Breaking the Budgetary Taboo: German Preference Formation in the EU's Response to the Covid-19 Crisis

Abstract

In the face of the coronavirus pandemic, the German government embraced a major shift towards a grants-based EU recovery fund relying on common European debt. How can we explain this impetus, especially in view of the reticent German fiscal stance in previous years and in the early stages of the pandemic? To elucidate this question, this paper provides a qualitative inquiry into German preference formation during the spring of 2020. Theoretically, it reconciles liberal intergovernmentalist and discursive accounts of preference formation in the context of EU politics stressing the intertwined nature and simultaneousness of preference formation in the national and European arenas. We hypothesise that, along with material self-interests, the construction and framing of the pandemic as a certain type of crisis was key. Examining the stances taken by the federal government, commercial groups and key EU actors such as France and the European Commission, our findings point to a rapid preference realignment in German political and economic circles. Overall, the analysis suggests that especially in times of crisis, assumed national preferences are subject to reconfiguration thus allowing for contingent political responses.

Speaking to a greatly reduced press corps in Berlin on 18 May 2020, Germany’s Chancellor Angela Merkel presented the blueprint for what would become the European Union’s (EU) most important response to the Covid-19 (corona) pandemic. Next to her on a large screen, cut in from Paris, was France’s President Emmanuel Macron. The French-German initiative of 18 May foresaw the European Commission borrowing €500 billion in the financial markets and allocating the money in the form of grants to the EU regions and economic sectors worst hit by the pandemic (Bundeskanzlerin 2020b). Practitioners and observers, including Germany’s Finance Minister Olaf Scholz, interpreted the French-German initiative as Europe’s ‘Hamiltonian moment’ – in reference to Alexander Hamilton, the first treasurer of the United States of America who issued joint bonds on the part of the new federal government – and as a significant step towards a European fiscal union (Die Zeit, May 19, 2020). Indeed, the initiative was followed, only nine days later, by a Commission proposal on an EU recovery fund to be endorsed by the EU’s 27 heads of state and government at their European Council meeting in mid-July.

This episode offers a stimulating puzzle in several respects. From a theoretical perspective, the EU’s financial response to the pandemic allows exploring yet another crisis, if we understand by crisis a moment where the EU faces an acute collective decision-making problem. On the one hand, due to the historic recession, a possible breakdown of the Eurozone and the Schengen area, and a manifold ‘rebordering’ of the EU’s single market (Genschel and Jachtenfuchs 2021), pressure for a coordinated response was high. On the other hand, member states’ preferences strongly diverged on which instruments should be activated or newly created, and decisions were likely to modify the status quo of resource distribution in important ways. The corona crisis thus adds yet another instance to the EU’s ‘poly-crisis’ (Zeitlin, Nicoli, and Laffan 2019) and the academic debate about which factors best explain the changes to the EU polity and the overall dynamics of the European integration process (Csehi and Puetter 2021; Schmidt 2020).

Analytically, the French-German initiative from May 2020 marked a remarkable deviation from previous policy stances, especially with regards to Germany, which had excluded EU debt and insisted on financial transfers between member states only in the form of loans. Since the Eurozone crisis of 2009-2012, common bonds had been a taboo in Germany and other Northern EU creditor countries because a mechanism allowing greater fiscal risk-sharing was believed to open the way to moral hazard and excessive spending. Still in early April 2020, the German government denied the need for a large European response to the corona crisis and it even turned down discussions about loans from the European Stability Mechanism (ESM), the Eurozone’s bailout fund. On 7 April, the Financial Times reported that Germany and its Northern ‘allies’ of equally fiscally conservative EU countries were satisfied with small European measures and would oppose common debt. With EU politics apparently locked again into the creditors versus debtors confrontation, how can we explain the eventual shift in German preferences, away from a deep-seated reluctance vis-à-vis a Transferunion towards a major impetus involving a quasi-doubling of the EU’s fiscal capacity and the creation of common debt?

To answer this question, we trace German preferences during the most critical phase of the pandemic, that is from early March 2020, when the virus started spreading across Europe, until late May 2020, when France and Germany (followed by the European Commission) presented their proposals for the EU’s financial response. In theoretical terms, we critically revisit the debate on national preference formation in the EU in order to bring together two perspectives which too often have been considered to compete or even contradict each other, namely liberal intergovernmentalism and discursive institutionalism. We hypothesise that established national preferences, shaped to a significant extent by economic interests and material concerns, need to be understood in relationship with the deliberative process of problem construction, that is, the perception and framing of a given event as ‘crisis’. Furthermore, this process unfolds simultaneously across the national and the EU political arenas. Our qualitative case study supports this hypothesis and shows how German political and economic preferences were rapidly re-shaped from a national focus rooted in a creditor country’s stance towards a European response implying institutional innovations and increased EU resources.

In the first section, we give a synthetic overview of the recent theoretical debate on national preference formation against the background of the EU’s poly-crisis and stress how liberal intergovernmentalist explanations of EU politics and a discursive approach highlighting problem construction can complement each other. Next, we present the methodology and data underpinning the analysis. In the third section, we trace German preference formation and show how national political and economic preferences were (re)shaped to support a debt-financed EU recovery fund. We conclude with a summary of our main findings and suggestions for future research.

The EU’s Poly-Crisis And The Theoretical Debate On Preference Formation

The outbreak of the coronavirus pandemic has fed into the vivid theoretical discussion about the EU’s responses to a succession of ‘crises’. As scholars have sought to explain how EU leaders proved able to reach agreements and give new impetus to integration or, to the contrary, how Europe got caught in political deadlock, this debate returned to a classical theme of European integration theories, namely preference formation. Not least because of its elegant sequential framework of national preference formation, intergovernmental bargaining and institutional choice, and its corresponding straightforward causal assumptions, liberal intergovernmentalism (LI) has taken up the position of a ‘baseline theory’ in European integration scholarship (Moravcsik 2018). Essentially, LI assumes rational member states to pursue national material (that is, economic and financial) interests within the EU context. Member states will promote European cooperation if and to the extent that they expect individual gains to outweigh the involved costs and risks. National governments, and political leaders in particular, are the most important actors. Prior to their EU-level bargains, governments determine national preferences, which reflect the demands of important domestic commercial groups and the country’s macro-economic (fiscal) position (Moravcsik 1998).

During the EU’s poly-crisis over the past decade, scholars regularly have applied LI theory. Regarding the Eurozone crisis, for example, Schimmelfennig (2015) has argued that their better fiscal position gave the ‘creditor’ countries around Germany more bargaining power, enabling them to get their way in intergovernmental negotiations and policy reforms and passing on adjustment costs to ‘debtor’ countries. Moreover, comparing the changes in integration during the Eurozone crisis to those during the migration crisis, Biermann et al. (2019) contend that different member-state preferences to support or object EU action in these two crises resulted from different national exposures to financial and migratory pressures. The different preferences, in turn, led to distinct actor constellations and bargaining dynamics. We concur with the assumption that member states’ most basic concern in EU-level interaction is to promote national preferences and that national leaders are the most important actors in this endeavour. Below, however, we point out several shortcomings in the way LI accounts for preference formation in EU politics.

Considering German preferences over the integration of core state powers at the EU level since the introduction of the Maastricht Treaty in 1993, Freudlsperger and Jachtenfuchs find that ‘in cases of high interdependence, German state elites consistently supported the supranational regulation of national capacities’ (2021, 129) and that only in existential crises did Germany endorse European capacity-building, albeit under intergovernmental control. Bringing with it significant capacity-building only partly under the control of the Council (through the European Semester) and breaking the taboo of common debt issued through the EU budget, the corona recovery fund goes a long way in pushing such preferences to their limits. Moreover, as these authors stress, the recognition of an ‘existential crisis’ alters preference formation in important ways. Building on these debates, we address three weaknesses of LI in particular.

First, most accounts using LI to explain the EU’s crisis responses do so after national preferences had long been formed and the bargaining results have become obvious. While they are concerned with concrete moments in time and clearly defined outcomes of European integration, they hardly look at the very process of national preference formation itself in the face of a crisis (Biermann et al. 2019; Schimmelfennig 2015). This is understandable, given Moravcsik’s (1998, 36) claim that LI theory works better the more ‘intense, certain, and institutionally represented and organised’ societal interests are and the more certainty there is about the substantive implications of a particular choice. Arguably, however, the ‘normal history’ of European treaty amendment rounds against which LI had been established, differs greatly from moments of crisis characterised by high levels of threat, uncertainty, and urgency. Especially in the early stages of an escalating crisis, policymakers usually struggle to come up with the adequate measures and to agree on policy solutions (Boin et al. 2016, 7–9).

Second, following earlier critiques of LI (Csehi and Puetter 2021; Puetter 2012), we contend that the formal sequence of national preference formation followed by intergovernmental bargains does not capture the nature of EU crisis politics. The metaphor of ‘two-level games’ suggests that political agreements at the international (EU) level result from rational calculations by national negotiators and overlapping sets of preferences forged at the domestic level (Putnam 1988). However, the boundaries between national and EU politics have become increasingly blurred as public spheres, deliberative processes, and communication tend to overlap both vertically and horizontally. As a result, the conceptualisation of preference formation as a ‘two-level game’ must be updated and conceived rather as a ‘double game’ (Crespy and Schmidt 2014) or a ‘nested game’ (Fontan and Saurugger 2020) taking place simultaneously in the national and the EU arenas.

And third, we put into perspective LI’s assumption that national preferences are primarily determined by powerful domestic interest groups and, once established, remain stable during the intergovernmental bargains. While we agree that political leaders are the key agents and that they aim at acting rationally on the basis of cost-benefit calculations, we contend that those are far from straightforward. In the case of the corona crisis, for example, the magnitude and implications of the pandemic for individual member states and the EU as a whole were very unclear at the beginning. Tellingly, it took economic interests several months to assess their situation and formulate specific demands towards their governments for EU-wide action. Agents therefore find themselves torn between the sense of urgency to act, on the one hand, and the difficulty to assess the potential consequences of a given course of action, on the other.

To tackle these issues, we bring in discourse, and more specifically problem construction, as a key mechanism shaping preference formation. A major tenet of constructivist approaches to politics is that agents must decide and act against the background of uncertainty, which opens the way for communicative and ideational factors to influence policymaking (Béland and Robert 2010). This does not mean that agents do not form their preferences in a self-interested fashion, but rather that the definition of those interests itself occurs through a social and political construction. Emphasising how preferences are not only shaped but also legitimised, contested, and reconfigured through discourse, Schmidt’s discursive institutionalism (2008, 2020) has been similarly influential in the study of EU policymaking. In this perspective, discourse is both (a) an ideational process aiming at generating meaning over events and political action and (b) an interactive process through which agents seek to persuade each other and influence policymaking. To make our claim more specific, we focus on problem construction as a particular type of mechanism which powerfully shapes preference formation.

Problem construction is a widespread explanatory factor in various streams of political science, ranging from mainstream policy analysis (Kingdon 1984) to narrative approaches to policymaking (Jones and McBeth 2010), including the EU (Daviter 2007) or the study of grassroots mobilisation (Benford and Snow 2000). Focusing on problem construction through perception and framing does not only challenge a central premise of rationalist approaches, namely the existence of fixed and stable preferences (Schmidt 2008, 304; see also Puetter 2012). It further implies that material interests and discourse (perception and framing) do not exist separately from one another. Rather, they are co-constructed as agents’ preferences evolve when they find and define what their self-interests are in a given situation.

Combining LI’s focus on preference formation with a discursive perspective stressing the importance of communication, we hypothesise that the way in which the pandemic – or the corona ‘crisis’ – was constructed, namely as a political problem requiring a common European answer, shaped the reconfiguration of German preferences in the pivotal months of spring 2020. Problem construction implies both the perception and framing of the pandemic by key political actors, which, especially in its early stages, were the federal government and Chancellor Merkel in particular. While perception refers to the cognitive and ideational apprehension of events, framing points to the discursive contents and instruments in the deliberation about problems and solutions.

Defining ‘what the problem is’ is closely linked to how agents perceive threats and opportunities as they seek to make rational calculations about ‘what to do’ (Jones 2009). Especially during crises, which imply an acceleration of political time and where decision-makers need to act fast in response to pressing problems, national preferences including financial and economic concerns are difficult to overview and assess. ‘Crisis’, after all, is not a given fact: Arguably, the higher the level of uncertainty about the impact and further implications of a certain threat, the more perception, interpretation and framing matter (Boin et al. 2016, 3–18). Importantly, the pandemic was largely perceived as an exogenous, symmetric shock – and framed as such. However, if Covid-19 indeed suddenly affected all EU member states, its impact (in terms of death toll, for example)¹ was never completely symmetric since wealthier countries were clearly less badly affected than poorer ones. This was particularly the case in the early stages of the pandemic in spring 2020, which is our period of interest. Figure 1 below illustrates our hypothesised intertwined role of material interests and discourse in problem construction during the formation of national preferences, especially in times of crisis.

Figure 1. The role of problem construction in national preference formation during ‘crises’.

Another central claim of discursive institutionalism is that the explanatory power of discourse must be considered in specific institutional settings. The EU, with its multi-level governance structures and the multitude of different actors involved, arguably represents a distinct institutional setting for preference formation. Empirically, this means that deliberations take place in Germany and at the EU level at the same time. Consequently, we follow Schmidt’s (2008, 310–313) distinction between a ‘coordinative’ discourse and a ‘communicative’ discourse. Coordination points to the policy sphere and elites interacting with one another about the understanding of events and the construction of policies; it is especially prominent at the EU level where leaders deliberate to come to an agreement (or a compromise) on ‘what is going on’ and ‘what to do’. Communication, on the other hand, refers to the political sphere and elites engaging with parliamentarians and the broader public about the necessity, appropriateness, and legitimacy of specific policies; it is especially important in the domestic arena where leaders must make sense of what has been decided ‘in Brussels’.

Two further points are important. First, what are the scope conditions for our hypothesis to hold? In principle, problem construction always matters for preference formation. However, we expect it to matter most (a) in the early stages of an escalating crisis and (b) in a context of high member-state interdependence. In the early stages of a crisis, characterised by the novelty of a situation and the absence of established facts and procedures, the level of uncertainty is highest; by contrast, once frames are established, re-framing issues proves more challenging. And in the event of a shock affecting one or several EU core policy regimes – Eurozone, Schengen area, and single market – several member states seek to take the lead in the framing contest and have an interest in particular political action. Second, our argument does not make Germany a unique case. Instead, we concur with Freudlsperger and Jachtenfuchs (2021) that Germany is mostly ‘a member state like any other’. In that sense, the tracing of problem construction and framing in other countries, for instance, the Netherlands, might prove conducive to contrasted preferences regarding EU fiscal politics. What is equally true, however, is that Germany – due to its overall size, large fiscal resources, and central position in the EU – will be particularly called upon by other actors to take a stance on the issue at stake.

Methods And Data

As pointed out by Csehi and Puetter (2021), the bulk of the literature on preference formation in the EU relies on stylised accounts of events. When seeking to understand how preference formation occurs in a contextualised manner, however, individual qualitative case studies remain key, at least as a first-stage methodology (Kassim, Saurugger, and Puetter 2020). Exploring recent important events, we tap into this line of scholarship, conducting a qualitative inquiry to account for German preference formation during the pivotal period between March and May 2020.

We put a particular focus on the precise sequencing of events, the position-taking by crucial actors, the perception of the pandemic, and the policy implications of its framing. The actors considered essentially are German commercial groups and political parties, the federal government, the French government, and the European Commission, all of which were closely involved in German preference formation during the corona crisis and the deliberations on the EU recovery package. Reflecting our notion that national preference formation and EU-level bargains happen simultaneously, we thus take a close look at the crucial actors at different levels of government. While new intergovernmentalist scholars focus on multilateral deliberations within the (European) Council (e.g. Puetter 2012), we shed light on the political dynamics happening at different levels of government including the national, bilateral, and EU arena. We find that coordination between the German government, the French government, and the European Commission in the early stages of the crisis proved essential and in fact necessary for the consensus later reached among all member state governments.

Using a process-tracing methodology, which is summarised in Table 1 below, we identify three important analytical sequences – namely crisis acknowledgment, crisis rescaling, and crisis management legitimation – unfolding between March and May 2020. In each sequence, we observe how perception and framing contribute to constructing the pandemic as a common problem affecting all member states and requiring a solidary response. We further use the categories of ‘coordinative’ and ‘communicative’ discourse to stress the intertwined deliberative and legitimation dynamics across the German and the EU arenas (see Table 1).

Table 1. Process tracing methodology.

	Sequence 1 Crisis acknowledgement	Sequence 2 Crisis rescaling	Sequence 3 Crisis management legitimation
Shaping mechanism in preference formation	Problem construction Assessing national preferences	Problem construction Reconfiguring national preferences	Problem construction Pursuing renewed national preferences
Arenas and types of discourse	National/German arena
	Coordinative and communicative discourse focused on national level	Emergence of communicative discourse focusing on European nature of the crisis	Salient communicative discourse centered on EU-level solutions
	EU arena
	Conflicting coordinative discourse, little communicative discourse	Emergence of coordinative discourse focusing on European nature of the crisis	Coordinative discourse focusing on specific EU policy solutions
Observable manifestations	Absence of EU in discourse, national understanding of pandemic’s implications Calculations according to own immediate affectedness and fiscal situation Activation of national instruments and EU instruments inherited from previous crises	Nature of crisis (type, origin) and implications (emotions, solidarity) Calculations: impact of crisis (new economic forecasts) and implications (nat. costs and benefits of common EU response) Deliberations on new EU instruments to tackle recession	Deliberation about appropriate EU policies Striving for domestic support/legitimacy Strategic deliberations aiming at enforcing national preferences

Process tracing is the key method employed in qualitative within-case analysis and focuses on identifying causal mechanisms leading to the outcome of interest (Beach and Pedersen 2019). A crucial point is that process tracing implies considering counterfactuals, seen as ‘plausible alternatives’, which however did not occur in the studied case (Collier 2011, 825). In our case, we see no determinism as to why the EU recovery fund had to come about. Had the initial, tentative framing suggested by Dutch political elites (as we explain below) met resonance among other EU leaders, the crisis may well have been framed as an asymmetric shock and led to different cost-benefit calculations. The specific process of problem construction we see unfolding in our case therefore is crucial. Process tracing has been a bridge-builder between positivist and interpretivist scholars in their way to think about causality and inference (either causal or descriptive; cf. Vennesson 2008). In line with this bridge-building stance, we adopt a soft conception of causality in order to shed light on both the ‘why’ and the ‘how’ questions and with the broader aim of ‘teasing out the more fine-grained distinctions and connections between alternative theoretical schools’ (Checkel 2005, 16).

The reconstitution of events and position-taking as well as the pertaining framing of problems and solutions relies on a variety of sources which we triangulate. First, we systematically reviewed the German press (Frankfurter Allgemeine Zeitung, Handelsbatt, Süddeutsche Zeitung), French press (Le Figaro, Le Monde) and international press (Euractiv, Financial Times, Politico) deploying the search words ‘Germany’, ‘EU’, ‘corona’ and ‘recovery’ for the time period of interest. Their extensive reporting on EU governance matters was crucial not only to trace the course of events but also to find out when exactly the different actors took a stance publicly. Second, we retrieved the public speeches from Chancellor Merkel and Finance Minister Scholz to trace how the German government’s positions developed – and changed – during the time under scrutiny, how the crisis was perceived and framed, and when and how frames were communicated to the German parliament and public. And third, we conducted 13 semi-structured interviews with German, French, and EU officials as well as German policymakers and business representatives. The interviews proved extremely helpful to refine the narrative account with unofficial information about position-taking and perception of the crisis, which eventually led the German government to advocate a major EU fiscal stimulus. To obtain relevant information, the interviewees were guaranteed confidentiality.

Tracing German Preferences In The EU’s Response To The Pandemic

Sequence 1: Crisis Acknowledgment – Assessing and Asserting National Preferences in the Immediate Reaction to the Crisis

With the pandemic fully breaking out in Europe, German policymakers were primarily concerned with taking action in the national context. Measures taken in the EU arena essentially concerned the activation of existing policy instruments. Yet, there were already signs that deliberations and negotiations would not simply reproduce conflict lines between ‘creditors’ and ‘debtors’ and that the looming corona crisis would be ‘different’ in several ways. In this sequence, both the coordinative discourse among governing elites and the communicative discourse towards the German public reflected an acknowledgment of the pandemic and its implications essentially in national terms.

When the novel coronavirus began to rapidly spread across Europe from early March 2020, EU member states took primarily national measures for its containment. One country after the other closed its borders, de facto suspending free movement inside the Schengen area and causing interruptions to economic supply chains and labour migration in Europe’s single market. The decision in mid-March by Germany and others to ban the export of medical equipment to the hardest hit countries in the EU’s South led to outrage and the accusation of national egoism. In her televised address to the nation on 18 March, Merkel spoke of the pandemic as the greatest challenge for Germany since the Second World War but did not mention the word ‘Europe’ a single time. Similarly, in a joint paper by the ministers of finance and for economic affairs on measures to fight the economic damages caused by the virus, the EU was only given a subordinate focus and role (Bundesfinanzministerium 2020).

At the EU level, Finance Minister Scholz on 16 March even turned down discussions about loans from the ESM as inadequate and ‘premature’ (Reuters, March 17, 2020). Due to its own large fiscal resources, the German government could, and did, provide massive support programmes for its economic sectors and citizens.² Governments’ takes on the crisis and their positions on European measures at this point reflected their immediate affectedness and own budgetary situation, with the hardly hit, primarily Southern member states advocating common fiscal action, while the, at first, rather spared Northern countries suggested a more cautious approach. The division of member states into different, seemingly opposing camps thus was about to reproduce patterns in EU fiscal policy seen in the past, notably in the heated debates about Euro bonds, hence the joint issuance and liability of government debt (Matthijs and McNamara 2015).

In the EU arena, at this stage we see mainly conflicting coordinative discourses and no consistent communicative discourse towards the national publics. At the European Council meeting on 17 March, the first since the outbreak of the coronavirus in Europe, leaders noted the unprecedented nature of the crisis and called for economic support for the hardest-hit regions. However, they strongly disagreed on the terms and scope of this support. The discussions gathered further pace when a week later, nine countries including Italy and Spain but also France, in a letter to the European Council President, called for the introduction of ‘corona bonds’ (Euractiv, March 25, 2020). At the next European Council on 26 March, leaders again clashed over the adequate European response, with Germany and other Northern countries strictly opposing common debt and now advocating the use of ESM loans.

At the same time, reports revealed how, during a Eurogroup meeting in late March, the attempt by the Dutch Finance Minister to re-activate frames known from the past blaming the Southerners for the sheer scale of the crisis in their country, triggered strong criticism among his peers (Politico, March 27, 2020); a few days later, Wopke Hoekstra had to apologise publicly. Several interviewees (1, 5, 9) stressed that, given the emotional context of spring 2020 including thousands of deaths and severe restrictions to the social and economic life, ‘this crisis was different’ in that everyone was affected and nobody could be held ‘responsible’ for the unexpected outbreak of a global health emergency stemming from China.

The German government’s focus on the national level corresponded with that of its economic lobby. Statements issued by the leading producer groups in March and April were equally concerned with the government’s actual or further measures in the national or even sub-national context. This echoes Benoit and Hay’s (forthcoming) claim that the pandemic breakout triggered a ‘“sovereigntist reflex” privileging the expression of national capacities and national self-reliance’ as opposed to a ‘more “perforated” understanding of sovereignty stressing the interdependence of peoples and states, both geographically and institutionally’. We find that the latter line of reasoning rapidly gained traction, leading the German government to reconfigure national preferences.

Sequence 2: Crisis rescaling – Reconfiguring National Preferences by Perceiving the Nature of the Crisis as a European One

With the number of corona infections and deaths on the rise, German policymakers got more vocal on Europe, both at the national and the EU level. The pandemic was perceived as a common threat requiring common action. This went hand in hand with important economic interests progressively taking a stance in favour of a large-scale European financial response to the escalating crisis. Both in the German and the EU arenas, we find coordinative and communicative discourses signifying a rescaling of the pandemic and its implications towards the EU level.

Engaging with a Europe-wide communicative discourse, Scholz, together with Foreign Minister Maas, in a joint article published in several European newspapers on 5 April called for common European action ‘in the spirit of solidarity’ (Auswärtiges Amt 2020). A similar shift occurred in the coordinative EU arena. Siding with a group of other member states in their call for corona bonds, France had increased the pressure on the German government to reconsider its policy stance and play a more active part in the EU crisis management. Given their different initial takes and positions, German and French policymakers and civil servants from mid-March started intensifying their bilateral consultations trying to sound out common ground (see also Seidendorf 2020). The two countries, thanks to their ‘embedded bilateralism’ in Europe (Krotz and Schild 2013), can draw from a unique set of close political, administrative, and diplomatic ties. While for Germany the ‘natural’ candidate to align its position with was France (Interview 10), the French – themselves advocating an ambitious EU response involving joint borrowing – were aware of the German ‘red lines’ when it came to fiscal risk-sharing (Interviews 1, 8).

A first result of this bilateral coordination became visible on 9 April when the Eurozone finance ministers agreed on a corona support package worth €540 billion consisting of loans and guarantees for workers, enterprises, and health-related state expenditures (Eurogroup 2020). According to both insiders and observers, the German and French finance ministers, Scholz and Le Maire, were crucial for brokering the deal, overcoming the resistance of other countries to parts of the package (Interviews 5, 13). As the Financial Times put it on 13 April, the ‘coronavirus crisis revives Franco-German relations’. The package also included a request to the European Commission to soon present a proposal for an EU ‘recovery fund’.

In Germany, too, the coordinative discourse within economic circles was undergoing fundamental change. As the crisis was affecting more member states more severely than previously expected, German businesses and industry perceived the pandemic as a threat to German material interests. In view of the ongoing border closures, their biggest focus had become to safeguard trade and supply chains as well as labour migration in Europe (Interview 4). Fears of a ‘rebordering’ of the European single market (Genschel and Jachtenfuchs 2021), in combination with Brexit, and uncertainties about the cohesion and stability of the Eurozone, led to a reorientation of previous priorities. A major concern shared by political and economic actors (Interviews 4, 10) was the prospect of Italy – at the time the hardest hit EU country in terms of absolute infection numbers and, due to its precarious fiscal situation, facing sharply rising interest rates – not managing to refinance its state expenditures, thereby causing distortion in both the Eurozone and the single market.

As a consequence, the German BDI, the French MEDEF, and the Italian Confindustria – the three leading industry associations in the three largest EU countries – on 12 May presented a joint position paper in which they expressed their concerns about the economic impact of the pandemic, risking jeopardising fair competition in Europe. Especially for the BDI, demands for ‘a level of public support previously unknown in times of peace’ and for a large-scale EU investment ‘with a balanced ratio of loans and grants’ (BusinessEurope 2020) differed from statements issued in the past. Unlike the Eurozone crisis, for instance, the corona pandemic was putting at risk Germany’s own economic stability and sales markets. Studies showed that German exports to other EU countries during March 2020 – the first month of the EU-wide lockdown – dropped by eleven percent, the largest decline in 30 years (Handelsblatt, May 8, 2020).

The notion and call for greater European solidarity now featured more prominently than ever before in the German political discourse. Speaking to the Bundestag on 23 April, Merkel called for European ‘solidarity’, also in financial terms, since this crisis was ‘no one’s fault’ and threatened Germany’s own economic situation (Bundeskanzlerin 2020a); here, we find moral and interest-based reasoning to merge into each other. Merkel further stated that her government was now ready to make significantly higher contributions to the next Multi-annual Financial Framework (MFF), the EU’s long-term budget, which had to be agreed before the end of 2020. Only a few weeks earlier, negotiations on the MFF had failed because Northern countries around Germany had called for a reduced EU budget. The corona crisis and the corresponding discourse clearly had changed the political context and Germany’s policy priorities. Explaining her new stance, Merkel argued that the corona pandemic was the EU’s ‘ever-biggest crisis’ and as an external event affecting everyone so that a common European response was both needed and appropriate (Bundeskanzlerin 2020a). Some days before, she for the first time in public had hinted to be in favour of issuing EU bonds, which however must happen within the limits of the existing treaties and the regular EU budget (Politico, April 20, 2020).

By the end of April, we thus see a considerable change in German political discourse and action. Next to the economic concerns about the unprecedented recession in Europe, perception mattered in that the crisis was interpreted and internalised as exogenous it its origins, no one’s fault, and affecting everyone. Notions of moral hazard and free riding, which had locked in previous debates about EU fiscal risk-sharing, did not apply and were not invoked. Furthermore, we find a broad consensus between the German government and domestic economic actors in, eventually, advocating a comprehensive EU fiscal stimulus. But other than LI would suggest, the timing of events shows that economic actors did not determine government action. It is only in late May and early June that both the Association of German Employers (BDA 2020) and the Association of German Chambers of Industry and Commerce (DIHK 2020) took a clear stance on European measures by welcoming the French-German and subsequent Commission proposals for an EU recovery fund. More important for the formation of German preferences was how the pandemic, its nature and (material and moral) implications were perceived by the government, constructed as a common threat requiring common action, and the simultaneous intergovernmental deliberations, most notably at the bilateral level with France.

Sequence 3: Crisis management legitimation – Pursuing National Preferences by Framing a Necessary Common Response to the Crisis

In this section, we further show how preference formation in Germany was closely intertwined with bilateral and broader European deliberations. In the domestic arena, the federal government embedded the definition of a specific course of action for crisis management in a communicative discourse. Framing the country’s commitment to a solidarity-based European policy response as a necessity rooted in both economic and moral motives, Merkel and her cabinet were striving for legitimacy and support from political parties and the broader public.

From the outset, Commission President von der Leyen called for an ambitious financial response to the pandemic through a European ‘recovery instrument’ linked to the next MFF (Von der Leyen 2020). This position was underpinned by three motivations: First, there was a need to re-launch the negotiations on the MFF which had broken down just before the outbreak of the pandemic in February 2020. Member states thus would have to come back to the topic and find a solution anyway. Second, the Commission sought to overcome the intergovernmental deadlock and heated debates paralysing the European Council between corona bonds and ESM loans with alternative recipes (Interviews 3, 7, 9). And third, linking a future recovery instrument with the EU’s regular budget was more likely to achieve German support than new fiscal instruments outside the Community framework (Interviews 2, 5, 13).

From late March, the coordination between the Commission and the capitals intensified. Beyond the usual contacts between Commissioners’ cabinets and Directorate-Generals, especially Budget and ECFIN, von der Leyen held close direct talks with the German Chancellery and the French Elysée, the primary purpose being to figure out options for how to link a recovery fund with the next MFF. Within the Commission, a small group of officials was set up and tasked to design several proposals under strict confidentiality (Politico, July 15, 2020). Deliberations on concrete steps, meanwhile, happened at the highest political level, with Merkel, Macron, and von der Leyen being directly involved (Interview 3).

Reflections on the form and details of the EU recovery fund also continued inside member states. Already on 1 April, the French Finance Minister Le Maire had called for the issuance of common debt as an ‘exceptional and temporary’ measure and a stand-alone fund (Financial Times, April 1, 2020). For Germany, however, the Le Maire plan as it stood was unacceptable since it came too close to ‘corona bonds’. Instead, the German government was determined to create a recovery fund within the regular EU budget and through a decision to lift the EU’s ‘own resources’, which would be a temporary measure, require unanimity, and would hold member states liable only for their share of the budget (Interviews 2, 5, 13). The government thus underlined long-time German positions of limited national liability for EU fiscal resources and the need to secure broad parliamentary ratification and control (cf. Howarth and Schild 2021). Progressively, through their close bilateral contacts and the accompanying operational work by the Commission, French and German positions by mid-May had come close to each other.

Meanwhile, framing the crisis as a public health emergency in the first place and a common threat requiring a common response, the federal government sought to convince important domestic political actors and legitimise the EU policies to come. The high death rates across Europe and the directly noticeable consequences of the pandemic for individual businesses and citizens, in turn, made German parliamentarians and the broader public receptive for the arguments of this communicative discourse (cf. Bremer, Genschel, and Jachtenfuchs 2020). Despite Germany’s unprecedented commitments in EU fiscal policy and the breach of its well-known budgetary taboo on common debt and direct financial transfers between member states, there was no party-political or popular backlash.

Indeed, the eventual French-German initiative from 18 May was welcomed across the German political spectrum. Despite their rejection of a European Transferunion, stressed numerous times in party programmes and speeches, the Conservatives from CDU and CSU endorsed the recovery fund. Picking up the framing of the government, their parliamentary group issued a statement only hours after the plan’s announcement, underlining their ‘support [to] Chancellor Merkel and President Emmanuel Macron’s joint initiative as a major contribution to European solidarity in the corona crisis’ (CDUCSU 2020). As several interviewees explained, notably the conservative parliamentarians had been briefed regularly on the ongoing French-German talks by Merkel herself, receiving the assurance that the recovery fund would be a one-off instrument exclusively for the fight against the pandemic and would not imply Germany to become liable for other member states’ debt (Interviews 6, 11, 12).

As a bilateral initiative, the Macron-Merkel plan did not specify how exactly the money would be raised, distributed, and paid back. On 27 May, only nine days after the French-German advance, the Commission presented its ‘official’ proposal for an EU recovery fund – termed ‘Next Generation EU’ – together with a renewed suggestion for the next MFF. The proposal largely built on the French-German plan, with the Commission to the €500 billion in grants adding another €250 billion in the form of loans (European Commission 2020). Originally, the proclamation of its proposal had been scheduled for the last week of April. It got postponed several times due to remaining technical details, but also for political reasons to sort out and meet national concerns (Interview 7). With their prior initiative, France and Germany provided some much-needed political weight and momentum to the upcoming Commission proposal, preparing the stage for the final EU-level negotiations to come (Krotz and Schramm 2021).

The reconfiguration of German preferences does not mean that the government ceased to act in a self-interested manner. On the one hand, in view of the nature, immediate impact, and further implications of the corona crisis – not least for Germany itself – the government perceived, and framed, common and comprehensive EU financial measures as the most appropriate and efficient response. On the other hand, the financing of the proposed recovery fund, as well as its limits in terms of size and duration, reflect longer-term German material interests and positions in EU fiscal politics. We again find moral and interest-based motives coming together as the federal government strategically deployed a coordinative as well as a communicative discourse which proved successful in pre-empting any broader dissent in the German public sphere.

Conclusion

This paper has traced German preferences with regard to what became the EU’s most important response to the corona pandemic. Doing so, we have contributed to the broader theoretical debates on preference formation in EU politics, especially in times of ‘crisis’. Starting from a liberal intergovernmentalist understanding of preference formation, we suggest that a rationalist approach can be enriched – rather than contradicted – by a discursive perspective emphasising how problem construction is inherent to preference formation.

The empirics presented show how the perception and framing by political elites served to construct the very nature of the crisis as one requiring a common European response based on solidarity (as a principle) and a comprehensive recovery fund (as a means of action). In an emotionally loaded context, German political elites framed the crisis as a public health emergency and an exogenous shock not involving any responsibility or wrongdoing. At the same time, policymakers made clear Germany’s own economic interests in a comprehensive European response and that the form of this response would largely be on German terms. Discourse and material interests therefore went hand in hand, with problem construction serving to underline, rather than overlook, economic concerns as German preferences were reconfigured.

In theoretical terms, the paper highlights three types of findings. First, we have traced the very process of national preference formation in the case of Germany. Especially in situations involving high levels of threat, urgency and uncertainty, perception and framing matter for what the problem is and what should be done. Second, we provide evidence for how national preference formation and EU-level deliberations fed into each other. In the corona crisis, the German government engaged in particularly close contacts with France and the European Commission seeking common ground for a joint position and to enable EU-wide action. And third, we have shown how national economic interest groups aligned with, rather than shaped, government positions. When rapid answers to pressing problems are needed, national governments (continue to) enjoy considerable leeway from other domestic actors.

What are the further implications of our findings for European integration theory and EU fiscal politics? Due to the overall dynamics during the crisis and the broader European compromise on the recovery package, we would expect perception and framing to have featured prominently in other member states as well. It would be particularly interesting to look at preference formation in those countries which featured as the ‘frugal four’ (Austria, Denmark, the Netherlands, Sweden) prior to and during the European Council deliberations on the MFF and Next Generation EU in July 2020. A related line of investigation concerns the similarities and differences between the 2009–2012 Eurozone crisis and the 2020 corona crisis. Such comparative analyses could increase the external validity of our theorised mechanism about the role and impact of discourse on national preference formation and would help refining its scope conditions. In addition, and beyond problem construction, is there a deeper process of cognitive and/or normative change at stake among decision makers from one crisis to the other? If some ‘learning’ has taken place (Ladi and Tsarouhas 2020), what lessons exactly have been learned, by whom and for how long?

In a different vein, our paper also raises the question of the significance of the recovery fund agreement for the future of the EU and its fiscal policy. There are various calls already from France and EU institutions to turn the recovery fund into a permanent instrument. Some scholars too see a ‘paradigmatic change’ to happen (Schmidt 2020). However, the recovery fund itself is limited in scope and explicitly designed as a one-time instrument. Thus, there are good reasons to assume that a lot will depend on the ‘recovery experience’, that is, how effectively the money will be spent and targeted towards future-orientated policies. Whether Next Generation EU represents a critical juncture in the EU’s socio-economic governance therefore remains an open question. Finally, we do not see the pandemic as a unique crisis. As already pointed out by experts, sanitary and environmental disasters are likely to become more frequent in the future. The episodes of floods, droughts and other plagues ensuing from climate change will affect all countries, although Southern European economies will be especially affected by global warming, which in turn might reinforce existing imbalances (Politico, July 2, 2021). The discursive construction of ‘crises’ and the management of interdependence will therefore be a key matter for EU politics in the future.

Acknowledgements

Previous versions of this paper were presented at seminars and workshops at the Université catholique de Louvain, the European University Institute, the Swiss Political Science Annual Conference, and the Universität Konstanz. We are very thankful for the valuable feedback and further suggestions we received from numerous colleagues, especially to Daniele Caramani, Claire Dupuy, Clément Fontan, Philipp Genschel, Adrienne Héritier, Antoine Lapeyre de Cabanes, Dirk Leuffen, Frank Schimmelfennig, Virginie Van Ingelgom, and Thomas Winzen. We are also indebted to the two anonymous reviewers who led to us to considerably sharpen the first draft of the paper.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Amandine Crespy

Amandine Crespy is Associate Professor of Political Science & European Studies at the Université Libre de Bruxelles (ULB) where she is affiliated with the Cevipol and the Institut d’Etudes Européennes (IEE). She is also Visiting Professor at the College of Europe (Bruges). Her research deals with socio-economic governance and policies in the European Union with a focus on the role of ideas, discourse and conflict. She is the co-editor of ‘Governance and Politics in the Post-Crisis European Union’ (2020, Cambridge University Press) and the author of ‘Welfare Markets in Europe’ (Palgrave, 2016) and ‘The European Social Question: Tackling Key Controversies’ (Agenda, 2022).

Lucas Schramm

Lucas Schramm is a Ph.D. Researcher at the Department of Social and Political Sciences at the European University Institute in Florence. He holds a master’s degree in European Political and Administrative Studies from the College of Europe in Bruges. In his doctoral dissertation, he analyses past and more recent political crises in the process of European integration, seeking to explain the variation in crisis outcomes. His recent publications include ‘An Old Couple in a New Setting: Franco-German Leadership in the Post-Brexit EU’ (with Ulrich Krotz, Politics and Governance, vol. 9 no. 1, pp. 48-58) and ‘Exit from Joint-decision Problems? Integration and Disintegration in the EU’s Recent Poly-crisis’ (European Review of International Studies, 2020, vol. 7 no. 1, pp. 2-27).

Notes

1 The number of cumulative deaths per million inhabitants is significantly higher in EU ‘debtor’ countries (e.g., Hungary: 3125, Bulgaria: 2872, Italy: 2155, Poland: 1996, Spain: 1832, Portugal: 1758) than in better-off ‘creditor’ countries (e.g., Austria: 1200, Germany: 1106, the Netherlands: 1075, Denmark: 450, Finland: 189). Figures as of 15 September 2021 from database Our World in Data, https://ourworldindata.org/coronavirus/

2 For instance, Germany accounted for half of the emergency state aid approved by the European Commission in the first months of the pandemic (Euractiv, May 4, 2020).

Appendix: Interviews conducted

1. Official, French Elysée, 19 November 2020

2. Official, German Finance Ministry 1, 23 November 2020

3. Official, European Commission 1, 23 November 2020

4. Representative of German Industry, 26 November 2020

5. Official, German Finance Ministry 2, 26 November 2020

6. German MP/ Finance Ministry, 4 December 2020

7. Official, European Commission 2, 4 December 2020

8. Official, French Finance Ministry, 9 December 2020

9. Official, European Commission 3, 8 January 2021

10. Official, German Federal Government, 8 January 2021

11. German MP 1, 14 January 2021

12. German MP 2, 15 January 2021

13. Official, German Finance Ministry 3, 15 January 2021