Balancing EU social and economic governance through performance management

ABSTRACT This article studies how the Macroeconomic Imbalance Procedure (MIP) is reconfiguring social and economic policy-making in the European Union (EU). The MIP was meant to tackle competitiveness-based imbalances but today addresses a broader range of social concerns. Scholars agree this fits in a trend of depoliticizing governance through tools of performance management. But they disagree whether this marks the ‘socialisation’ or ‘neoliberalization’ of governance. We contribute to this debate by drawing attention to the politics of performance management in the context of neoliberal European integration. We argue performance management has been driven by two tensions: i) macroeconomic divergence and intergovernmental dissensus and ii) EU institutions’ contradictory technical responsibilities and political ambitions. We conclude that the expansion of social indicators in the MIP has undermined the coherence of ‘neoliberal’ regulation. However, its connection with a project of the European Commission’s empowerment means it is unlikely to improve prospects for greater social protection.


Introduction
In 2011, as part of the 'Six-Pack' of measures that made up the so-called post-crisis New Economic Governance (NEG), the European Union (EU) introduced the Macroeconomic Imbalance Procedure (MIP). 1 Many European officials hoped the MIP could achieve the density of governance in macroeconomic affairs that the Stability and Growth Pact (SGP) had done for fiscal policy (European Commission, 2008). In a 'revisionist consensus' (Collignon, 2013), EU policy-makers had come to see macroeconomic imbalances as a major factor in the unfolding Eurozone crisis, reversing the 'benign neglect' they had received in the years previously (e.g. Blanchard & Giavazzi, 2002). 2 The MIP is a performance management tool that uses a scoreboard of indicators to monitor imbalances, leading to progressively tougher levels of European Commission surveillance, and even sanctions for non-compliance with recommendations.
Puzzlingly for a procedure pitched as an early warning system of current account and competitiveness imbalances, and in the face of stiff opposition from national governments, scholars have tracked how the European Commission has promoted a number of 'social and employment indicators' in the MIP (Zeitlin & Vanhercke, 2014, 2018. Observers criticised the early MIP for its asymmetrical approach to macroeconomic surpluses and deficits, leaving deficit countries vulnerable to post-crisis neoliberalization through competitiveness-focused structural adjustment programmes (Scharpf, 2021). But the addition of social indicators suggests it has been repurposed. Today it is a broader instrument seeking to make interventions through a range of economic and social indicators.
The use of performance management as a tool of social policy has long been recognised by scholars of neoliberalization (van Apeldoorn & Hager, 2010;Nunn, 2020). Performance management has been seen as a depoliticising instrument that ingrains a competitive logic into the delivery of public welfare that is recast as 'workfare'. In the EU, this has been observed to fit in a trend whereby performance management techniques have driven a growing fusion of economic and social policy, such as seen in the MIP (Puetter, 2012;. Scholars have argued that this fusion reflects the thorough neoliberalization of EU social policy, as questions of employment or healthcare are recast in terms of their competitiveness contributions or fiscal burdens (Wigger, 2019).
For many of these scholars, the possibility of contesting the neoliberalization of social policy through performance management has been through opportunities of re-politicising the implementation of workfare by on-the-ground policy officials (Nunn, 2020). What is curious about the case of the MIP, however, is that the addition of social indicators was widely seen by policy actors as a politicising move, but from the top rather than the bottom. To the frustration of national policy-makers, the addition of social and employment indicators was viewed as a project of the European Commission that was widely seen as politicising the MIP and diluting its focus on competitiveness. As such, whereas the literature currently understands the use of performance management in social policy as a depoliticising tool to instil neoliberal competitiveness, the MIP presents a case where social policy performance management has been used to disrupt a thoroughly neoliberal policy instrument.
To make sense of this, we argue that critical literature should look more precisely at the political institutional dynamics behind the turn to performance management in the neoliberal era. We argue that two tensions at the heart of socio-economic governance shape performance management in the EU: i) intergovernmental dissensus between the EU's divergent national economies; ii) and EU institutions' overlapping and contradictory technical responsibilities and political ambitions.
It is a well-established observation that intergovernmental dissensus in the EU set the historical context for the turn to performance management (Schäfer, 2004). Less recognised, however, are its consequences for the European Commission's role in the EU. With the rise of performance management, the European Commission has found its institutional position shifting from a 'policy entrepreneur' driving integration forward through grand plans to a 'policy manager' responsible for detailed policy surveillance of national economies (See Chang & Monar, 2013). But while the Commission's responsibilities have become more technical, scholars have noted the institution's political ambitions and legitimacy pressures have intensified (Majone, 2014;Mérand, 2021).
It is within this context, we argue in relation to the MIP, that the Commission has been politically instrumentalising its technical surveillance responsibilities as a response to the legitimacy crisis it has faced after the euro crisis. As we conclude, this has undermined the coherence of 'neoliberal' economic governance that many criticise the EU for. But its significance as a project for the Commission's own institutional empowerment means it is unlikely to improve prospects for a 'European social model' as some scholars may hope.
Our analysis proceeds in three steps. First, we elaborate and position our argument in relation to the literature on performance management in a neoliberalizing Europe. Second, we develop our theoretical perspective on the political significance of performance management by placing it in the historical context of European capitalism and EU governance since the 1990s. Third, a final section will zoom in on the contemporary case study of the MIP. This section will process trace the 'socialisation' of the MIP across the negotiation and implementation of the procedure. Data for this analysis was gathered from EU policy documents, including national position papers exchanged within the 2010-2011 Van Rompuy Task Force on Economic Governance, Commission Communications and Staff Working Papers, and reports from the EU legislative process that passed the two regulations that underpin the MIP. In addition, we report from seven interviews conducted as part of this research. These seven interviews are listed at the end of the paper. The interviews were conducted in Brussels, Frankfurt, and London in 2017 and 2018 with EU institution officials (European Commission, European Parliament, European Council) and member state officials in permanent representations to the EU. Data from interviews was used to add greater analytical depth on the informal political context and dynamics of the MIP's creation and implementation.
2. State of the art: is performance management market-making or policy learning?
European integration is de jure asymmetric in favour of market integration. As Fritz Scharpf has noted, market integration is treaty-led and supported by law and courts, while market correction is largely supported by only political will requiring unanimity or qualified majorities among member states (Scharpf, 2010). Despite this asymmetry, social policy has nonetheless developed into a major site of governance driving political and academic debate.
Giving shape to social policy governance, EU policymakers have historically turned to a variety of instruments to coordinate policies that have been seemingly impossible to collectively agree on. These instruments have included practices like benchmarking and ranking, peer-review, and other indicatorand number-based governance techniques which were epitomised by the so-called Open Method of Coordination (OMC) of the 2000s. Collectively these instruments work according to a logic of performance management: EU institutions collect, organise, and evaluate member states' policy performance data in order to facilitate intergovernmental policy coordination. Rather than adopting binding rules, EU governance is run as a managerial machine of policy analysis and recommendation, issued with varying degrees of compulsion. The euro crisis brought significant changes to the role of these practices. In particular, the 2010 creation of a single policy coordination architecture for the EU known as the 'European Semester' not only fused economic and social policy into the same governance process, but also deepened the level of EU intrusion into member state policymaking (de la Porte & Heins, 2016). Amandine Crespy and Georg Menz point to a stark divergence in opinion on the significance of performance management (Crespy & Menz, 2015, p. 183). First, policy-and sociologically-oriented political economists like Jonathan Zeitlin, Philippe Pochet, or Caroline de la Porte have seen performance management as a potentially innovative way to develop EU social policy within the constraints of intergovernmental dissensuseven if concrete manifestations like the OMC have not realised their democratic, deliberative, or pragmatic potential (de la Porte & Pochet, 2002;de la Porte & Nanz, 2004;Zeitlin, 2005). Second, a 'Cologne School' of Political Economy from scholars like Fritz Scharpf, Martin Höpner or Armin Schäfer have perceived performance management as a reflection of EU social policy's subservience to law-led market integration (Scharpf, 2010;Höpner & Schäfer, 2012).
To this debate we add a third Critical Political Economy (CPE) approach of scholars who link performance management techniques to the structural interests of transnational capitalinfusing policy-making with a market ethos that only symbolically gestures to social concerns ( van Apeldoorn & Hager, 2010;Bruff, 2017;Wigger, 2019). This section will review these three perspectives on performance management to situate our argument.
Sociologically-oriented political economists have expressed hope that non-binding modes of performance management can help progress social policy and address the asymmetry of EU governance. Through concepts of learning and experimentation, Zeitlin (2005) argues that the exposure of states to comparative data on issues such as social exclusion or youth unemployment can trigger policy improvements. The postcrisis fusion of social and economic governance has consequently been observed by Zeitlin and Vanhercke to have led to 'a partial but progressive 'socialization' of the [European] Semester' since its inception in 2010 (Zeitlin & Vanhercke, 2018, p. 152). In particular, the two scholars have observed a 'growing emphasis on social objectives' and increased inclusion of socially oriented actors in policymaking (p. 167). We agree that social issues have been deeply integrated into EU policymaking through the layering of coordination mechanisms from the Lisbon Agenda to the European Semester. But at a time of growing social unrest, stagnant living standards, and ingrained inequality, we are still left with the puzzle of why this growing 'socialisation' through performance management has not translated into improved social protection in Europe. We agree with Zeitlin et al. on the importance of inter-institutional struggles over the direction of EU policymaking, but argue that it is vital to link these struggles to the underlying tensions and contradictions in European capitalism to draw out the significance of performance management.
For Cologne School political economists, performance management through indicators, peer-review, and learning, hits up against the political divergences in the heterogenous EU polity that constrain the possibility of a redistributive social policy (Scharpf, 2010). For all the observations over social policy governance's institutional density, these scholars argue that the EU is structurally unable to pursue a redistributive social policy (Höpner & Schäfer, 2012). This is because, even at times of social democratic majorities in the main EU institutions, there is a consistent veto against EU-level redistribution owed to the heterogeneity of welfare models (Scharpf, 2010). For these scholars, the adoption of performance management by the social democratic governments of the late-1990s was a testament to the weakness of social policy vis-à-vis market integration rather than anything that can potentially act as a force of market correction (Schäfer, 2004).
In short, the policy-oriented literature of Zeitlin and others see performance management as a necessary depoliticization of socio-economic governance allowing EU institutions and domestic stakeholders to creatively collaborate on pragmatic solutions to social and economic ills. The Cologne School, as well as Crespy and Menz, see this depoliticization as a dangerously constrained and insufficient set of tools to overcome marketled European integration and the subsequent liberalization of domestic political economies.
We agree with the Cologne School's emphasis of the intergovernmental constraints on the development of redistributive social policy. Nevertheless, we recognise that performance management has had a profound effect reorienting the very nature of EU governance. Indeed, given the unlikely conditions, it is surprising the extent to which social policy governance has extendedas revealed by the more sociological approaches discussed above. While the Cologne School rightfully corrects overly optimistic assessments of the performance management's progressive potential, it is missing a deeper investigation of the political dynamics of EU policy-making that has propelled economic and social policy through techniques of performance management.
A third CPE school directly addresses these political dynamics of performance management in the EU most powerfully. We see our argument as contributing to and building on these approaches. For a number of years, CPE scholars have observed how performance management has bolstered neoliberal integration through techniques of depoliticisation (van Apeldoorn et al., 2009;Overbeek & van Apeldoorn, 2012). Along these lines, van Apeldoorn and Hager (2010) argued that techniques like social policy benchmarking were 'market-making' rather than 'market-correcting' through their transformation of policymaking from a collective democratic exercise into a market-like race to climb competitiveness rankings. More recently, scholars have begun to take a more careful look at the role of management practices to processes of neoliberalization (Knafo et al., 2018;Eagleton-Pierce & Knafo, 2020). Such perspectives build on a deeper analysis of neoliberalism as not a project of state withdrawal, like its ideologues argued, but a state-led project remaking welfare institutions in market-conforming ways (Cahill & Konings, 2017).
Alex Nunn (2020) has added a novel perspective to these CPE arguments by highlighting the possibilities of contesting the disciplinary effects of neoliberal performance management. For Nunn, performance management is a 'multi-scalar' technique of neoliberalism that depoliticises governance to 'insulate central decision makers from critique about the experience of policy delivery' (Nunn, 2020, p. 9). Nunn argues that performance management works to fix lower levels of the policy-making machinery into regimented performance obligations to 'insulate these scalar relations from everyday political questioning' (Nunn, 2020, p. 4). This is never, though, a closed effort for Nunn. Rather, the complexities and contradictions of policy delivery mean that 'the work that lower scale managers do in implementation becomes highly politicized' as the latter navigate the failures, contradictions, and trade-offs of performance management (Nunn, 2020, p. 15). As such, while performance management is seen as a key tool of neoliberal depoliticization, it is for Nunn seen as one latent for re-politicisation through the work of onthe-ground policy officials on the receiving end of performance measures.
The MIP, though, poses a puzzle to this reading of depoliticization-at-the-top, repoliticisation-on-the-ground. As we show below, the 'socialisation' of the MIP was not a project of the national governments on the receiving end of performance measures -who were firmly against it. But rather it was through modifications made by the European Commission who was technically responsible for the MIP's indicators at the top. In contrast to Nunn's depoliticization-at-the-top narrative, we therefore explore the possibility of the re-politicization of performance management by political-administrative institutions at the top of multi-level governance frameworks. The novelty of this is to show what a more detailed and political reading of governing institutions can reveal compared to the historical materialist understandings that reduce performance management to broader capitalist developments.
The next section will build such an analysis, offering our historically grounded theoretical perspective on the rise of performance management in the EU. This allows us to highlight the closely entwined processes of depoliticization and re-politicization in the specific circumstances of performance management's rise and implementation in the EU.

Theoretical perspective: making sense of EU performance management
In this section, we bring together the insights from the broad literature on EU performance management to build a novel interpretation of why it arose and what it has politically signified. To do this, we argue that the politicisation/depoliticization dynamics of performance management must be placed within a historical context of intergovernmental dissensus and the changing institutional legitimacy pressures that have shaped integration since the 1990s. We argue that the turn to performance management in the EU was never a straightforward process of depoliticization. Instead, managerial techniques were used to reframe political conflict in the heterogenous polity of the EU. Our theoretical perspective elaborated below follows a well-established critical tradition of historically-driven theorising that emphasises placing developments in their historical socioeconomic context over building abstracted theoretical models (Cox, 1981).

Intergovernmental dissensus and macroeconomic divergence
The first step of our theoretical perspective is that performance management was an historically specific innovation that arose from the political economic context of European integration in the late 1990s where three trends came together: macroeconomic divergence in the new Economic and Monetary Union (EMU); intergovernmental dissensus between new social democratic governments of Europe; and the growing movement of New Public Management.
As the EU's membership has grown, increasingly different models of capitalism were placed into a single polity. Whereas many had presumed that the EMU would lead to a process of convergence between these different models of capitalism, the 'one-sizefits-all' structure instead entrenched a process of divergence leading up to the euro crisis (Becker & Jäger, 2012;Höpner & Lutter, 2018). In the context of these heterogeneous models of capitalism, EU governance has been constrained by the divergence of interests it has bred. Germany, and other countries of Northern, Central and Eastern Europe, with a credence to ordoliberalism, an objective of macroeconomic stability, and an export-oriented economic model have rejected transnational fiscal transfers out of moral hazard fears (Baccaro & Benassi, 2017). In the United Kingdom, hostility to deeper European integration and preference for a finance-led trajectory ultimately resulted in Brexit (Lavery, 2017). In the Nordic countries, a tradition for minimalist employment laws has quashed support of extensive EU social policy (Mailand & Arnholtz, 2015). The result has been a gradual shift of EU governance from a process of negotiating convergence to a supranational state, to coordinating disagreement between variegated forms of capitalism.
It was in this context that techniques of performance management first found favour in the mid-1990s. After the adoption of the Maastricht Treaty in 1992, accelerating market integration, the EU was hit by growing unemployment, fiscal austerity, and monetary strife. Whereas the push to the EMU was dominated by largely conservative and Christian Democrat governments, the political climate changed in the mid-1990s as the negative fallout of market integration swept social democrats into office across the EU. The EU's social democrats in the 1990s were largely united on the need to invigorate a European social policy. But, as a consequence of macroeconomic divergence, they were politically divided on what to do about it.
Leading 'left' social democrats such as French Prime Minister Lionel Jospin and German Minister of Finance Oskar Lafontaine advocated a harmonisation of labour and welfare standards across the union. Contrastingly, representing an insurgent Third Way/Die Neue Mitte blend of social democracy, UK Prime Minister Tony Blair and German Chancellor Gerhard Schröder offered more of a compromise between social and liberal varieties of capitalism. Together, Blair and Schröder dismissed the harmonisation of labour market and welfare regulations, and instead called for a focus on employability and labour market activation (Pollack, 2000). In a joint platform ahead of the 1999 European elections, the two leaders outlined a vision for EU social policy thateffectivelycalled for the mobilisation of performance management. Drawing on New Public Management arguments widely circulating at the time that the 'state should not row, but steer', the joint platform eschewed supranational convergence in favour of inter-governmental coordination focused on a discourse of flexibility, performance, and effectiveness (Blair & Schröder, 1998).
It was out of this specific context of the mid-1990s clash between social democrat governments representing very different models of capitalism that the first significant steps to performance management were taken. The major turning point in this, from which EU social policy has since developed, was the Lisbon European Council (Daly, 2012). The adoption of a new 'Open Method of Coordination' (OMC) at Lisbon marked a qualitative shift in EU social policy from convergence to coordination. This shift was, however, more than simply the accommodation of reluctant British and German governments to further EU regulation through voluntarist policy coordination (cf. Schäfer, 2004). The key innovation of the OMC was to add a managerial twist to EU social policy governance by shifting the terms of coordination firmly onto a logic of performance management through the key practice of benchmarking.
Benchmarking was not new to the EU. It was previously proposed by the European Round Table of Industrialists (ERT) in the mid-1990s; a development CPE scholars identified as exemplifying the neoliberalization of EU governance ( van Apeldoorn, 2000;Holman, 2004). For us, going beyond these CPE scholars, the significance of the OMC was to transform performance management from the peripheral policy tool of benchmarking as proposed by the ERT into a core regulatory principle of social policy (Arrowsmith et al., 2004). Benchmarking allowed governments to agree on big targets for employment or social inclusion, but left the means to reach them indeterminate. The basis of this indeterminacy was the pivotal shift of the OMC to focus on policy performance. The OMC sat alongside existing Treaty-based processes by developing timetables, collecting statistical data, compiling indicators, and analysing member states' social and employment systems. By focusing policy activity on the performance of social indicators, rather than the negotiation of social regulation, the OMC provided a means to address policy issues at the highest level of the EU, while leaving questions of policyaction indeterminate. In doing so, the turn to performance management provided a fix that dealt with the political pressure to adopt an EU social policy, to compensate for the backlash to market integration post-Maastricht, while navigating the sensitivities of clashing intergovernmental perspectives on what social policy should look like.

Balancing technical responsibilities and political ambitions
The coming together in the late 1990s of capitalist divergence, intergovernmental dissensus, and NPM zeitgeist provides the driving explanation for the emergence of performance management. The specific way it has developed, however, requires us to focus on the consequences performance management had for the EU's key supranational institutionthe European Commission. Key to this is our argument that performance management did not simply depoliticise governance, but instead reframed political conflict along managerial lines.
The introduction of performance management through the OMC at Lisbon was firmly an intergovernmental process. Resistance from Third Way social democrats to policy convergence was, in part, a resistance to the supranationalisation of welfare. The turn to coordination rather than collectivism transformed the Commission's role from a prospective transnational government administering a European welfare state, to a chief information processor reliant on its technical services to run a performance management machine. This is an observation well recognised in the specialist public administration literature in terms of a shift from the Commission as a 'policy entrepreneur' to a 'policy manager' (Laffan, 1997;Chang & Monar, 2013). Puzzlingly, though, as the Commission has become increasingly reduced to a technical role, scholars have also widely observed an increasingly assertive and political Commission Presidency in recent years (Becker et al., 2016;Dinan, 2016;Kassim et al., 2017;Mérand, 2021). The Juncker Commission, in particular, was described as 'the most political Commission ever' through a combination of a Vice-President system, a claim to a political mandate drawn from the informal Spitzenkandidaten process of 2014, and a willingness to flex its administrative capacities (Peterson, 2017).
Driving the Commission's increasing political ambitions has been a seismic shift since at least the euro crisis in the institution's legitimacy pressures. EU legitimacy is traditionally judged along Fritz Scharpf's terms of input/output (Scharpf, 1999). Problematically for the Commission, as an unelected and largely technocratic institution it has few sources of democratic inputs to base its legitimacy on. Moreover, as its role has shifted to that of a policy manager, it has few visible and exclusive competences upon which its policy output can be judged. Instead, Vivien Schmidt (2013) has argued that the European Commission's largely technical responsibilities has meant its legitimacy has been based on assessments of its 'throughput'. This includes issues like the quality, transparency, and openness of its governance procedures. In the multilevel EU, the Commission has thus relied on a legitimacy 'intermediation' that outsources input legitimacy to national actors and stakes its credentials on principle-agent judgements of its 'throughput' (Scharpf, 2013).
However, as EU policy interventions have become increasingly felt at the level of the everyday (e.g. through sovereign debt bailouts), the EU's existing legitimation mechanisms have begun to break down (Majone, 2014). Increasingly, the Commission has been exposed to the democratic pressures of mass politics, as national and transnational media questions their actions and calls key figures to account. Despite the growing exposure of the Commission to mass politics, there has been continued resistance from national governments to either enhance its sources of democratic input, 3 or provide it with sufficient governing capacities such that its output effectiveness can be judged.
Instead, we argue that as the Commission has had to handle the legitimacy fallout of the euro crisis, it has focused on politicising its technical tasks under EU performance management. Specifically, this politicisation has taken the form of the Commission attempting to expand the remit of social policy in its policy surveillance work. The Commission has a long history of mobilising the idea of the European Social Model (ESM) in pursuing a political agenda. For Jepsen and Pascual (2005), the ESM is an ideational vehicle for EU institutional legitimation and the construction of supranational regulation. The ESM brings together the normative assertion of Europe's unique socio-economic qualities (compared to the USA) with 'scientific conclusions' on globalization-induced shifts in employment and industrial relations towards a 'knowledge-based' economy (Jepsen & Pascual, 2005, p. 233). Building on this tradition, the Commission has attempted to deepen its institutional status and mass political appeal by politicising its technical tasks within EU governance through normative appeals and procedural shifts centred on expanding the remit of social policy. Whereas performance management was originally a vehicle to mediate intergovernmental dissensus, it has become a means for the Commission to manage its legitimacy pressures.
In 2010, for example, the Barroso Commission rushed out the new social strategy of Europe 2020 as its first term was drawing to an end. Facing a backlash at a perception that the Commission was doing nothing besides supporting austerity in response to the emerging financial crisis in late 2009, Barroso sought to make the quantitative targets of Europe 2020 personally attached to his second presidency (Armstrong, 2012, p. 214;Copeland & Daly, 2015, p. 147). Similarly, a new 'European Pillar on Social Rights' (EPSR) was a project devised substantially by the Juncker Commission, 'without involving the Member States through the SPC [Social Protection Committee] and the Employment Committee (EMCO)' (Sabato & Corti, 2018, p. 57). For the Commission, the Social Pillar and its 'social scoreboard' was meant to expand the prominence of EU social policy governance (European Commission, 2017). Sabato and Corti note, however, that as a tool developed solely within the offices of the Commission Presidency, the scoreboard's unclear role and methodology is 'likely to weaken, not strengthen, the monitoring of Member State employment and social protection performance' (Sabato & Corti, 2018, p. 58).
The takeaway from this, we argue, is that while the Commission has increasing political ambitions and legitimacy pressures, since at least the 2000s its tasks have become more technical. The result has been that the Commission has attempted to politicise its technical tasks through the discourse of the European Social Model as a way to demonstrate its policy activism and relevance. As such, Schmidt's observation of the Commission's throughput sources of legitimacy takes on a strange twist. The Commission increasingly does not seek to be judged in terms of the quality of policy surveillance (as the throughput concept implies). Instead, the Commission's legitimacy claims are based on the extent to which it can integrate its own political objectives into EU performance management. In the post-crisis context, this has been through efforts to shift the focus of performance management onto social issues, organised around the ESM discourse.
This complicates the CPE reading of performance management as depoliticization-atthe-top, repoliticisation-on-the-ground. Taking a closer account of the political institutional dynamics of the EU in the context of neoliberal integration, we argue the politicisation of performance management at the top has been a crucial but overlooked dynamic. In the next section we provide a closer analysis of these dynamics as they have unfolded in the case of the MIP.

'Socialising' the MIP in the New Economic governance?
The MIP was introduced in 2011 to target Eurozone member state imbalances in areas like current accounts, private debt, and unit labour costs. In our analysis, we provide particular attention to the MIP's scoreboard of indicators. David Bokhorst has recently argued that the scoreboard has played a marginal role in the MIP compared to the bilateral and vertical engagement between EU officials and domestic actors as part of the procedure more generally (Bokhorst, 2019). Several of our interviewees supported this interpretation, playing down the significance of the MIP's indicators.
Unlike Bokhorst, it is not the purpose of our analysis to gauge the importance of the scoreboard for pragmatic domestic reform. Instead, we are interested in the political significance of performance management as a mode of governance. As such, we argue that the political conflicts that have made the application of the scoreboard difficult offer interesting clues into the (de)politicisation dynamics of performance management. In fact, the non-applicability of the scoreboard in the enforcement of domestic reforms follows perfectly in the tradition of EU performance management. As our previous sections have shown, performance management grew out of the inability among European leaders to agree on collective policies. As such, we argue, the purpose of the MIP scoreboard as a performance management technique was to reframe the question of imbalances along managerial lines in order to signal the EU's political orientation on the socio-economic policy balance rather than develop purposeful instruments. The purpose and significance of the scoreboard, therefore, is political rather than legal or policy-pragmatic.
Our analysis reviews the inter-institutional negotiation of the MIP as part of the 'Six-Pack', and the early implementation of the instrument, in terms of the two tensions outlined above. In order to make explicit the impact of these tensions on concrete political struggles, we also outline the key institutional actors in the story and analyse their role in promoting and/or criticising the addition of social indicators in the MIP's scoreboard.

Unemployment indicator and intergovernmental dissensus
The EU's identification of macroeconomic imbalances as a potentially existential problem for the Euro narrowly preceded the financial crisis (European Commission, 2006, 2008. In the wake of the financial crisis, as Eurozone turmoil escalated and deepened, EU leaders acted on Commission proposals from 2008 for an instrument governing imbalances (European Commission, 2008). From the very outset, there was a shared understanding that imbalances should be targeted as a problem resolved through national structural reforms and not international transfers. Beginning from this shared understanding, intergovernmental and inter-institutional debates focused on the nature of macroeconomic imbalances to judge national economies against. While a consensus was emerging on the prioritisation of domestic reforms and national responsibility, there was disagreement over what constituted an imbalance. These debates reflected the existence of divergent capitalist economies as well as divergent ideological outlooks.
A prominent feature of this clash was the question of the 'social' sources of imbalances. While national governments and the Commission's technical services were quibbling over competitiveness, the European Parliament was pushing for an expanded understanding of imbalances that addressed the social components of economic governance. The Parliament's employment committee argued the procedure 'should address the prevention and correction of macroeconomic as well as of social imbalances on an equal basis' (European Parliament, 2011, p. 3). A raft of proposed indicators of social imbalances came from the Parliament during the negotiations on things like unemployment, income inequality, environmental externalities, poverty, and social exclusion (Interview 7).
For the Commission's Directorate General for Economic and Financial Affairs (DG ECFIN), the Parliament's social proposals were a non-starter. In the Commission's first proposed scoreboard in November 2010 there was no mention of social indicators at all, instead focusing on traditional macroeconomic issues of current account balances, labour costs, and debt (European Commission, 2010). The Commission's initial rejection of social issues reflected a number of concerns. For DG ECFIN, the purpose of managing imbalances built on their original study of pre-crisis imbalances (European Commission, 2008). This analysis took a largely technical perspective on imbalances, arguing a future policy instrument should mirror the SGP in being rules-based and focused on policy coordination in the pursuit of stability rather than any vehicle for political objectives. As DG ECFIN stated, in merely a footnote in an explanatory report introducing the MIP, social indicators were too broad and were better addressed through the Europe 2020 strategy 'where they are indeed taken into account' (European Commission, 2010, p. 6). Insisting on the omission of social and employment indicators from the MIP in the early negotiations, DG ECFIN viewed them as a political proposal rather than anything that could usefully shine a light on the emerging problem of macroeconomic imbalances.
By the end of 2011, however, after the MIP had passed the co-legislative process between the Council and Parliament, the Commission partially reversed its position to include some consideration of social issues. Specifically, a Commission working paper providing an updated design of the scoreboard included unemployment as an indicator for the MIP (European Commission, 2011). While not including the broad range of indicators that had circulated in the European Parliament's discussions, the proposed unemployment indicator nevertheless broadened the MIP's treatment of labour issues from squarely a concern of cost competitiveness to a thin work-focused consideration of social welfare. For their part, the Economic and Financial Affairs Council (ECOFIN) of national ministers welcomed the inclusion of an employment indicator in their conclusions on the Commission's proposed scoreboard design ( ECOFIN Council, 2011). DG ECFIN's and the European Council's sudden acceptance of the MIP's unemployment indicator was arguably because of the asymmetric way in which it was added to the procedure. While employment issues were considered in the most visible part of the MIP in the scoreboard, it was made clear by the Commission and Council from the beginning that employment issues did not have the same procedural weight in triggering the disciplinary steps of the MIP. Rather than an imbalance in its own right, the unemployment indicator was used, in the words of the Commission, to 'understand the potential severity of macroeconomic imbalances' (European Commission, 2011, p. 4). Crucially, unemployment was introduced merely to contextualise 'real' macroeconomic imbalances. As a result, breaching the early warning threshold on the unemployment indicator did not trigger further Commission analysis in the form of an In-Depth Review (IDR), unlike other indicators like the current account, private debt, or unit labour costs. 4 This was despite the poor performance of a number of countries on the unemployment indicator at the time of the MIP's adoption.
The unemployment indicator's presence in the MIP as a visible, but subordinate, indicator speaks to how social policy has been melded with economic policy through performance management in the EU in the navigation of intergovernmental dissensus. As with the turn to the OMC in the 1990s, social issues were an area felt pressing to coordinate (in this case through including them in the MIP's scoreboard). But they were deemed unacceptable to be subject to supranational legalized or administrative pressure (by ensuring they couldn't trigger the MIP's procedural steps). Reflecting this, national government interviewees reported that while the inclusion of unemployment in the scoreboard of indicators was misguided, ultimately member states found such indicators 'harmless to include' since it was made clear to not have procedure consequences (Interview 2). Making unemployment visible as one indicator among many satisfied political demands to consider social issues. Yet, by ensuring that social indicators could not trigger further steps within the MIP, it was far from certain how such visibility could lead to tangible political action.

The European Commission's politicized championing of social indicators
As the MIP was implemented from 2012, the wrangling over the addition of an unemployment indicator quickly looked trivial. From 2013, the Commission (led by the institution's political actors in the College of Commissioners and the Presidency) pushed for the expansion of social and employment indicators in the scoreboard and country analyses. First, in October 2013, the Commission Presidency directed the introduction of 'auxiliary indicators' in the MIP focused on poverty and social exclusion, youth unemployment, long-term unemployment, and labour market participation (European Commission, 2013). This was then taken further in 2015 when three employment indicators (on the activity rate, long-term unemployment, and youth unemployment) were added to the MIP's headline scoreboard (European Commission, 2015a).
Drawing on our theoretical perspective on performance management elaborated above, the deepening of a social and employment component to the MIP reflected the growing political legitimacy crisis facing European institutions as the euro crisis escalated. The MIP had been negotiated in 2010 to remedy the ballooning imbalances that had arisen following the adoption of the EMU. It came too late, however, to usefully respond to the sovereign debt crisis unravelling across Europe following the shock to transatlantic finance. As the enormity of the euro crisis grew, political responses took shape largely outside of the EU's normalised governance instruments through hastily negotiated bailouts and unprecedented monetary interventions (Tooze, 2018, pp. 422-446). The involvement of EU institutions in coercive bailout negotiations brought with it an unparalleled level of political exposure, and growing criticism of the unelected officials of the Commission and dictatorial finance ministers of the Eurogroup in determining the sovereign fate of fiscally stricken countries like Greece and Ireland (Varoufakis, 2017). As a result, commentators at the time widely spoke of a growing legitimacy crisis knocking at the door of the EU's supranational institutions (Meadway, 2012). The Commission itself commissioned scholarly perspectives on remedying this legitimacy crisis (Schmidt, 2015).
To manage this political exposure and legitimacy crisis, various institutions of the EU fell back on traditional appeals to the ESMresponding to depictions of them as austerity-obsessed by expanding a social agenda that could deal with rampant youth unemployment and growing poverty across the EU. In June 2013, European Council Conclusions declared that 'the social dimension of the EMU should be strengthened' (European Council, 2013). Two years later, the 2015 Five President's Report reiterated the point, expressing a cross-institutional consensus that 'employment and social concerns must feature highly in the European Semester', including the MIP (European Commission, 2015b). Former Employment Commissioner László Andor provided external validation for the move, picking up a narrative first deployed in the European Parliament during the MIP's legislative negotiations that 'a social dimension cannot be achieved without a mechanism for collectively monitoring and preventing employment and social imbalances' (Andor, 2013).
The pursuit of this social agenda through the Commission's technical instruments of policy surveillance such as the MIP scoreboard followed earlier abortive Commission efforts to spearhead greater collective EU policy responses to the euro crisis. In particular, the Barroso Commission's 2011 proposal for jointly issued Eurobonds fell on deaf ears with national governments, largely ending the Commission's role as a policy entrepreneur in directing Eurozone crisis responses (Matthijs & McNamara, 2015). Instead, in a pattern discussed above in relation to Lisbon, the Commission's efforts at flexing its political muscles were directed through an instrumentalization of its technical responsibilitiesin this case its capacity to influence the MIP's scoreboard and country analysis reports.
The Commission's political leadership in the Presidency and the College have been at the forefront of the project to expand the use of social and employment indicators in the MIP (Interviews 3 and 4). One interviewee reported that the expansion of social indicators was 'something coming top down from the very top' as part of an effort by the 'Commission to show they were doing more' besides pursuing austerity (Interview 3). A national official concurred, reporting dissatisfaction with their addition, remarking it was a 'politically motivated choice' of the Commission (Interview 5).
This national official's dissatisfaction was widely shared. Indeed, the Commission's political pursuit of a social agenda through the MIP triggered a coalition of opposition. National finance ministries in the ECOFIN Council were resoundingly critical of the move. In 2013, the ECOFIN Council was hesitant about the addition of auxiliary social indicators to the scoreboard, commenting that any changes to the MIP needed to 'carefully preserve the nature of the procedure' and its focus on traditional macroeconomic risks of finance and competitiveness ( ECOFIN Council, 2014, p. 14). After the Commission's intensified push to emphasise social issues within the MIP from 2013 onwards, the Council was more forthright. The ECOFIN Council Conclusions of January 2016 expressed 'concern about the inclusion by the Commission of three additional employment indicators to the main scoreboard given the need to preserve the effectiveness of the scoreboard … social and labour market indicators are not relevant for identifying macro-financial risks' ( ECOFIN Council, 2016, p. 4).
The economically-oriented policy actors who predominantly undertook the technical work of the MIP saw the adoption of social indicators as simply bad economics in a procedure focused on imbalances. If Social issues were seen as alien to the economic question of imbalances. One national official spoke for many to say that 'they were not relevant in this specific context of the procedure' (Interview 5). It was not just the ECOFIN Council of national finance ministers that resisted the broadening of the MIP, but also the technical services of the Commission's DG ECFIN. One official said of the social indicators that 'if the idea is to prevent new imbalances then these indicators would not tell you very much' (Interview 3) While the opposition of economically-oriented policy actors to the socialisation of the MIP is perhaps unsurprising, opposition also stretched to actors within employment and social policy networks one would expect to have welcomed the move. Many within these networks were concerned that the inclusion of such indicators represented an attempted power grab from finance ministries (Interview 3). Conclusions from the Employment and Social Policy Council (EPSCO) in June 2015, for example, argued that issues 'concerning employment and social issues must be maintained within the EPSCO remit and should not systematically become part of the MIP process' ( EPSCO Council, 2015, p. 4).
The concern that the fusion of social indicators within the relatively legalized and economically-focused MIP would make national governments accountable to the EU on social policy issues united a broad range of national actors. Whereas the usual venues of social policy through Europe 2020 or the broader European Semester are rooted in voluntary action, the MIP's mechanism for implementation potentially leads to sanctionseven if the MIP in practice has been applied less aggressively. The integration of social indicators into the MIP was, therefore, seen by national officials to potentially open the door to closer scrutiny and accountability to EU institutions on a vast range of new policy areas related to social inclusion and unemployment.
Managing these concerns, while social and employment indicators grew in visibility in the MIP scoreboard, DG ECFIN made sure the new indicators were clearly placed in a subordinate position compared to the more traditional competitiveness-focused indicatorsjust as they had done with the original unemployment indicator. The Commission explicitly asserted in 2015 that social and employment indicators 'will not have legal implications nor change the focus of the MIP. Flashes of the new [social and employment] indicators would not be read as implying, by themselves, an aggravation of macro-financial risks, and consequently will not trigger further steps in the MIP' (European Commission, 2015a, p. 6). Instead, the purpose of social indicators is to 'allow better understanding of the risks of such imbalances … to identify policy measures to correct imbalances while minimising their social consequences' (European Commission, 2013).
Despite their vocal opposition, in the end national officials accepted the Commission's drive to expand the social orientation of the MIP. Many national governments felt political pressure to accept the social indicators, lest they were portrayed as an opponent to the ESM as trumpeted by the Commission. As one national official described, 'once someone has suggested it [social indicators], it lives its own life, and it is difficult to oppose' (Interview 6). After securing confirmation that the social indicators would have no procedural ramifications, their inclusion was ultimately accepted by national officials (Interview 2).
Making unemployment and social issues visible as a few indicators among many accommodated the Commission's political project aligned with the decades old ideal of the ESM. Yet, by ensuring that social indicators could not trigger the escalation of the MIP's sanctions, it reproduced the structural and institutional asymmetry of European integration. In the case of the MIP, the EU merely monitored the employment performance of member states without implying that any action would followin fact, real action through the MIP's legal mechanisms was explicitly ruled out.
A notable consequence of the politicking over the MIP's scoreboard by the Commission has been its general weakening as the hierarchical policy instrument of neoliberalization it was designed to be. The MIP was originally envisaged as a legal tool of crisis prevention, capable of spotting and then remedying macroeconomic imbalances. The scoreboard was meant to play an important role in this. It was not meant to be an automatic trigger for policy action, but rather a prominent early warning for a growing macroeconomic imbalance that could lead to a clearly defined escalation of procedural steps. In part because of the Commission's politicisation of their technical responsibilities surrounding the scoreboard, however, it has been progressively played down as a relevant factor of the MIP. As one national official commented to us, the scoreboard became 'a Christmas tree, [that] everybody wants to put something on' (Interview 1). Indeed, a 2017 European Court of Auditors report of the MIP outlined that, despite it originally being envisioned as leading to escalating enforcement steps, 'the scoreboard is of limited value at the AMR stage and none thereafter' (European Court of Auditors, 2017, p. 44). 5 The fate of the scoreboard reflects a broader development of the MIP that it has never worked as the legal instrument it was designed to be. Despite numerous countries tipping over into excessive imbalances, the MIP's disciplinary tools have never been used. The Commission has instead produced ever increasingly imaginative categories to describe macroeconomic imbalances as anything but 'excessive' to avoid a confrontation with states over the imposition of sanctions through the so-called Excessive Imbalance Procedure.
For the Commission, this sacrifice of the technical-legal potency of the MIP has reflected the dilemma it faces between its technical responsibilities and political legitimacy problems. The Commission has been stuck in a bind since at least the euro crisis that its legitimacy has been increasingly questioned through its involvement in dubious bailout practices, but it has been consistently denied new policy instruments or accountability mechanisms through which it could derive stronger claims of legitimation. Instead, as national governments remade EU governance through the crisis, the Commission was relegated to largely technical responsibilities of policy-surveillance -such as managing the MIP. Facing a continued legitimacy crisis, the Commission's pursuit of social indicators in the MIP is, as a result, a reflection of how it has explored ways through which it could instrumentalise its surveillance tasks for political purposes as a new basis upon which to ground its legitimacy claims.

Conclusion
Since the late-1990s, EU governance has been characterised by performance management and a growing fusion of economic and social policy-making. Rather than developing collective, supranational policy instruments in social and economic affairs, the EU has been governed through a constellation of practices that coordinate intergovernmental disagreement through quantitative techniques like benchmarking. Scholars have been divided on what this trend signifies. For some scholars, it reflects the potential 'socialization' of EU governance, where social and economic affairs are addressed on more equal terms (Zeitlin & Vanhercke, 2018). For others, it has reflected a thorough 'neoliberalization' of EU governance, where social issues are evaluated only in competitiveness or budgetary terms .
In this paper, we have contributed to studies of EU socio-economic governance by combining a close inter-institutional reading of struggles over the shape of EU performance management with a broader political economic reading of European integration. We have argued that the distinctive development of EU governance through performance managementwhere numbers rule, and social and economic policy are increasingly fusedis a reflection of two tensions: i) macroeconomic divergence and intergovernmental dissensus over the purpose of governance and ii) EU institutions' overlapping and contradictory technical responsibilities and political ambitions.
Looking into the specific case of the MIP, we have highlighted how discussions on the new procedure transformed from an emphasis on economic and technical aspects to much more political discussions on its social function. Rather than viewing this as a back and forth of de/repoliticisation, we argue it reflects how performance management has reframed political conflict to deal with the constraints imposed by intergovernmental dissensus. As we have shown, the European Commission's navigation of the multiple tensions arising from capitalist diversity, intergovernmental dissensus, and institutional reconfigurations led it to use the socialisation of technical policy surveillance to portray itself as a force of compromise rather than an austerity dogmatist in the heat of the euro crisis.
While we do not find a significant 'socialization' of the EU through performance management, our paper qualifies the arguments of CPE scholars who see the EU on a path to deeper 'neoliberalization' through depoliticization. The Commission's highly political pursuit of social objectives in the EU's performance management regime has blunted the edges of EU governance as a force of neoliberal structural adjustment. Testament to this is the failure of the MIP to bite as the legal disciplinary tool it was designed to be. Yet, because the Commission's efforts to boost social policy have been largely tied to reinforcing its own institutional legitimacy, the prospects for a reinvigoration of social protection justice remain slim.
Our argument has important implications for analysing the major changes underway to EU governance in response to the coronavirus pandemic. Performance management through scoreboarding is integral to the EU's pandemic response in the 'Recovery and Resilience Facility'. Whereas performance management has historically been an alternative in the EU to more instrumental forms of governance, the Recovery and Resilience's Facility is directly connected to explicit financial investments made into Member States on the condition of achieving performance requirements. The argument we have made here concerning the inter-institutional politicking that surrounds EU performance management suggests that scholars should be alert to how the new financial leverage of the Facility interacts with the longstanding tensions of EU governance we have discussed in this article.

Notes
1. The MIP's legal base is in two regulations: Regulation (EU) 1176/2011 on the prevention and correction of macroeconomic imbalances sets out the MIP procedure and applies to all EU countries covered by the MIP and Regulation (EU) 1174/2011 on enforcement measures to correct excessive macroeconomic imbalances specifies a sanction mechanism to enforce MIP recommendations for euro area countries 2. Indeed, the MIP closely mirrors the SGP in its design and functioning. 3. The Spitzenkandidaten process, for example, (where the Commission President was selected from the Parliament's largest party group) was discontinued after only one cycle in the face of national government pressure. 4. IDRs are reports published by the Commission to study a country deemed to be experiencing imbalances under the MIP in more detail. 5. The Alert Mechanism Report (AMR) is the first step in the annual cycle of the MIP where the Commission screens the scoreboard to determine which countries are facing imbalances.
Ian Alexander Lovering is a postdoctoral researcher at King's College London. His research looks at neoliberalism in the Eurozone and public sector reform.