Nein to ‘Transfer Union’: the German brake on the construction of a European Union fiscal capacity

ABSTRACT This paper argues that, on the development of European Union (EU)/Eurozone fiscal capacity German governments have consistently engaged in foot-dragging. Few German state elites have ever supported European fiscal capacity building beyond the EC budget. Following the outbreak of the Eurozone sovereign debt crisis, German governments agreed to the creation of financial support mechanisms only with reluctance. We see a case of continuity in German policy preferences driven by consistent ordoliberal and, specifically, moral hazard concerns. The important long-term change that we can observe relates to the growing importance for successive governments of avoiding audience costs driven by sceptical public opinion and the rise of a challenger party in German politics. German government support for a massive EU fiscal response to the economic crisis triggered by the COVID-19 pandemic demonstrates an exceptional policy position in favour of temporary financial mechanisms involving no fiscal transfers among member state governments.


Introduction
This special issue sheds light on two key arguments about Germany and European integration. The first is that Germany, once a major driver of integration, transformed into a more 'normal' member state 'which reacts to patterns of interdependence by comparing the costs of maintaining the status quo with the price for change' (Freudlsperger and Jachtenfuchs, this issue). Since the agreement on the Treaty on European Union (TEU, Maastricht Treaty), German governments adopted a much more instrumental position on the transfer of core state powers to the European level Paterson 2010, 2019;Cole 2000). The main reasons advanced by academic and other observers to explain this shift are the sensitivity of German governments to the immediate material costs and to the audience costs of pursuing core state power integration. Thus, the second argument: as both types of costs are typically lower -at least for governments and taxpayers -in supranational regulatory policy-making than in European-level capacity building, German governments should display a strong preference in favour of supranational regulation of core state powers. Germany departs from this stance only when faced with imminent and credible threats of sectoral or overall disintegration. Only in cases of very high and escalating status quo costs might German governments prove willing to contribute to European-level capacity building. Overall, however, the editors of this special issue expect Germany to have a strong preference for the toolkit of the 'regulatory state' (Majone 1996).
In this contribution, we analyse German preferences on the integration of core state powers in the field of fiscal policy, arguably a key test case for the propositions put forward in this special issue. Specifically, we focus upon supranational fiscal capacity building. As capacity building over the past two decades was mainly about fiscal instruments for the Eurozone and less so on the overall fiscal capacity of the EU -its yearly budget and its Multiannual Financial Framework (MFF) -we focus principally upon the former. We also analyse German preferences on proposals for increasing the EU's/ Eurozone's fiscal capacity through common debt issuance with joint liability. In order to assess the claim that the Maastricht Treaty constituted a watershed in terms of German preferences, we also examine German preferences revealed in earlier debates on EU fiscal capacity-building prior to the 1990s.
This contribution proceeds as follows. The following section discusses the key causal arguments based on domestic economic and political cost-benefit calculations as a basis for national preference formation. We also consider the ideational sources of German preference formation -in particular ordoliberal thinking -in order to situate the domestic cost arguments in a broader debate on preference formation. The third section deals with the development of German policy on EU/Eurozone fiscal capacity building during the sovereign debt crisis (SDC), followed by a fourth section focused on the shift in German preferences during the COVID-19 pandemic. The final section assesses the empirical record against the background of the guiding hypotheses and concludes.
Our research design combines a long-term overview of German policy spanning fifty years with an analysis of case studies of agreed or proposed financial support mechanisms since 2010. The former allows us to assess the hypothesis of the Maastricht Treaty as a watershed in Germany's European policy. The short case studies on bailout funds, the debate on joint debt issuance, the Eurozone budget and the fiscal response to the pandemic shed light on the explanatory factors accounting for the varying degrees of German support for fiscal capacity building. Methodologically, we rely on a triangulation of secondary academic and primary official documentation, quality newspaper sources and descriptive statistics, notably opinion poll data, parliamentary vote tallies and the EMU Choices dataset. 1

German preference formation: domestic cost-benefit calculations
Following the theoretical framework outlined in this special issue's introduction, we take domestic cost-benefit calculations as our starting point. Costs mainly come in two forms, material costs -immediate or longer term -and political costs such as domestic audience costs. In fiscal policy, material costs of EU level capacity building can take many forms: direct and indirect, short-and long-term. Among the direct and short-term costs are national contributions to (new) European fiscal capacities, immediate costs of lending money to other member states and higher borrowing costs for current account surplus and creditor countries in case of joint liability for public debt. The higher these (anticipated) costs and the share of them a member state has to contribute, the lower -ceteris paribus -their preference for supranational fiscal capacity building. In line with hypothesis 1.1. in this special issue's introduction, we expect Germany to advocate EU-level regulation of national fiscal policies in order to deal with the externalities of interdependence (Freudlsperger and Jachtenfuchs, this issue). However, we expect the preference for fiscal capacity building to co-vary with status quo costs. In line with hypothesis 1.2, we predict that '[s]tate elites support the build-up of supranational capacities solely in cases in which prohibitive costs arise from the regulatory status quo of a given field' (Freudlsperger and Jachtenfuchs, this issue). We expect this support for fiscal capacity building to last only as long as state elites perceive the costs of fiscal capacity building to be lower than the costs of non-action, especially when dangers of sovereign defaults and bankruptcies of systemically important financial institutions imply potentially high shortterm costs. German governments must also strike a balance between the short-term costs of fiscal capacity building and the longer-term benefits of a stable Eurozone framework serving the export interests of the economy.
Despite the possibility of diverging perceptions of costs and benefits between state elites and high-ranking government politicians concerning the development of EU/ Eurozone level financial support mechanisms, we have detected no public disagreements nor indications of major non-public differences between these officials on this topic. This broad alignment differs notably from the major publicly-expressed disagreements between the Bundesbank and government politicians on other aspects of European economic governance and notably on the European Central Bank's (ECB's) nonconventional monetary policy and -previously, during the 1970s and 1980s -on the development of currency support mechanisms.
We also expect anticipated longer-term risks of fiscal policy instruments and choices to influence German government preference formation. These anticipated risks stem principally from the perceived moral hazard created and include: risks associated with public grants and guarantees for loans handed out by the Eurozone's rescue funds; risks of banking crises with cross-border contagion effects due to wrong incentives, notably linked to the availability of a European financial backstop for bank resolution; and the moral hazard effects of separating risks, liability for them and control over risky decisions between actors and levels of decision-making.
Another longer-term consideration likely to influence German state elites' preference formation is the perception of the distributional implications of EU-level fiscal capacity building. We expect the support for fiscal capacity building to vary with the perception of the longer-term distributional implications of fiscal policy choices. In case they are not considered to be fiscally neutral between the member states over the medium and longer term, support for them varies with the perception of being on the 'winning' or 'losing' side in terms of the instrument's distributional impact.
Domestic audience costs influence preference formation in three different ways: through the channel of public opinion and economic voting, through the parliamentary arena and through inter-and intra-party party competition (Târlea et al. 2019). In addition to material costs, we predict audience costs to be an important explanatory variable. We expect the distribution of attitudes in and political demands of mass publics to influence the preference formation of governments in three ways: through the effects of economic voting based on voters' cost-benefit calculations anticipated by governments; through the channel of parliamentary representation, the effects being stronger in political systems with powerful parliaments; and through the channels of inter-and intraparty competition. The willingness of the government to subscribe to deeper integration by way of fiscal capacity building should vary with the level of competition the parties in government face from Eurosceptic fringe parties. Furthermore, the greater the level of intra-party division in governing parties on fiscal integration issues, the lower the government's preference for fiscal integration should be. In case division increases over time inside governing parties, the transaction costs of holding the party and ultimately the coalition together also increase. Hence, parties in government seeking to stay in office have a strong incentive to refrain from supporting highly contested forms of fiscal policy integration.
Fiscal capacity building is more visible and contested in Germany -as the largest net contributor to the EU budget -than steps towards more intrusive fiscal regulation due to potential distributional consequences. We thus expect capacity building to be opposed by mass publics, in particular when politicized by challenger parties, in line with the introduction's expectation 2.1: 'Mainstream parties oppose the integration of core state powers if challenger parties succeed in politicising its material or sovereignty costs among the wider public' (Freudlsperger and Jachtenfuchs, this issue).
In addition to material and audience costs, other sources of domestic preference formation have been discussed in the literature. Powerful sectoral interests and business associations, of key importance for domestic preference formation according to liberal intergovernmentalism (Moravcsik 1998), have been shown as being of little importance for the German government's positions at key moments of the euro crisis . Greater importance has been assigned to ideational sources of German preference formation both on the road towards EMU and in particular during the Eurozone crisis years. In order to avoid any omitted variable bias, we also examine these ideational sources of preference formation and relate them to the analytical framework put forward in this special issue. Risse et al. (1999) point to Germany's deeply ingrained European identity in order to explain German governments' willingness to embark on the journey towards a currency union even though its costs and benefits were not reliably calculable ex ante. Assuming the stability of this identity variable over time, we expect a strong German preference for preserving a major step of European integration into which German governments invested considerable domestic political capital in the past. However, German preference formation included careful cost-benefit calculations with regard to the best way to construct EMU -a particular mix of regulatory and capacity-building measures. Germany's European identity offered little guidance in this regard.
A number of scholars identify ordoliberal beliefs and a sound money paradigm as important ideational sources of German preference formation both on the road towards EMU (Dyson and Featherstone 1999, 20;McNamara 1999) and during the Eurozone SDC (Brunnermeier, H., and Landau 2016;Matthijs 2016;Nedergaard and Snaith 2015;Schäfer 2016). The impact of ordoliberal ideas on German preference formation should translate into a clear preference in favour of a rules-based system as ordoliberals think in terms of economic constitutions and not in terms of discretionary economic policies -Ordnungspolitik instead of Prozesspolitik. However, with regard to concrete policy measures and instruments, ordoliberal thinking often provides insufficient guidance. Jacoby (2014) observes that German ordoliberals could be found on different sides of major policy debates during the SDC years. He concluded that the impact of ordoliberal thinking on German policy choices tended to be overestimated. And Lars Feld, a prominent ordoliberal and holder of the Walter Eucken chair in Freiburg, stressed the strong dose of pragmatism informing Germany's policy choices to save the Eurozone during the SDC, thus limiting the influence of ordoliberal ideas (Feld, Köhler, and Nientiedt 2015).
However, we argue that the influence of ordoliberal thinking about German preferences on European fiscal capacity should have made itself felt in three ways. First, it should translate into a strong preference for a rules-based system and hence an emphasis on regulating national fiscal policies (Schoeller and Karlsson, this issue). Second, policymakers influenced by ordoliberal ideas should stress moral hazard concerns. Third, ordoliberal ideas should shape material cost-benefit calculations, and most importantly their time horizon. We expect policymakers adhering to an ordoliberal economic paradigm to put more weight on longer-term cost-benefit calculations compared to policymakers that do not adhere to this German tradition. 2

Maastricht as a watershed?
In the 1970s, the European Communities took decisions on balance of payments assistance instruments and experienced debates on the pooling of currency reserves. Although these were not about fiscal policy, they had many similarities as they dealt with Community-level capacity building with potentially strong redistributive implications between current account surplus and deficit countries. In the final phase of the Bretton Woods monetary system, the German government accepted the creation of balance of payments support instruments, including the Medium-Term Financial Assistance Facility. Created in 1971, this facility provided a framework to pool preallocated bilateral loans between member states, but was only once used for Italy in 1974 (Stieber 2015). In response to the 1973/74 oil price shock, the German government supported the establishment of a Community Loan Mechanism in 1975. This allowed for the issuing of European Community bonds (up to US$3bn.) on private capital markets (Horn, Meyer, and Trebesch 2020). Throughout the 1970s, most discussions on Europeanlevel fiscal capacity building focused upon funds to support stability in the parities between currencies participating in the Snake monetary mechanism. A European monetary cooperation fund was created in 1972 but this involved no paid-in capital. Towards the end of the decade, France pushed for the creation of a European Monetary Fund (EMF) which would have required paid-in capital from the member states. However, the German government block the fund, which met with strong resistance from state elites, most notably from the Bundesbank (James 2012, 188-89).
The 1970s also saw some debate on the need for a larger European budget underpinning a move towards monetary union (see, for example, Marjolin 1975, 5). The MacDougall Report (1977) suggested a maximum option of a European budget performing stabilisation and redistributive functions of 7.5 to 10% of the Community's GDP. Germany consented to substantial increases in European fiscal capacity when the Community doubled its structural funds in the Delors I multiannual financial framework (MFF) (1988-93) and added a new Cohesion Fund in the context of the Maastricht Treaty negotiations. However, these were above all cases of side payments to buy off the resistance of Southern member states, in particular of Spain, on the Single Market programme and Monetary Union (Moravcsik 1998). During the negotiations leading to the TEU, the prevailing 'sound money paradigm' and monetarist ideas left little space for debates on EU level fiscal capacity building (Schlosser 2019, 27-31). The Commission proposed a financial assistance mechanism serving the purpose of stabilisation, not redistribution, but met with opposition from Germany, the Netherlands and the UK (Dyson and Featherstone 1999, 732). Hence, when the Eurozone crisis hit, it 'exposed brutally the "fiscal void" of the monetary union' (Schelkle 2014, 105) because the redistributive fiscal capacities of the EU budget, in particular the structural and cohesion funds, were not well suited for dealing with asymmetric shocks.
During the 1990s, German government willingness to finance consensus-building in negotiations on the EU's budget decreased substantially. Germany's increased net contribution became more contested domestically due to post-reunification federal budgetary tensions (Anderson 1999;Harnisch and Schieder 2006;Becker 2014, 258-92). In MFF negotiations, Germany aligned with other net contributors to push for a relative decrease of the MFF ceiling as a share of EU GDP in the 2000s and 2010s compared to the 1990s. Overall then, we observe strong elements of continuity on fiscal capacity building before and after Maastricht with a hardening of the German stance on the EU's MFF in the aftermath of reunification. Advocating budgetary discipline at the EU level was a core element of German European policy for decades (Becker 2014, 241-45).

Financial assistance in the eurozone crisis: bailouts and rescue funds
During the repeated debates on Eurozone fiscal capacity building from the start of the SDC, Germany consistently displayed a highly restrictive approach. This is true for the first Greek bailout in 2010, the establishment both of temporary and permanent stabilisation facilities ('rescue funds'), on the issue of a financial backstop for the Banking Union's Single Resolution Fund (SRF), and in the debate on a separate fiscal capacity for the Eurozone in 2011/12 and from 2017 onwards. The same holds true for the debates on mutualizing fiscal risks through the introduction of Eurobonds in 2010-12. The timing of events during the Greek SDC in spring 2010 indicates the primacy of domestic political concerns directing German policy. The Merkel government delayed decision-making on the coordinated bilateral Greek loan facility because of an important regional state election in North-Rhine Westphalia on 9 May 2010, fearing heavy losses for the Christian Democrats, which increased the material costs of the Greek bailout (Jones 2010).
During the subsequent negotiations on rescue funds, this 'too little, too late' pattern consistently characterized German policymaking (Meiers 2015;Webber 2019, 56-105). According to the EMU Choices dataset, 3 German governments almost systematically adopted the most restrictive position of all Eurozone national governments in the policy space on contested fiscal capacity building issues. Initially, the Merkel government was not supportive of the loan programme for Greece. In the debate on debt relief for Greece, it opted for a private sector only solution, without public contribution. Regarding the European Financial Stability Facility (EFSF), which the Merkel government wanted to be only temporary, Germany initially opposed the idea of providing this fund with public loan guarantees (Gocaj and Meunier 2013). In 2011, the German government opposed both the proposal to increase the EFSF's lending capacity and proposals to add instruments to its toolbox. In policy debates and negotiations on the establishment of the permanent European Stability Mechanism (ESM) in 2011, the Merkel government was successful in its demands that the mechanism's lending capacity be limited to €500bn., with the maintenance of strict conditionality for all ESM lending.
On the governance of the new fiscal instruments, Germany was in favour of intergovernmental institutions. This holds true both for the EFSF, a special purpose vehicle under private law, and for the ESM, based on an intergovernmental treaty outside the EU legal framework as the UK prevented its adoption under Union law. 4 This preference is fully in line with the expectation that German state elites, when consenting to EU-level capacity building in core state powers, opt for forms of governance providing them with control and veto power over how these capacities are used (Genschel and Jachtenfuchs 2014;2016, 51;Freudlsperger and Jachtenfuchs, this issue). Key decisions in the ESM's Board of Governors require unanimity (Art 5, para 6 ESM Treaty). The intergovernmental nature of the ESM 'implies that any participating country guarantees only its share of the bond issue which is determined by its share in paid-up ECB capital. There is only "several", no "joint and several" guarantee', thus restricting the ESM's lending capacity (Schelkle 2014, 110). Despite strong domestic political opposition, functional pressure drove German government consent to establish rescue funds. Prohibitive status quo costs when the Eurozone's survival was at stake -notably with the rise in euro periphery member state sovereign debt yields -made the German government change its position on rescue funds, resisting moral hazard concerns and disregarding high domestic audience costs (Schimmelfennig 2015). 5 Risk premia at unsustainable levels, steeply rising contagion risks and market tensions provided the tipping points that made the government change its stance. Pragmatism trumped ordoliberalism (Feld et al. 2015). This observation corroborates the hypothesis stated in the introduction that German governments are willing to consent to supranational capacity building in cases of 'full blown regime crises' but 'solely in cases in which prohibitive costs arise from the regulatory status quo of a given field' (Freudlsperger and Jachtenfuchs, this issue).
Another example of the close link between German support of fiscal capacity building at the EU level and increasing status quo costs and functional pressures to act is provided by the debate and decision-making on a financial backstop for Banking Union's Single Resolution Fund (SRF) in the event the SRF's resources are depleted. When the Eurozone crisis reached its apex in June 2012 in the context of the Spanish banking crisis, Germany accepted indirect bank recapitalisation via the EFSF. Germany subsequently accepted direct bank recapitalisation through an ESM instrument limited to €60bn. that was adopted in December 2014. In 2013, the German government supported a Council decision to establish a financial backstop to the SRF as an ultimate instrument for bank recapitalisation (Council 2013). Incorporated into the ESM Treaty reform agreed upon in the Eurogroup in November 2020, this ESM backstop is to replace the mechanism's direct recapitalisation instrument by the beginning 2022. The German government insisted on the backstop's fiscal neutrality -that is, avoiding transfers among member states -in the medium-term. After the intensity of the crisis diminished from mid-2012 onwards, the German government adopted delaying tactics with regard to this fiscal instrument and resisted an earlier start for the financial backstop for years (Schild 2018). Berlin successfully pushed for the prior reduction of risks on bank balance sheets, finally considered to be sufficient in November 2020.
This highly conditional support for EU-level fiscal capacity building during the SDC might be explained both by moral hazard concerns inspired by ordoliberal ideas and by the immediate and longer-term material costs of the crisis. Despite the existential crisis facing the euro, the German government preferred to leave the bulk of the burden of crisis management to the ECB which took decisive steps and performed a key role to rescue the euro in 2012, following President Mario Draghi's 'whatever it takes' speech of July and the subsequent design of the Outright Monetary Transactions (OMT) programme. Both Chancellor Merkel and Minister of Finance Schäuble lent their support for the ECB's 'fiscal integration by stealth and default' (Schelkle 2014, 116). The German government proved willing to sacrifice ordoliberal principles in that this ECB policy clearly entailed the risk of politicising the central bank's role by blurring the lines between monetary and fiscal policy. The functional alternative of huge fiscal capacity building or risk sharing via Eurobonds would without doubt have been much more visible, contentious and hence politicized at the domestic level, entailing high domestic audience costs. We interpret this choice as reflecting a higher value attached by the German government to the goal of diminishing domestic audience costs compared to the goal of strictly adhering to ordoliberal principles.

The debate on a Eurozone budget
The creation of a Eurozone specific budget was placed on the European agenda in 2012, promoted by the Commission and taken up in the Five Presidents' Report on the completion of EMU (Juncker et al. 2015). The governments of France and Southern European countries sought the creation of a budget to serve stabilisation purposes by providing transfer payments in crisis situations. The Merkel government expressed cautious support but promoted 'contractual arrangements' between the European level and a member state, combining financial incentives with a commitment by the recipient to undertake structural reforms. Given pronounced intergovernmental divisions, the idea of a Eurozone budget and the proposal on 'contractual arrangements' dropped off the agenda.
The creation of a Eurozone specific budget returned to the agenda in 2017, when French president Emmanuel Macron made the case for a Eurozone budget to perform allocative, redistributive and macroeconomic stabilisation functions (Macron 2017). The German Grand Coalition responded by promising to 'support the specific budgetary resources for economic stabilization and social convergence and for the support of structural reforms . . . which can be the point of departure for a future investmentrelated budget for the [Eurozone]' (Coalition Agreement 2018, 12-13). In June 2018, the 'French German roadmap for the euro area' specified the goal of setting up a 'Eurozone budget within the framework of the European Union to promote competitiveness, convergence and stabilization . . . starting in 2021ʹ (French and German Government 2018). The budget figure proposed by Chancellor Merkel (Merkel 2018) was, however, considerably lower -in the lower two-digit billion euro range -than President Macron's proposal of several hundred billion euros (Macron 2017).
The German government preferred the negotiations on a Eurozone budget under the label 'budgetary instrument for convergence and competitiveness' (BICC) to take place in the context of negotiations on the EU's MFF (2021-27), thus giving all 27 EU member states a veto. Any proposal going beyond a small budgetary instrument stood no chance to be acceptable to the German coalition government. The Eurogroup agreed to provide the BICC only €17bn. over the 2021-27 period. When the Commission tabled its proposals on a EU Recovery Instrument and a revised MFF proposal on 27 May 2020, it dropped the Eurozone budget proposal that had, in any event, been watered down to 'homeopathic insignificance' (The Economist, 23 May 2020). The German government's cautious support for Macron's Eurozone budget idea can be interpreted as a case of symbolic capacity building to accommodate French preferences without incurring either important material or domestic political costs.

German opposition to common European debt issuance
Apart from the temporary EFSM (see note 2), German governments never accepted common European debt issuance to stabilize the Eurozone -prior to the COVID-19 pandemic. In June 2012, Chancellor Merkel came out unequivocally against so-called 'Eurobonds', declaring that there would be no full debt sharing 'as long as I live' (Spiegel Online,27 June 2012). Merkel's uncompromising stance aligned fully with an extremely hostile public opinion towards Eurobonds, in particular among supporters of the centreright CDU/CSU. At the voter level, there was broad cross-party consensus on this issue (see Table 1). Another part of the explanation for this German government position is to be found in the growing discontent regarding the bailouts and rescue funds in the ranks of the coalition parties during the SDC, leading to increasing difficulties to secure parliamentary majorities in favour of financial assistance capacities without the support from opposition parties (see Table 2).
The CDU and CSU's long-standing policies (Freudlsperger and Weinrich 2020) and domestic audience opposition largely explain the limited willingness to lend support to European level fiscal capacity building potentially leading to fiscal redistribution. At the height of the SDC, public opinion was overwhelmingly hostile to risk mutualisation proposals, particularly Eurobonds (see Table 1). This translated into ever fiercer interand intra-party competition on the limits of German solidarity and risk sharing in the Eurozone (Schieder 2014). 6 A growing number of dissenting voices from the ranks of the governing Christian Democratic (CDU/CSU) and liberal (FDP) coalition (2009)(2010)(2011)(2012)(2013) restricted the government's room to manoeuvre as the governing coalition no longer enjoyed a secure majority in parliament to ratify European level commitments on financial assistance and loan guarantees. The FDP, fighting for the party's survival, hardened its stance on solidarity issues and the Bavarian CSU played with soft Eurosceptic discourses, both categorically rejecting Eurobonds and any other move towards a 'transfer union' (Wimmel 2012). The growing internal resistance within the governing coalition against risk mutualisation can be seen in the increasing number of dissident votes in the Bundestag's decisions on rescue schemes (see Table 2). Any proposal that could be interpreted domestically as a step towards some form of 'transfer union' stood little chance of attracting either sufficient public support, or the backing of parties in the governing coalition. 7 Several explanatory factors account for the German government's hostility towards common debt issuance as with a substantial Eurozone budget. Increased inter-party competition due to the rise of the right populist, Eurosceptic Alternative für Deutschland -which won 92 seats in the 2017 German Bundestag elections -increased the pressure on the CDU/CSU to adopt policies to limit the drift of voters to the AfD. Moreover, the party in government most open to European fiscal capacity building -the Social Democratic Party -experienced strong decline in voter support from 2017 and its worst crisis for decades. Its influence within the Grand Coalition diminished. SPD minister for economic affairs, Sigmar Gabriel, came out in favour of a Eurozone budget (Gabriel and Macron 2015), while finance minister, Olaf Scholz, also of the SPD, later unsuccessfully advocated a European unemployment reinsurance fund (Handelsblatt, 25 November 2018). Their ideas, however, never reflected the agreed German government position, as the CDU/CSU continuously adopted a more restrictive position on risk sharing and transfers in the Eurozone. The CDU/CSU did not change, but rather hardened its stance of opposing fiscal capacity building in the 2010s as a consequence of the rise of the challenger party AfD. This lends some support to hypothesis 2.1 (Freudlsperger and Jachtenfuchs, this issue): 'Mainstream parties oppose the integration of core state powers if challenger parties succeed in politicising its material or sovereignty costs among the wider public'.

The Covid-19 crisis: releasing the brake
A major German policy shift, which took most observers by surprise, came with the COVID-19 pandemic. In the early moments of the crisis, the German government's positioning was fully in line with its previous cautious positioning. When the Italian government called for the temporary introduction of common EU debt (so-called 'coronabonds') and found support from a coalition of nine EU member states, led by France (Euractiv 2020), this proposal 'met with a familiar "Nein"' from the German government (Spiegel International, 27 March 2020). Chancellor Merkel as well as the ministers of economics, Peter Altmaier (CDU), and of finance, Olaf Scholz (SPD), ruled out 'coronabonds', pointing instead to the unused capacities of the ESM as the preferred instrument for lending to those EU member states most affected by the pandemic (Financial Times, 27 March 2020). Following this deep split, which found France and Germany openly siding with opposing coalitions, Franco-German bilateralism started working again. In May 2020, France and Germany set the European negotiation agenda by agreeing a joint proposal on a large-scale post-COVID-19 recovery fund to disburse grants of €500bn., financed by joint debt instruments (Bundesregierung 2020). This was taken up by the European Commission in its own proposal on an EU recovery plan (Next Generation EU, NGEU) and its key component, the Recovery and Resilience Facility, and found its way into the final agreement of July 2020. The latter foresees €750 bn. of NGEU expenditures outside the MFF of which €390bn. are to be disbursed as grants (European Council 2020).
In defence of the Franco-German proposal and the NGEU, Chancellor Merkel emphasised the extraordinary, historical nature of the economic and social challenges faced by the EU and its member states. She repeatedly pointed to the real danger of European single market disintegration and the unravelling of the wider integration framework (Merkel 2020). Hence, in the context of an economic crisis of unprecedented magnitude, the German government broke two taboos. First, it advocated grants in addition to loans in favour of member states most hit by the economic impact of the crisis. Second, it proposed that these grants should be financed by allowing the Commission to borrow massively on financial markets on behalf of the EU -implying joint liability for debts, which Germany had strictly ruled out during the Eurozone debt crisis. Merkel stressed that the recovery plan was to be 'pandemic-focused and temporary' (Merkel 2020, our translation) and welcomed the Commission's proposal to integrate this extraordinary instrument and its temporary nature into a decision on EU Own Resources to be ratified by all member states. By so doing, the Chancellor and her government sought to avoid domestic debate on a permanent 'transfer union'. Moreover, the government insisted on debt issuance to take the form of common European debt and borrowing by the Commission to avoid a mutualisation of debt contracted by the member states. 'We always said that we would not accept the idea of the German budget guaranteeing Italian expenditures', according to a high-level diplomat quoted in Le Monde (18 July 2020, our translation). Unlike the adopted mechanisms to tackle the SDC, German public opinion offered greater support to these recovery instruments. A majority of 51% favoured the Franco-German recovery fund proposal (34% against) (Spiegel 2020). This ties in to hypothesis 2.2: 'Mainstream parties support supranational capacity-building only if citizens regard the material and sovereignty costs involved as justifiable or even desirable' (Freudlsperger and Jachtenfuchs, this issue). However, 56% wanted financial assistance to be restricted to lending, whereas 36% supported grants in addition to lending (Politbarometer 2020).
It is noteworthy that the German government did not wait until functional pressures and escalating status quo costs -rapidly deteriorating economic situations in a number of member states, increasing risk premia on sovereign debt and the looming danger of banking crises -left little choice other than to take bold measures. It acted relatively early in the pandemic in order to mitigate its economic impact.
Prior to and during the European Council's negotiation marathon on the NGEU and the MFF 2021-27 from 17-21 July 2020, Chancellor Merkel joined with French president Macron to undermine the attempts of the 'frugal four' (Austria, Denmark, the Netherlands and Sweden), joined by Finland, to cut significantly the amount of grants. The member states compromised on €390bn. in grants -significantly less than the proposed €500bn. This was nonetheless a major step in fiscal capacity building with Germany in the unfamiliar role as pace-setter. Moreover, the German government, contrary to the expectation presented in the introduction (Freudlsperger and Jachtenfuchs, this issue), departed from its longstanding position in favour of strict intergovernmental control of the disbursement of funds and conditionality. Instead, it agreed to empower the Commission with a key role to manage the disbursement of grants and loans with only limited control by the member states in the framework of the European Semester.
Can the Franco-German proposal and support for NGEU then be considered to reflect a major preference shift and to be a decisive step for the German government towards accepting the idea of a transfer union? We answer in the negative. The European Council's conclusions reflect longstanding German preferences on a key point: 'Given that NGEU is an exceptional response to those temporary but extreme circumstances, the powers granted to the Commission to borrow are clearly limited in size, duration and scope' (European Council, 2020). At the same time, the repayment of the €750bn. debt to be borrowed by the Commission on capital markets is not temporary and may last until 2058. In order to increase the EU's repayment capacity, the European Council's conclusions foresee the option of eventually establishing new EU own resources, either levies (a carbon border adjustment mechanism and/or a digital levy) or taxes (Financial Transaction Tax). With its consent to both EU-issued debt (albeit temporary) and its openness to consider new types of own resources, the German government appeared prepared to take two important steps towards Fiscal Union. This major change can only be explained by the extraordinary nature of the socio-economic and political challenges created by the COVID-19 pandemic.
However, it would be problematic to present this either as a step to 'transfer union' or German acceptance of such a move. The German government conceives of its own and the EU's response as being a temporary fix to deal with rapidly increasing functional pressures to act, in the context of the most severe recession and economic crisis since the Second World War and the most serious risk of European disintegration tendencies, in line with hypothesis 1.2. (Freudlsperger and Jachtenfuchs, this issue). The road taken implies a fair sharing of fiscal costs and risks at the Union level as common debt issuance does not mean joint and several liability. Germany's share of the risk is limited to its share of contributions to the EU budget. The integration of the grants into the EU budget limits the domestic political costs because direct transfers between national budgets are avoided. Hence, this policy change is not a far-reaching change in the underlying, long-standing German preferences on EU fiscal capacity building (Freudlsperger and Jachtenfuchs, this issue).

Conclusion
Our analysis of German government preferences on the integration of a core state power in fiscal policy strongly supports key hypotheses developed in this special issue's introduction. German government support for EU fiscal capacity building was much weaker than support for the development of EU fiscal policy regulation (Schoeller and Karlsson, this issue). From a comparative perspective, Germany found itself almost consistently at the extreme end of the policy space on contentious issues, either opposing new fiscal capacities altogether (in particular Eurobonds) or trying to set clear limits to their size and range of functions until 2020. In line with hypothesis 1.2, Germany's willingness to support fiscal capacity building crucially depended on the prohibitive costs of the regulatory status quo in acute crisis situations. Until the history-making decisions on the COVID-19 pandemic recovery fund, German governments insisted on intergovernmental forms of integration ensuring member state control and veto power when agreeing to new fiscal instruments such as rescue funds. Facing situations of high regulatory status quo costs and potentially high audience costs for fiscal capacity building, German governments, in line with the introduction's expectation, choose the option of fiscal capacity building only as 'temporary or emergency fixes for a Union on the brink of collapse' (Freudlsperger and Jachtenfuchs, this issue). Furthermore, German governments displayed a strong and unwavering resistance to any form of permanent redistributive EU-level fiscal capacity building. Only the German pace-setting role in 2020 on the EU recovery fund and huge grants financed by joint European debt issuance diverged from this pattern. This pace-setting role removed Germany from the extreme end of the policy space and put it closer to a pivotal middle position among the EU member states. But for most of the time, German preferences on fiscal capacity building in the EU and the Eurozone were almost a mirror image of the French ones. Paris advocated a 'fiscal union' approach, emphasizing solidarity values, calling for risk sharing instruments, the joint issuance of debt, a huge Eurozone budget and taxation powers for the EU (Schild 2020;Târlea et al. 2019).
How do we explain this empirical record? The timing of decisions in key moments of the sovereign debt crisis lends strong support to cost arguments, both domestic costs -immediate material costs and (potential) audience costs -and status quo costs. Over time, domestic cost-benefit calculations grew in importance due to processes of politicisation of EMU topics and the rise of the right-wing populist AfD. During the worst years of the SDC, a public opinion hostile to risk sharing and transfers and inter-party as well as intra-party competition led to the hardening of government positions on instruments with clear redistributive implications, inter alia Eurobonds. German public opinion was supportive of stronger integration on fiscal regulation (Schoeller and Karlsson, this issue) but opposed fiscal capacity building, which contributed to the rise of a challenger party, the AfD, the raison d'être of which was to fight the government's EU-level 'rescue policy' -in line with hypothesis 2.1. This process of politicisation fuelled the discontent of mass publics and fed intra-party conflicts in Christian Democrat and FDP ranks. Acting in line with mass publics based on largely shared preferences, German governments and state elites disregarded domestic opposition to fiscal capacity building only at particular moments, when the survival of a key policy regime, in our case EMU, was at stake. As soon as the imminent danger of the policy regime collapsing appeared to have passed, government support for fiscal capacity building largely evaporated and it once again tended to oppose fiscal capacity building.
What about ideational factors? Ordoliberal ideas, in particular moral hazard arguments, clearly pervaded German government and state elite discourses. These arguments did not merely provide a smokescreen to hide material interests. They influenced the way in which state elites perceived and weighted short-term and medium-to long-term material costs, tilting the balance more to the latter. A key ordoliberal principle guided the government's approach throughout the SDC: namely, that there cannot be joint liability (for sovereign debt or the cost of bank resolutions) without clear supranational control. This ideational component of preference formation sets Germany apart from other member states with 'hawkish' fiscal policy preferences, such as the Netherlands or other net contributors to the EU's budget. However, an ideational explanation based on ordoliberal ideas must be supplemented by other explanations to account for the development of German government preferences. Escalating status quo costs prevailed in critical moments. To a limited extent, pragmatism trumped ordoliberalism when the Eurozone faced the abyss.
The editors of this special issue argue that after Maastricht, Germany moved from its position as a keen driver of European integration to a more 'normal' member state (paper 1). We argue that, on fiscal capacity building, German governments were never a keen driver of integration. In this respect, we see a case of continuity rather than change. An important change that we observe relates to the growing attention given by successive governments to sceptical mass publics and to the rise of a challenger party. The COVID-19 pandemic changed this -at least temporarily. An economic shock of an unprecedented nature and beyond national responsibility triggered a major policy shift. The German government, after initial hesitation, acted as a pace-setter on fiscal capacity building for the first time -rather than a foot-dragger -against the background of a more permissive public opinion and a right-wing populist party experiencing a decline in voting intentions. However, we do not interpret this preference change as signalling a durable shift towards German government support for a 'transfer union' which still meets with strong scepticism across the German political spectrum and notably in the ranks of the CDU/CSU, the FDP and the AfD. The perceived status quo costs -the potentially very farreaching and damaging consequences of failing to act decisively -trumped both ordoliberal ideas and short-term electoral cost-benefit calculations and fears of economic voting against the coalition government. As with the creation of the EFSF and the ESM a decade earlier, the variation in perceived status quo cost was the key explanatory factor to account for the level of the German governments' support for supranational fiscal capacity building. When push comes to shove, the unwavering German commitment to the European integration framework and the short-and long-term material benefits of preserving the single market and monetary union encouraged the government to accept extraordinary fiscal capacity building of a temporary nature. However, the German government maintained its 'nein' to a permanent 'transfer union'.