⁴⁷ The trigger for this intervention was economic – the increasing realisation that spiralling health costs and their increasing share of national budgets were a threat to fiscal sustainability. This led to an increasing interest of actors like the Directorate-General for Economic and Financial Affairs in health, as well as a gradual increase in the number of health-related country specific recommendations, from 3 in 2011 to 16 in 2014 and 14 in 2015. ⁴⁸

Published Commission papers in the field of social protection have provided greater insights into the process by which country specific recommendations for health are adopted. According to the Commission, the primary indicator for country specific recommendation development in the health field are states where trends in healthcare spending indicate a shift to an above median fiscal sustainability gap. ⁴⁹ The primary indicator driving the development of these recommendations relates to fiscal sustainability. While other Directorates-General may provide input, country specific recommendations are drafted by the Directorate-General for Economic and Financial Affairs when they are based on the Stability and Growth Pact or Macroeconomic Imbalance Procedure. The final country specific recommendations are adopted by the Economic and Financial Affairs Council configuration of the Council, with a qualified majority vote necessary for Member State amendment (via, for example, more socially minded actors in the Social Protection Committee or EMCO committees). In health, as in other areas of the Semester, economic actors thus predominate at the key stages of the process, with social actors seeking to correct their action facing significant decision-making hurdles.

The governance imbalance between ‘socially’ and ‘economically’ minded actors also has a substantive dimension. If country specific recommendations take the Macroeconomic Imbalance Procedure or the Stability and Growth Pact as their legal basis, they are discussed solely in the Economic and Financial Affairs Council. If they are based on Articles 121 and 148 of the Treaty, i.e. the larger Lisbon 2020 integrated guidelines, they are normally additionally discussed and adopted in the Employment, Social Policy, Health and Consumer Affairs Council. This is likely to have a real bearing on the types of recommendations eventually adopted, i.e. social affairs Ministers have a very different – and most likely, more protective and flexible – view of country specific recommendations with a social component than do finance ministers.

The difficulty may be the tendency to issue country specific recommendations addressing social policy topics under the Macroeconomic Imbalance Procedure or Stability and Growth Pact, thus limiting the input of social actors. The Social Protection Committee and EMCO’s analysis of the 2016 round of recommendations demonstrates that 19 of the country specific recommendations from that year on social protection and social inclusion were issued as Macroeconomic Imbalance Procedure recommendations (in spite of the absence of poverty and social exclusion indicators from the relevant legal and policy framework). ⁵⁰ Looking further back, Sonja Bekker, examining the recommendations from 2013, points out that 39 of 78 employment and social policy country specific recommendations were legally linked to either the Stability and Growth Pact or the Macroeconomic Imbalance Procedure. ⁵¹ The political battle over the Semester is therefore also a battle about displacement, i.e. over the lens under which social policy should be examined. Rather unsurprisingly, actors such as the Social Protection Committee argue for a more prominent role for the integrated guidelines (and hence the Employment, Social Policy, Health and Consumer Affairs Council) in the formation and adoption of social country specific recommendations.

The concern that this ‘battle’ is being lost by social voices is reflected in the substantive content of health country specific recommendations. Azzopardi-Muscat has analysed the content of these from 2011 to 2014, according to the Social Protection Committee’s three-part classification of health system indicators. ⁵² Do country specific recommendations relate to the fiscal sustainability of the health system, to access to it, or to quality of care? Of the 65 recommendations issued in this period, approximately two-thirds addressed the fiscal sustainability aspect and only one-third the other two aspects. ⁵³ This finding drives their conclusion that, in spite of Article 168 TFEU’s assurance that ‘a high level of human health protection shall be ensured’ in the Union’s activities:

A hierarchy of health policy objectives is emerging with sustainability becoming supreme to access and quality, departing from the balanced triad of policy objectives promulgated through the Open Method of Coordination or the focus on key values for European health systems. Health systems CSRs are framed as a means to the objective of ensuring sustainability of public finances and not as part of the pillar on combatting poverty and social exclusion. ⁵⁴

As much as non-fiscal goals are ‘mainstreamed’ into the Semester, they face an uphill battle, given the Semester’s procedural and substantive design, to gain policy priority over the Semester’s fiscal objectives.

Some of these concerns also apply when looking more specifically at the area of poverty and social inclusion. The Social Protection Committee and EMCO’s assessment of the country specific recommendations in 2016 notes, for example, the high divergence between a declining national performance in terms of the EU’s poverty targets coupled with neglect of poverty in the recommendations. In particular, they note a rise in the ‘at risk of poverty’ rate in 11 Member States. Yet only two states received country specific recommendations on poverty reduction, the same number as in 2015, with both of them tied to the Macroeconomic Imbalance Procedure. ⁵⁵

In the area of pensions, we also see a repeat of the lopsided focus on sustainability over other non-fiscal criteria observed in health. Nine pension country specific recommendations were issued in 2016, with six of them based on the Macroeconomic Imbalance Procedure. These recommendations emphasised fiscal sustainability and a greater link between retirement age and life expectancy. Only one recommendation addresses the adequacy of pensions. This reinforces other work, commenting on the high degree of uniformity and prescription noticeable in Semester recommendations relating to pensions, which tend to uniformly insist on later pensionable and retirement ages across national systems regardless of their distinct welfare systems and social indicators. ⁵⁶ This is continually repeated year by year in spite of evidence that 26 out of 28 Member States have already adopted provisions since the financial crisis for increasing the statutory retirement age. ⁵⁷ The Social Protection Committee/EMCO review’s answer is to make a plea for greater flexibility in how pension country specific recommendations are formulated and implemented:

The [Social Protection Committee] maintains that, in addition to adjustments to the statutory pension age, and given the competences of the Member States in the area of social protection in general and pensions in particular, other tools (such as restricting access to early retirement pathways, extending contributory periods, including a life expectancy factor in the benefit calculation formula and/or stepping up efforts in workplaces and labour markets to enable women and men to work longer) also represent valid policy options for increasing the effective retirement age and for adapting pension systems to the changing demographic and economic conditions. ⁵⁸

Why is the discussion over country specific recommendations so important, especially given the discussion above on the effectiveness of what are largely soft law measures? Such scepticism might be heightened by other studies – yearly European Parliament studies, for example, put the level of ‘full’ implementation of country specific recommendations at continued lows. ⁵⁹

One should not, however, look at the country specific recommendations process in isolation. This process also affects other parts of the Semester and of EU economic governance in general. Assessments of draft budgets (as enabled by the two-pack legislation) also refer to the conformity of budgets to the Macroeconomic Imbalance Procedure and integrated guidelines. ⁶⁰ Furthermore, conformity with CSRs has been given by the Commission as a reason not to pursue the corrective arm of the Macroeconomic Imbalance Procedure after the issuance of an excessive imbalance notification. ⁶¹ The CSRs (and perceptions regarding conformity with them) feed across economic governance (including to those procedures that carry more coercive elements). In simple terms, the displacement of social Europe via the Semester, or the tendency of the Semester to coordinate social policy through lenses other than social policy itself, is not some neutral rhetorical exercise but one that can still carry real effects.

A final question concerns whether more recent developments can correct the displacement trend. To move to the most recent 2017 round of country specific recommendations, there are some indications that the Juncker Commission’s continued promise to strengthen the social dimension of Economic and Monetary Union may finally be bearing some fruit. Examining the 2017 recommendations, the Social Protection Committee notes that it ‘presents a welcome shift towards a more balanced consideration of economic and social objectives, stronger emphasis on social policy priorities as well as on social impact in the implementation of policy reforms in all areas.’ ⁶² There are some further positive developments in the specific policy fields discussed above. In health, for example, we see a substantial increase in the number of country specific recommendations mentioning accessibility and quality dimensions of care. ⁶³ This may be accompanied by a more ‘hands off’ approach – the Social Protection Committee also notes that ‘the recommendations in relation to pension policy, in contrast to previous years, are less prescriptive and leave the necessary policy space for Member States to use the appropriate policy levers.’ ⁶⁴ Perhaps this can be taken as a sign, à la Vanhercke and Zeitlin, of social actors gaining traction, or at least of a more stable economic environment allowing room for concerns beyond fiscal sustainability to find their voice.

An initial danger concerns the contingency of this development. It is unclear, for example, whether any new-found balance between economic and social objectives would survive a future downturn in the Union’s economic fortunes. The wider danger may be that – even if economic and social conditions improve – a frame for social policy has been established via the Semester that is difficult to budge. It is interesting, for example, that the Social Protection Committee (a body of social actors) begins its assessment of the 2017 round with the observation that ‘social policy should be seen as an investment and a productive factor’. ⁶⁵ What in 2005 may have been a typical statement on social policy from a group of economy ministers has now found its way into how social officials themselves see social policy (and its broader societal function). At the very least, what earlier social policy scholars would have termed ‘de-commodification’, i.e. the attempt to remove certain core aspects of personhood from market conditions, is hard to reconcile with the Semester structure. ⁶⁶ Social policy is a matter of making individuals ‘fit for the market’, with the notion of using it to free them from dependence on economic conditions and the arbitrariness associated with swift changes in supply and demand, receding from view.

This suggests the paradox of socialisation thesis. It hopes to rescue the European Semester by capturing its processes for social voices. What, though, of the danger that social voices are themselves captured, or ‘socialised into’ the Semester’s wider logic of competitiveness and market fitness? A Semester that follows the 2017 trend – a moderately positive economic tide with the seeming potential to lift all boats – is also still vulnerable to falling back into the old debate about new governance: is it a process of socially embedding the market or of moulding social policy for economic ends?

Displacing displacement: back to the future?

Is there an alternative to displacement or to the ancillary role social policy under the European Semester so often plays? The most obvious alternative would be somehow to go ‘back to the future’. This would be the situation of the early 2000s, when social inclusion policy in particular was a much more distinct pillar: one with its own set of institutions, indicators, action plans and guidelines, and one with limited and defined interaction with other policy coordination processes. Such a strategy would seek to rescue social inclusion policy from the cognitive limits set by the European Semester discussed above, in particular its deep association with fiscal sustainability on the one hand, and convergence of EU economies on the other. This autonomy strategy – centred on the renewed social reporting process established in 2012 – would also surely involve focusing on improving the ‘structural entry’ of social actors ⁶⁷ , such as the Social Partners, into national and EU policy discussions (a major participatory hope of the original Social Open Method of Coordination but one that finds little resonance in the founding documents and guidelines of the Semester). ⁶⁸

Unfortunately, the Open Method of Coordination’s past also gives some hints as to the limits of this strategy. For better or worse, the Semester has become the centre of EU policy coordination. While it was once hoped that the Open Method of Coordination would spread into ever greater fields of cooperation, national governments have tended to resist its administrative burdens, preferring to avoid duplicating the reporting demanded by the Semester in other policy fields. Part of the very rationale of abandoning the Open Method of Coordination SPSI in 2010 was that it was in danger of slipping into irrelevance, or becoming a closed conversation of social policy actors increasingly ignored by the EU’s main centre of political gravity, oriented around the Eurozone. ⁶⁹ The dangers of going ‘back to the future’ are thus clear: such a strategy could yield a more autonomous form of social policy coordination but also one whose political purchase is reduced.

This is the dilemma of displacement: the search for new forums in which to articulate and coordinate social policy in the EU was also in part driven by the weaknesses of EU social policy in its more autonomous forms. Returning to a more autonomous social policy sphere – without an extensive political re-orientation of the Union as a whole – risks increasing the distinctiveness of EU social policy only to see that gain coupled with increasing marginalisation.

A further alternative may rest in the ‘Pillar of Social Rights’. ⁷⁰ At its most optimistic, such a Pillar would precisely seek the larger political re-orientation of the Union necessary to filter down and alter the substantive orientation of EU policy coordination itself. One could imagine two paths to such a Pillar: either one that expands the Union’s competences in the social field, addressing ths current imbalance between market-making and market-correcting competences, or else one that simply seeks to shield national social policy from the negative effects of EU economic integration (a ‘constitutionally conditioned’ internal market). ⁷¹ This initiative is discussed in better and greater detail elsewhere in this special issue. ⁷²

It is sufficient here to state that it seems unlikely to form the basis of a broader re-balancing of the European Semester’s political context. The very notion of re-stating social rights (some of which have well-established juridical form) as principles seems an indication of how far the logic of policy coordination has reached into the very fabric of EU social policy. If part of the problem of Social Europe is its reliance on soft law instruments, which are too often outweighed by harder economic procedures, the Pillar of Social Rights replicates this phenomenon, with much of its content deliverable only via soft law, including the European Semester itself. ⁷³ In this sense, the Pillar does not escape the constraints of governance under the European Semester: rather, it generalises the mechanisms and instruments of the Semester – i.e. setting out broad principles, reporting, benchmarking, comparison and recommendations – towards EU social policy as a whole.

In this sense, EU social policy coordination has achieved a kind of Pyrrhic victory. On the one hand, policy coordination, as a mode of governance, has transformed itself from an innocent and inferior ancillary to ‘hard’ social law into one of the central instruments through which EU social policy is now being delivered. ⁷⁴ On the other, soft policy coordination has proven an even more effective vehicle for displacing social policy goals than its hard law cousin, reaching into policy fields (see the examples of health and pensions discussed above) that other forms of EU law never could. If ‘displacement’ has been the key story of social Europe’s last decade, open coordination has been one of its leading characters.