Models of Solidarity in the EMU. The Impact of COVID-19 After Weiss

Right in the middle of the Covid-19 pandemic, the German federal constitutional court (Bundesverfassungsgericht – FCC) issued a ruling that sent massive shockwaves through the continent. Not only did the Court question the legality of the European Central Bank’s bond buying program PSPP (Public Sector Purchase Program), but it also rejected the earlier decision by the CJEU in which this latter had found that program to respect EU law. The ruling is as such not directly concerned with Covid-19 measures, but it may have nonetheless important consequences thereon. In this contribution we will explore what those consequences may be. Apart from the direct effects on the ECB’s pandemic emergency purchase programme (PEPP), we zoom in on the ruling’s indirect consequences on the broader question of how to arrange solidarity in EMU. With regard to the latter, we contend that Weiss and the Covid-19 crises combined will test the basic models of solidarity the EMU relies upon: the models of individual fiscal responsibility, ECB based solidarity and the model of fiscal union. These models are assessed from economic, constitutional and democratic perspectives. © 2021. The Author(s). This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CC-BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. See http://creativecommons.org/licenses/by/4.0/.


Introduction
Right in the middle of the Covid-19 pandemic, the German federal constitutional court (Bundesverfassungsgericht -FCC) issued a ruling that sent massive shockwaves through the continent. Not only did the Court question the legality of the European Central Bank's bond buying program PSPP (Public Sector Purchase Program), but it also rejected the earlier decision by the CJEU in which this latter had found that program to respect EU law. This decision is therefore particularly important for the future of the Economic and Monetary Union (EMU), as it questions the margin of discretion the ECB enjoys and the kind of scrutiny it is subject to. The ruling is as such not directly concerned with Covid-19 measures, but it may nonetheless have important consequences thereon. In this contribution, we will explore what those consequences may be. Apart from the direct effects on the ECB's pandemic emergency purchase programme (PEPP), we zoom in on the ruling's indirect consequences for the broader question of how to arrange solidarity in EMU. With regard to the latter, we contend that Weiss and the Covid-19 crises combined will test the basic models of solidarity the EMU relies upon.
The notion of 'solidarity' we adopt in our analysis is a specific one, 1 and we do not mean to enter the debate on the interpretation of this concept, whether in absolute terms or with regard to its use in Articles 2 and 3 TEU. 2 Our analysis is based on a notion of solidarity that involves the willingness of Member States to limit their sovereignty over their budgetary or monetary policy to take into account the needs and the preferences of the other Member States. Therefore, we adhere to the line of thought that considers solidarity in the EU context as a reciprocal principle, one where the efforts performed by each of the participants in the game find their mutual justification -not in mere altruism, but -in a common objective such as market integration. 3 We believe this approach is particularly suitable to analyse the EMU in the light of the challenges COVID-19 will bring. First, it highlights that, even with limited or no direct fiscal transfers or mutualization of debt, integration is already bringing about, in forms that the public debate too often overlooks, an intense cooperation among Member States of the Eurozone. Second, it better allows to show that the development of more direct forms of transfers and mutualization would complete an existing design, rather than suddenly introducing solidarity out of context.
As one can see, this notion of solidarity does not need fiscal transfers to manifest its reciprocal nature. Constraints to public expenditure do not involve direct transfers among Member States, but they rely on the idea that limitations to budgetary autonomy can deliver a better equilibrium to the benefit of the EMU as a whole, and of each Member State individually. The same applies to monetary policy because, as we shall explain more broadly, 4 relinquishing this part of state sovereignty in favour of an EU institution also entails acceptance of an institutional framework whose immediate policy objective relates to the EMU as a whole, despite persistent diversity at Member States' level.
Against this backdrop, we distinguish three models of solidarity in the EMU. The three models are not alternative and can coexist, but they show to some extent different levels of development. Each model is characterised by a specific form of solidarity, as a consequence of the technique it adopts to foster integration among the participating Member States. The first one is the traditional model, based on individual responsibility and exclusion of risk sharing. This is the approach reflected in the Maastricht treaty and its subsequent amendments, as well as the Fiscal Compact. In this model, the solidarity aspect is embodied by the requirement that Member States contribute to the economic stability of the area as a whole and avoid becoming a burden on other Member States. This means that Member States retain broad margins to determine their economic policies, but these are subject to the constraints needed to ensure coordination. While the first model is centred on economic policies, the second one relies on the common monetary policy (and the instruments that brings along), and therefore has the central bank as its key actor. The solidarity element is based on Member States' participation in the central bank and in their subjection to a centrally developed monetary policy. 5 The third model is that of fiscal union and includes fiscal transfers, based on political decision-making. This model has direct consequences for national fiscal autonomy and financial liabilities.
This contribution will assess these models, their application in the EU, and how they have become tested by the Weiss decision and the subsequent Covid-19 crisis. For each of the three models, the paper provides a short outline (i) and an assessment from the perspectives of their economic consequences and context (ii), of their constitutional embedding (iii) and, finally, of their democratic legitimacy (iv). As the analysis will show, the typology carries an historic dimension, with the Maastricht model being the original model and the model of fiscal union mostly reflecting a possible future scenario (albeit with some elements already visible today). In this sense, subsequent models have demonstrated limitations of previous models. In particular, the economic crisis has highlighted the limitations of the first model of integration and has been followed by a more proactive role of the ECB and its monetary policy. The recent pandemic, combined with the FCC decision on Weiss, are now showing the limitations of the second model and are fostering the development of a third model to combine with the pre-existing ones. However, rather than a straightforward sequential rise and fall of these models within the EMU, we rather propose to see them as models that co-exist and will continue to do so as we argue in the latter part of this contribution.
Each of the models we identify is characterised by a specific interaction between monetary and economic policies. Monetary policy, whether conventional or unconventional, has of course remarkable economic consequences, as both the CJEU and the FCC acknowledge. Reciprocally, economic policies inevitably influence the effects of monetary measures. To put monetary and economic policies in context, we rely in this paper on the same analytical framework of the theory of optimum currency areas (OCAs). 6 As we shall see, this theory analyses the strengths and weaknesses of monetary unions in light of the complex interactions among monetary policy measures, job mobility, wage flexibility as well as budgetary and fiscal policies.
More broadly, we will use the theory of OCAs to contextualise not only the ECB programs, but also the role of solidarity strategies more in general (or the lack thereof). After all, these strategies can also be understood as stabilising mechanisms that counteract excessive imbalances during asymmetric shocks in the single currency area. Just like the theory of OCAs suggests, this kind of analysis is multivariate, because all the factors involved -such as monetary policy, job mobility and flexibility, budgetary policies and fiscal transfers -have an influence on the overall equilibrium of the eurozone. In this respect, each form of solidarity we identify can also be explained as a determinant (or an input) of this equilibrium, and therefore as a lever that policymakers can use to affect it.
To be sure, we refer to OCAs merely as a way to frame our discourse in the light of the interplay between monetary policy, on the one hand, and all the other factors that contribute to determining the economic development of the eurozone and its components, on the other hand. Whether one agrees that the eurozone is an OCA or not, that theory provides a useful framework for analysing the crucial but difficult role of central bank decisions when these apply across national borders. This is all the more so when, as is the case with the eurozone, those borders still reflect material differences among local economic structures and among local budgetary policies. For this reason, the Covid-19 outbreak, is having asymmetric effects on the eurozone, due to the different healthcare and economic systems this comprises, and will be a stress-test for the stability of the area in the years to come. 7

Outline of the model
The Treaty of Maastricht's model of Economic and Monetary Union can be traced back to the Delors report (1989) which in turn has drawn strong inspiration from the much earlier Werner Report (1969). The latter already foresaw that 'transfers of responsibility from the national to the Community plane' would be essential in the economic policy domain. In particular, the idea that monetary and economic policies are strongly intertwined but would be subject to differentiated processes of development can already be distilled from the Werner report. It is equally true, however, that already in the 1960s it was considered impossible to set up a monetary union without dealing at least to some extent with economic policies at the European level as well. 8 The model that emerged with the Treaty of Maastricht -based on a strongly federalized monetary policy, moderate EU influence on national budgetary policies and a rather unclear and undefined legal situation with regard to other economic policies -has thus had ample time to mature. It has been based on the idea that the common benefits brought about by a common currency should not be accompanied by a ' community of risk-sharing'. 9 The Treaties do not define the exact meaning and scope of economic policies. In principle this could encompass all government action intended to affect the economy in some way or other. 10 In the context of EMU, the focus lies on macroeconomic policies, which have been defined as general government policies to steer the economy as a whole, employment and labour market policies and a favourable balance of external trade. 11 As national budgetary policies may impact the stability of the common currency, the Stability and Growth Pact has set ' disciplinary' provisions on public debts and deficits. model is first and foremost solidarity to underpin convergence as a way to reduce the likelihood of asymmetric shocks and, therefore, to support the common currency. The Treaties include strong limitations on financing deficits as well. The prohibition on monetary financing (art. 123 TFEU) bans access to national central banks' and ECB's borrowing mechanisms. 12 This is intended to stimulate Member States to pursue sound public finances policies, which in turn contributes to the stability of the common currency. The socalled 'no-bail out' provision of article 125 TFEU works in a similar vein. 13 Thus, in this original model solidarity is primarily functional to the stability of the euro. This vertical solidarity is complemented by a horizontal solidarity dimension. This is captured by the above-mentioned no-bail out clause, which is not only a warning to individual Member States but also a safeguard for other member states not to have to assume liability for the debts of their peers. They may trust that other Member States equally contribute to the stability of the common currency. The 'rules-based' (as opposed to discretionbased) design of the EMU contributes to this horizontal solidarity as it ensures legal equality of the Member States (all Member States are subject to the same disciplinary provisions). Since the Treaty of Maastricht, EMU law has included a specific solidarity provision (now: article 122 TFEU) which equally demonstrates this solidarity model based on individual Member States' responsibility. 14 This provision creates the possibility for the Council to act, first, in case of 'severe difficulties' in the supply of certain goods (during the corona crisis, the supply of medicines and medical aid could qualify as such) and, second, financial assistance could be provided to Member States in difficulties. This second option is heavily qualified. Only difficulties caused by natural disasters or ' exceptional occurrences' which lie beyond control of the Member State concerned may qualify and the provision makes mention of the ' conditions' that may be attached. Thus, this EMU solidarity provision has been designed as an exception to the default situation in which Member States cannot rely on the EU, central banks or other Member States to address financial difficulties. The EMU legal system has never been a reflection of the original model in its purest form. The 'battle' between rules and discretion has not always been decided in favour of the former. Budget discipline contains discretionary elements as well, such as the determination whether an excessive deficit exists, and whether or not to impose sanctions. 15 Especially the Commission has thus not been able to abstain from political judgement. 16 In 2005, a new EMU regulation has been adopted, which downplayed the relevance of the 3% and 60% targets. 17 It designed a new obligation for Member States: the so-called medium term objective (MTO) which is differentiated for individual Member States. 18 These MTOs take a longer term perspective and consider the impact of structural reforms (for which a temporary exceeding of the budgetary norms may be necessary). Obviously, this brings especially the Commission more in political waters. In any case, rules-based governance has remained limited to fiscal policies. Policies to prevent and address macroeconomic imbalances and to reform national economies have hardly been subjected to concrete disciplinary norms at the EU level (perhaps with the exception of maximum unemployment rates). Instead, a very light coordination and monitoring regime, based on Broad Economic Policy Guidelines, was set up. This regime did not significantly alter the situation that the central locus for defining these policies remained national, even though such policies may impact the stability of the common currency as well. Some measures that have been adopted since 2008 have actually confirmed and strengthened national responsibilities. This applies inter alia to the obligation for Member States to identify national authorities responsible for making macro-economic forecasts and for assessing compliance with numerical fiscal rules and obligations to ensure their independence. Thus, independent national -fiscal -institutions have become an important element of EMU. Be it an implementation of the original model or a deviation thereof, the systematic flaws of EMU economic policy coordination have been obvious, mostly from the start already. The asymmetry between monetary and economic policies has probably drawn most attention. 19 A second issue has been that the enforcement system of the Stability and Growth Pact (SGP) was not designed to ensure its effectiveness. 20 This flaw was demonstrated in 2003 when German and French non-compliance of the budgetary norms ultimately did not result in sanctions being imposed. 21 The poor democratic quality of decision-making has been another flaw that has been particularly felt at the national level. Economic policies require a strong democratic basis as they typically involve redistributive choices. National parliaments enjoy budgetary rights which are usually protected by the constitution. The EMU has eroded the scope of decision-making for national parliaments, and this declining democratic quality of the decision-making has not been compensated at the EU level. 22 More generally, the EMU was built on market mechanisms rather than on democratic mechanisms. It was deemed crucial to subject Member States to market mechanisms to ensure correct incentives to define prudent public finances policies. An ever more optimistic belief in the market dictated that economies of the European Union would automatically converge. 23

Economic consequences and context
The model of individual responsibility relies on external constraints to foster convergence of the eurozone Member States' public finances and, to a lesser extent, economic conditions. This is meant to make the monetary union more homogeneous and therefore more resilient, on the correct assumption that macroeconomic imbalances represent a threat to its stability.
The reason why economic convergence makes monetary unions more resilient is closely connected with the threat that asymmetric shocks may represent for the stability of the system. As the theory of OCAs explains, exchange rates are a particularly handy tool to ensure adjustments of imbalances in the balance of payments among different economic areas. For instance, and looking only at the trade balance, whenever products and services from a country become more successful -whether for their intrinsic quality or for real wage suppression in their country of origin 24 -flexible exchange rates can work as a shock absorber, as they tend to make those products and services more expensive due to the appreciation of the currency of their country of origin. 25 On top of this, an expansive monetary policy by the country of destination can further reduce the competitive advantage of those products and services, thus helping to reach a new equilibrium. 26 While bringing substantial benefits in terms of reduced transaction costs (particularly through the elimination of exchange rate risks), 27 monetary unions eradicate these stabilisation systems. To be sure, the fact that monetary unions may have some drawbacks says very little, on its own, about their desirability. This depends on the net balance between the benefits in terms of market integration they help to achieve and the drawbacks stemming from the loss of a shock absorber. In this paper, we do not purport to take position on the longstanding debate on the possibility to qualify the EMU as an OCA. What matters to our analysis is, rather, the impact of solidarity, in its different shapes, on the efficiency of the EMU. Solidarity has intuitively the ability to reduce divergencies among different areas of the monetary union and to act as an adjustment factor in case of shocks. However, it is equally evident that solidarity can take various forms, and each of them can be measured against its ability to contribute to a more resilient EMU while, at the same time, to avoid moral hazard and ensure efficiency.
As we mentioned, one can frame the model of individual fiscal responsibility as a form of solidarity. Member States that had historically relied on competitive devaluation of their currency gave away this lever when they entered the EMU. 28 They obtained, in exchange, the ability to fund their public debt on financial markets at cheap real cost, even beyond a level justified by their budgetary conditions. 29 At the same time, the commitment by all Member States to stick to a sound budgetary policy can be understood as a form of solidarity toward all the other participants, and as a precondition for the EMU to deliver its expected results in the long run.
While most of the weaknesses of the EMU were clear from the outset, the expectation that the eurozone economies would converge buttressed the ambitious project of having a single monetary policy without a single budgetary and fiscal policy. Empirical evidence suggests, indeed, that monetary unions tend to foster the integration of the participating economies and to narrow the distances among them, thus reducing the impact of exogenous asymmetric shocks. 30 Value chains are increasingly operating on a cross-border basis, so that national boundaries do not necessarily correspond to the dividing lines that separate areas with diverging economic features. 31 Furthermore, collective wage bargaining appears to become more flexible, due to the enhanced credibility of the central bank commitment not to indulge in an accommodating monetary policy as an easy way to surreptitiously reduce the real value of salaries. 32 However, some divergencies remain in spite of the attempts to reach a higher level of harmonization and integration. All these factors still make the eurozone a non-homogeneous area in many respects. This is the case not only with the persisting weak coordination of economic policies, where EU efforts have fallen short of enforcing a strict budgetary discipline, but also with the differences in national capital markets, notwithstanding initiatives such as the Banking Union and the Capital Markets Union. 33 Take the Banking Union as an example. Excessive reliance on national supervision and the implicit bailout clause by Member States on banks established within their territories led to the ' doom loop' between those banks and state budgets. 34 This connection is not yet entirely broken, as demonstrated by the ongoing debate on the prudential treatment for exposures to sovereign debt. As long as the rules on risk-weighing will ensure a preferential treatment for sovereign bonds, home bias will continue to skew bank exposures and to trigger the risk of a vicious circle between bank crises and sovereign defaults. 35 This hesitance also contributes to deferring the enactment of a full-fledged European deposit insurance scheme (EDIS), which would on its turn have a shock-absorbing capacity. 36 The overall macroeconomic context of the EMU may magnify these inherent weaknesses, as the financial crisis of 2008 and its follow-up on the sovereign debt crisis demonstrated. In a system that removes exchange rate risk, investors from economically central areas with low inflation rates have stronger incentives to increase their exposure towards peripheral countries with higher inflation, as this ensures higher returns in times of economic growth. 37 This accumulation of risks reduces the overall resilience of the system, which manifests itself when an external shock, as the US subprime mortgages crisis, reduces investor confidence. 38 The subsequent capital flee from peripheral areas easily worsens the negative impact of deleveraging in a vicious circle that exacerbates the overall volatility of financial asset prices and may generate boom-bust cycles.
As the recent past has demonstrated, all these factors make the EMU, in its current form, prone to instability in case of asymmetric shocks. Whether the EMU is an OCA or not, there is therefore room for strengthening its resilience. The first model of solidarity relies on the Member States' commitment to coordinate their economic policies and to maintain a sound budgetary policy. Other forms of solidarity, as we shall see, depend more heavily on other tools. Among these, direct and indirect transfers of fiscal resources play an important role, if only as a benchmark to test the current state of the EMU.

Constitutional embedding
Arguably, the model of individual Member State responsibility has the best constitutional credentials. The model is, firstly, firmly grounded on the objectives of Article 120 and 121 TFEU. Since the Treaty of Maastricht, Member States are to conduct their national economic policies with a view to contributing to the achievement of the objectives of the Union (Article 120 TFEU) and, moreover, must treat economic policies as a matter of common concern (now: Article 121 TFEU). These provisions have, however, been further fleshed out in the SGP, which may be considered as a semi-constitutional regulatory package. A set of secondary law measures forms an indivisible whole with provisions of primary law: articles 121 and 126 TFEU. Additional constitutional pillars have been created by articles 123 and 125, thereby completing a comprehensive constitutional arrangement.
This strong constitutional embedding comes with some substantial downsides. As we will see later, it reduces flexibility to incorporate elements from other models to mitigate its negative effects. Furthermore, the model demonstrates a constitutionalisation paradox. The substantive norms enjoy constitutional protection but the enforcement system is weak (as has been discussed above). This is especially so as the CJEU decided in the 2003 saga involving Germany and France that the Council was indeed not obliged to follow Commission recommendations to impose sanctions in case of non-compliance. 39 In subsequent years, it has become increasingly clear that the centrality of political will as a key factor has severely damaged the efficacy of both the corrective and the preventive arm of the SGP. 40 This issue cannot be resolved by simply extending the constitutionalisation of the EMU to all aspects of decision-making and enforcement. To the contrary, the developments of the last decade have demonstrated that the rigidity of the provisions is a problem in itself as it prevents economic shocks to be adequately considered and addressed. Rather, policy discretion and flexibility have surfaced as essential aspects of economic decision-making. Thus, the constitutionalisation of the substantive norms may only exist by the grace of a weak enforcement system.

Democratic legitimacy
According to the Delors Report, the EMU legal framework should be properly embedded in the democratic process, but mostly lacked concrete proposals on how to achieve this. 41 The EMU model as adopted by the Treaty of Maastricht has indeed had poor democratic credentials. Monetary policy, based on the single objective of maintaining price stability and carried out by the ECB, was to remain outside political-democratic decision-making structures. The ECB was designed as an institution with limited policy discretion and was primarily seen as the enforcement institution. For the coordination of economic policies, the lack of strong democratic mechanisms could be justified by the soft nature of the multilateral surveillance system and the lack of any risk-sharing elements. The latter would ensure that national macro-economic policies would not result in redistributive effects among the Member States. For fiscal policies the justification is again different and borrows from both of the other areas. The constitutionalisation of fiscal policies by setting numerical targets (the 3% budget deficit ceiling and the 60% public debt ceiling) effectively removed these aspects from regular political processes (at least insofar as budgetary policy choices would be at stake that would exceed those targets). An additional argument for limited democratic control could be derived from the 39  difficult enforcement structure (even though an excessive deficit procedure was set up). This would add to the argument that the locus of fiscal policy-making would essentially remain national. Nevertheless, EMU's lack of democratic credentials has been obvious, especially with regard to fiscal policies. The substantive provisions of budgetary restraint and the procedural and institutional mechanisms to enforce these have significantly reduced national budgetary autonomy. This has created a systemic tension between the SGP and national democratically embedded budgetary decision-making. Many national constitutions protect parliamentary rights on adoption of the budget, which aggravates this tension even more. 42 This tension could have remained a latent one for a long time, but after the start of the 2008 crisis its impact has been brought fully into the limelight. This has been, first, the result of the tightening of the existing EMU model (e.g. by streamlining the Excessive Deficit Procedure) which has made the effects of EMU law on the Member States more probing. 43 For another part, it has been attributable to the increased complexity of the decision-making structures which makes democratic control more difficult. 44 Thirdly, the measures that have been adopted since 2008 include some which do not fit the traditional model anymore. The ESM treaty, creating a first form of fiscal solidarity, is perhaps the most prominent in that category. It has created particular democratic issues, especially for parliaments of member states receiving financial assistance, but also for parliaments of the contributing countries.

Outline of the model
It is by no means obvious that the ECB would be the central actor in any kind of solidarity-based model. Indeed, the ECB has been set up as an institution independent from political influences, with price stability as its main objective. Thus, stability of the common currency, rather than any type of solidarity between or within Member States euro, defines its mandate. The instruments at the ECB's disposal include setting of the interest rates and providing liquidity to financial institutions. What the ECB is explicitly prohibited to do is to extend credit facilities to the EU, to the Member States or to any of their institutions, and to carry out primary market purchase of sovereign debt (Art. 123 TFEU). As explained in the previous section, this is an expression of the model based on individual responsibility and prevents the central bank or national central banks from interfering with this model. Until the economic crisis that started in 2008, the ECB was therefore neither in its objectives, nor in its decision-making and instruments directly connected to national policies. However, a single currency and the allocation of monetary policy choices upon an independent central institution entail, as such, some forms of risk sharing. Therefore, the functioning of a monetary union inevitably relies on some solidarity mechanism, which has often gone unnoticed in the public debate.
This form of implied solidarity is based on the tools used by the ECB (together with the ESCB) to pursue its monetary policy -their role is comparable to that of a steering wheel in car, so it is convenient to briefly recap them before analysis how they can deliver different outcomes in terms of solidarity. These tools include the accounts opened for credit institutions and the collateralised credit operations, which define the interest rate corridor on the interbank market and are also the key tool for the management of the minimum reserve banks have to hold (Arts. 17 and 19 Protocol No 4 to the TFEU). 45 Open market operations (Art. 18 Protocol No 4 to the TFEU) allow liquidity-providing transaction for short terms (as is the case with the main refinancing operations -MROs) or for longer periods (as is the case with the longer-term refinancing operations -LTROs). 46  post-Covid correspondent, the pandemic emergency longer-term refinancing operations (PELTROs), also rely on open market operations. 47 There are some conditions that these tools, and the model based on 'ECB solidarity' more in general, have to meet to be effective. Chief among them is the credibility of the monetary policy, to which also central bank independence is to a large extent functional. For instance, if the commitment to maintain a low inflation rate is credible, trade unions will reflect it in their requests for salary increases, thus making an inflation spiral less likely. 48 The solidarity mechanisms that are built into the EMU are quite technical in nature, which can explain why the public debate has overlooked them for a long time. 49 This situation changed dramatically as a consequence of the economic crisis. It did so at a particular moment in time, the 26 th of July 2012, when then ECB President Mario Draghi released a famous speech in London in which he announced that the central bank would do 'whatever it takes' to preserve the unity of the Euro. The historical importance of that message can be understood from two perspectives, in line with the underlying idea of this article that technical matters -whether legal or financial -deliver remarkable results in terms of solidarity among Member States (or the lack thereof).
First, the speech was a commitment to resort, in case of need, to unconventional monetary policy. The tools of this optional program were labelled, just one week after the London event, 'Outright Monetary Transactions' (OMTs). OMTs are open market operations, but they are not meant to increase liquidity. 50 Rather, they rely on selective purchases of sovereign debt issued by distressed Member States that are receiving support from the European Stability Mechanism (ESM). This limitation (conditionality) is also meant to ensure that solidarity does not lead to moral hazard. 51 The OMT program was never activated, but it delivered results, nonetheless, just by looming in the background. The credibility of the ECB message was such that a simple declaration of intention sufficed to drive the sovereign spreads down, and to make the activation of the program unnecessary. 52 Second, public opinion in peripheral Member States of the eurozone hailed the message as an anticipation of a more accommodating monetary policy from the ECB, which was then seen as a crucial support for stabilising the economy. 53 This two-pronged impact of Mr Draghi's declaration demonstrated the importance of monetary policy as a solidarity mechanism in the eurozone. A discontinuance in the access to debt markets by a large peripheral Member State would in fact potentially destabilise the entire monetary union. In this regard, the ECB ability to narrow down the sovereign spreads and, more in general, to ensure stability on sovereign bond primary markets -although through secondary market purchases -partially compensates the weakness of the first model of solidarity, based in individual Member States' responsibility.
While the OMT programme was never activated, other nonconventional monetary policy measures equally carried out through open market operations came under the spotlight, namely the Asset Purchase Programmes (APPs). APPs increase the monetary base to foster economic expansion in periods of sluggish growth and to reduce prices (and volatility) for the assets they target, which are also removed from the banks' balance sheets (Quantitative Easing -QE). 54 This is expected to make loans cheaper, to boost investments and, therefore, to bring the inflation rate close to the 2 percent target the ECB established. 55 Among the APPs are the Public Sector Purchase Programme (PSPP), 56 which was the disputed matter in Weiss, and its post-Covid equivalent, the Pandemic Emergency Purchase Programme (PEPP). 57 While the APPs have to date fallen short of bringing the inflation rate at the desired level, their impact on financial market stability and on the overall economic system has been remarkable. To some extent, the ECB has been increasingly seen as a substitute of Member States when these fell short of properly coordinating their economic policies. The ESCB action has been indeed very effective, as the OMT case demonstrates. Part of this effectiveness seems to derive from the ECB ability to adopt strategies that, while sometimes subject to internal debate, 58 do not depend on the intricate political decision-making processes that have demonstrated to be arduous in the other models of solidarity. In this regard, the ECB role allows to by-pass political choices that are particularly sensitive from the point of view of sovereignty. This ability has made the second model of solidarity a victim of its own success, as it charged such model with expectations it might not always be in a condition to deliver.

Economic consequences and context
The creation of the EMU strengthened the ties among the economies of its participating Member States. In doing so, it also introduced some forms of solidarity that find their raison d'être in the economic rationale of a monetary union. These mechanisms can be seen as a result of the EMU and, simultaneously, as the gears that make the EMU work. The first part of this subsection addresses the economic implications, from a solidarity perspective, of the EMU. The second part shows how the Weiss judicial saga has highlighted important limitations -if only form the point of view of legal certainty -that may curb the effectiveness of this model of solidarity.

EMU tools and solidarity
An intuitive consequence of a single currency area like the EMU is that, while higher than in central areas, inflation rates in peripheral members are normally lower within monetary unions, if compared to their level predating the single currency. 59 To the extent that it results from sound and time-consistent monetary policies, rather than from a reduced bargaining power of trade unions, positive but moderate inflation is considered to improve the overall welfare of the economic system because it reduces the implied taxation of increasing prices and facilitates investments. 60 Unexpected low inflation is typically bad news for debtors, and sovereign debtors are no exception. 61 However, members of a monetary union can often take advantage of low interest rates on their sovereign debt, which partially offsets the side-effects of low inflation rates on debt sustainability. 62 In this respect, peripheral countries enjoy the benefit of the ECB credibility in its commitment to maintaining sovereign spreads at a level that reflects the market perception of default risk but not the risk of devaluation of national (proto)currencies. This is, in particular, the level of the sovereign spreads that the ECB has committed to 55  maintain with the OMT program, on the basis that sovereign interest rates reflecting redenomination risks would lead to excessive, hence unjustified, risk premia. 63 Narrowing sovereign spreads that investors consider, rightly or wrongly, at a level that reflects the available information is very close to a (perfectly legal) 64 form of market manipulation. As sovereign debts of different Member States are partial substitutes on the market, 65 open market operations of this kind are to some extent a zero-sum game where part of the market demand for sovereign debt of core countries can be driven to the periphery, in what can easily be seen as a solidarity mechanism. 66 This partial substitution effect is amplified when the open market operations in the context of the APPs deviate, in their relative size, from the proportional stake of the participating Member States in the ECB capital (capital keys). The ability to deviate from the capital keys is a characterising feature of the OMTs, which consist by definition of selective purchases, 67 and is an important element of the PEPP.
Selective purchases and other deviations from proportionality to capital keys lead to a differential treatment for different area of the EMU. In this respect, the ECB-based model displays some flexibility to support solidarity. A similar deviation from the typical model of a monetary union can be seen in the allocation of the assets purchased under the PEPP, and of the exposures these determine. Open market operations carried out under the PEPP reverberate -for an amount equal to 80 percent of their value -on the balance sheets of the relevant National Central Banks, as these purchase eligible securities of issuers established in their own jurisdiction. 68 In this case, however, the deviation reproduces a relevant feature of the individual responsibility model, because it is mean to curb risk-sharing among different parts of the EMU. 69 Finally, it is worth mentioning the implications of TARGET2, the real-time gross settlement system (RTGS) of the eurozone, which clears the interbank cross-border payments of the area. Every time a transfer of money occurs across the border of Member States, the bank whose customer ordered the payment may try to refinance the imbalance on the interbank market. However, when this and other systems for the provision of liquidity dry up -as is the case in a financial crisis -the imbalance is reflected, and settled, on the accounts the ECB maintains on behalf of the central banks of the countries involved in the transfer of money. 70 Outside a monetary union, this kind of imbalance in the balance of payments may grow only until the central bank of the country suffering the outflow can rely on foreign exchange reserves. When the ability of the central bank to stabilise the exchange rate of its currency is exhausted, the central bank will have to adjust. 71 This is not the case in Europe, because national central banks do not have an enforceable legal claim on the credits they have vis-à-vis the ECB, and the other central banks, as reflected in the TARGET2 imbalance. 72 In this regard, TARGET2 has been defined as ' a balance of payments equilibrating mechanism inside the common currency area'. 73 One can easily see that this mechanism delivers a form of solidarity. Imbalances in the balance of payments can be the result of current account or of capital account deficits. As far as TARGET2 reflects a current account deficit, it can be interpreted as a system that grants a refinancing facility from core (safer) countries, whose central banks create base money, to peripheral (unstable) countries, where that money finances consumption. 74 This allows more gradual adjustments of current account deficits. Quite symmetrically, the TARGET2 balance component that displays capital account deficits reflects the ability of commercial banks based in core countries to reduce their exposure towards the periphery -which they piled up during expansionary phases -with limited losses. 75 While TARGET2 imbalances potentially encounter no limits in their size (as long as they are collateralised), the same does not hold true for most of the building blocks of the EMU. The main limitations monetary policy faces in performing the role of a solidarity mechanism are a consequence of Art. 123 TFEU, which as we mentioned prevents overdraft and other forms of credit facilities as well as direct (primary market) purchase of sovereign debt instruments. Far from being a legal restriction deprived of an economic logic, this limitation is inherently connected to the creation of a monetary union where central bank, rather than national governments, has the control of the monetary base. 76 It also strengthens Member States' incentives to maintain a sound fiscal policy and protects the ECB independence, because the fear of losses on sovereign bonds may be an incentive to maintain an expansive policy to support the ailing sovereign debtors. 77 One of the most problematic legal questions originating from Art. 123 78 The crucial, but uneasy, dividing line lies with the need that secondary market purchases do not give primary market dealers any material confidence that they can resell their securities on the secondary market with little or no risk. 79 Functional to this objective are measures like a blackout period between the issue of a security on the primary market and the ESCB purchase on the secondary market, combined with an intended opacity on the precise dates and volumes of open market operations. 80 However, in spite of these safeguards, it is undeniable that secondary market operations have an influence on primary market interest rates. While this is unavoidable and, hence, not in contradiction with Art 123 TFEU, deviations from capital keys aim to selectively reduce sovereign spreads. The CJEU has deemed this feasible in the OMT also because access to that program is subject to strict conditionalities, 81 while this is not the case with other APPs. For instance, the point was not disputed in Weiss because the PSPP is grounded on proportionality of purchases, with a double cap of 33% per ISIN code and per issuer. 82 However, in its follow up on Gauweiler, the FCC reiterated its disagreement with the CJEU as it stated that the selectivity of purchases was perhaps the most problematic feature of the OMTs from the point of view of monetary policy. 83 In Weiss, it deemed proportionality to the ECB capital keys an essential element for the PSPP to comply with Art. 123 TFEU. 84 This makes the pending challenges before the FCC against the PEPP, 85 which allows for deviations from the capital keys with less conditionalities compared to the OMT programme, 86 particularly difficult to handle from the perspective of the no monetization clause.
All in all, the uneasy relationship between proportionality to capital keys and Art. 123 TFEU is a result of persistent excessive divergencies within the eurozone. This shows a broader limit of the EMU as a solidarity mechanism. In a monetary union that is composed of nonhomogeneous economies, the central bank may easily find itself in the difficult condition to determine a monetary policy that is suited to some areas of the union, but not to others. Absent shock-absorbing factors -such as flexible wages, labour mobility, or direct fiscal transfers -which can partially replace the effects of flexible exchange rates, asymmetric shocks may misalign the growth pace of different Members of the Union. In this context, areas with diverging economic dynamics would need equally diverging monetary policies, but this would be of course at odds with the very idea of a monetary union. 87 Therefore, in a context where the economic dynamics are desynchronised, a single monetary policy might not fit all the areas of the currency union. When this is the case, accommodating the needs of some areas may easily make other areas worse off. Part of the economic literature concludes, for example, that the ECB monetary policy predating the 2008 financial crisis was excessively driven by the needs of the core areas of the eurozone, 88 while the peripheral zones were underweighted compared to the optimum. 89 Somewhat symmetrically, the FCC in Weiss claims that the ECB monetary policy of more recent years accommodated for the needs of peripheral countries to ensure these had continuous access to the market, while leading to unintended consequences whose negative effects reverberated particularly on core countries that did not need that support. 90 These conflicting claims show the risk that the ECB's role may become de facto more political. Every monetary policy decision is, of course, inherently political, but the ECB has weak democratic representation and grounds its legitimacy on its independence. It is therefore important that the ECB is not dragged into a condition to make decisions that go beyond to its mandate.

EMU and judicial scrutiny: implications for solidarity
The scope of the ECB's leeway in defining EMU monetary policy also depends on the interpretation of the legal constraints to its powers, and on the intensity of judicial scrutiny in this regard. As we mentioned, a call for a less deferent approach by the CJEU in its assessment of the ECB decisions recently came from the FCC in Weiss, under the assumption that this would be needed to protect the ECB independence. 91 The decision has stirred a lively political and academic debate, and we do not purport to opine on the merit of the FCC stance. 92 To the contrary, we plan to highlight the consequences of these judicial developments on subsidiarity. Different interpretations of proportionality can indeed deliver equally different results in terms of the ECB ability to accommodate for individual Member States' needs. The FCC claimed, in particular, that the CJEU adopted too lax a proportionality test when screening compliance with Art. 5 TUE, thus granting the ECB excessive discretion in balancing the multiple interests that monetary policymaking involves.
This judicial clash highlighted persistent uncertainties surrounding the criteria courts should adopt when scrutinizing ECB decisions, and it appears that the unsettled allocation of jurisdiction among the CJEU and national constitutional courts (or their equivalent, depending on the Member State involved) will remain an obstacle to future clarifications. To understand why, let us first consider the CJEU interpretation of the proportionality test.
In a nutshell, the CJEU grounds its assessment of the PSPP proportionality in Weiss on the basis of three guiding criteria, two of which are expressly stated at the outset of the assessment. 93 The first criterion is whether the open market operations are suitable to achieve the policy objective of the PSPP. 94 The second criterion is whether these measures are also necessary, as they should not go beyond what is required to deliver the expected outcome. 95 In the third place, the CJEU also evaluates whether the ESCB weighed up the interests involved by the PSPP so as to prevent disadvantages which are manifestly disproportionate to the objectives of the program. 96 This three-pronged assessment by the CJEU is in line with the approach that the same Court often follows when reviewing discretional acts. 97 However, it is worth highlighting that, in the CJEU decision on Weiss, the third criterion, which assesses whether the PSPP is proportional stricto sensu, only considers the magnitude of the risk of losses for the ESCB and its components, with no reference to the overall economic impact of the PSPP. 98 In the FCC's opinion, this limitation weakened the CJEU proportionality test. 99 More in detail, the FCC believes that the impact of the monetary policy should be balanced against its economic policy implications. If these implications turned out to be prevalent, the measure would fail the test. 100 This approach can easily be interpreted as a way to test whether a measure is excessive (or proportional stricto sensu), in addition to being both necessary and sufficient to achieve its declared objective. 101 What remains unclear in the FCC decision is, however, how this test should precisely look like as regards the implications of factoring the economic impact of monetary policy measures. A first reading of the narrow proportionality test as interpreted by the FCC is that this tool can help determine the boundaries between monetary and economic policy with a view to enforce the principle of conferral. 102 As one of the judges of the FCC senate that decided on Weiss stated, 'the constitutional justification of the independence of the ECB is […] limited to a primarily stability-oriented monetary policy strictu [sic!] sensu and cannot be transferred to other policy areas', 103 including economic policy. As a consequence, an APP should be deemed disproportionate if the economic policy effects of that measures are neglected. 104 A logical consequence of this approach would be to focus the analysis of narrow proportionality on the size of the economic effects of the monetary policy measure. In other words, a measure that delivered a limited monetary impact, but a major economic result, could hardly be considered as monetary in nature, as it would go beyond its merely supportive role (Art. 127 TFEU) and would encroach upon competences that the TFEU leaves to Member States, with the coordination role of the EU (Art. 119 TFEU). Therefore, a monetary policy measure could be regarded as disproportionate, under this interpretation, if its economic spill overs were materially larger than its monetary implications. 105 If this reading were correct, the question would however remain how precisely to distinguish the economic and the monetary effects of the disputed measures, as the two are often closely intertwined. 106 After all, this distinction is largely artificial and the silence of the Treaties on this point can also be considered as a form of strategic incompleteness. This silence could have been an easy way to hide potentially diverging views on the desired level of economic integration (and solidarity) underlying the EMU. In this respect, European and national courts now share the uneasy task of shedding light on a matter that the writers of the Treaties might have decided not to tackle, because clarifying it would have prevented an agreement.
There is however a second way to interpret the additional criterion the FCC suggests in Weiss. If the proportionality test is run to assess the rationality of the monetary policy measures, 107 the focus of the ECB and 94 Ibid, paras 73-8. 95  of the courts should be not so much on the size of the economic impact of the monetary policy measure, but rather on its net benefits. 108 If a measure is expected to deliver net positive benefits, it should be deemed rational. This is precisely how the ECB showed to interpret its mandate in the follow-up of Weiss. The minutes of the ECB Governing Council of 3-4 June 2020, which were meant to clarify the reasoning behind monetary policy choices, 109 provided a qualitative analysis of the upside and downside risks of the PEPP revision, and came to the conclusion that the benefits of the proposed enlargement of the programme outweighed the side effects. 110 Apparently, no consideration was paid to the fact that the economic policy consequences could be, in absolute value, larger than the -admittedly, hardly distinguishable -purely monetary ones. In this regard, the narrow proportionality test closely resembles a rationality check. And, indeed, one cannot but notice that even the FCC decision seems to focus on a selective list of potentially negative consequences, rather than on a broad-spectrum analysis of all the implications of PSPP on the economy as a whole. Nonetheless, even a broad agreement that narrow proportionality should be understood, in this context, as a rationality test might not be sufficient to support a common understanding of its operationalization. Just like any other form of efficiency considerations, this assessment would require a preliminary understanding of its precise contents.
In a first sense, efficiency can be understood as requiring that at least one of the participants in a system, such as a Member State in the eurozone, is better off, while none of the other participants is worse off (Pareto efficiency). 111 In a second sense, a measure can be deemed efficient -and therefore rational and proportional -if it delivers a net gain to the system as a whole. Therefore, the measure can make some participants worse off, as long as the better position of the winners more than compensates this (Kaldor-Hicks efficiency). 112 As one can easily see, both notions leave room for uncertainties. In a non-homogeneous monetary union, satisfying the needs of all Member States simultaneously may be very challenging. 113 At the same time, quantifying to what extent the upsides in a specific area can offset the downsides in another one is a tricky exercise.
The implications of these doubts concerning proportionality on the actual shape of solidarity within the EMU should be, at this point, quite obvious. Enabling monetary policy measures that deliver positive benefits regardless of the qualification of those outcomes as 'monetary' or ' economic' can offer the ECB broader margins of discretion. In a similar vein, looking at net positive benefits for the system as a whole, rather than for each component of the system, allows the adoption of policies which may result in second-best scenarios for some areas of the union, as long as they represent a first-best for the EMU as a whole. While this would enhance solidarity as we interpreted it above, many questions remain open on what approach should be followed.
Contributing to these uncertainties is also the uneasy interplay between the CJEU and national constitutional (or equivalent) courts, which makes the provision of clear guidance on the limits of the ECB mandate all the more difficult. In the light of the principle of conferral (Art 5 TUE), national courts -including the FCC 114 -tend not to recognize Kompetenz-Kompetenz to the CJEU, and therefore retain the power to police the application of the principle of conferral by the EU institutions under the ultra vires review or other similar doctrines. 115 At the same time, the CJEU is the only court in charge of interpreting the Treaties, including the boundaries of the powers allocated to the institutions (Arts. 19 TEU and 267 TFEU). As the recurrent invocation to judicial dialogue demonstrates, this system fails to deliver a clear hierarchy -not just of courts, but -of the logical statements that compose different judicial decisions. For this reason, the lack of hierarchy leads to self-referencing and, in turn, to undecidability of questions. 116 To be sure, perfectly hierarchical and non-self-referential systems are utopian -as Kurt Gödel demonstrated. 117 However, the current system falls short of even getting close to such a result. It is not hard to imagine that national courts and the CJEU may well take a different stance when defining the notion of efficiency that should underpin the ECB decisions (and the judicial scrutiny thereof). The ECB (for instance when deciding by majority) 118 and the CJEU may take and, respectively, support decisions that maximise the upsides for the EMU as a whole, at the risk of making some areas worse off. A similar approach can hardly be expected, however, from national courts of these areas.
It is no coincidence that all these doubts stemming from the FCC approach are strictly connected to the very notion of solidarity among different areas of the EMU. Their clarification would require policy choices that, due to their political nature, neither the ECB nor the CJEU can easily make without jeopardizing their independence -a remarkable unintended consequence if one considers the FCC intentions.

Constitutional embedding
A central bank that forges ties of solidarity when it was designed as a technocratic institution with a single responsibility to maintain price stability is bound to raise constitutional questions. Such questions have indeed arisen and have (with regard to the OMT and the PSPP programs) been addressed by the CJEU, as we mentioned. The main issues have revolved around the scope of the ECB's -indeed quite narrow 119 -mandate and the prohibition of monetary financing (Article 123 TFEU). 120 The latter prohibition relates to the ECB as well, as the CJEU in Pringle emphasized that it is only directed to the ECB and to other EU institutions. 121 This prohibition states in other words that Member States or the EU cannot borrow money from the ECB. 122 The CJEU has first had to rule on the OMT program following preliminary questions from the FCC in case Gauweiler. The German court considered that the OMT programme had crossed the red line of article 123 TFEU and was moreover was politically motivated. 123 The CJEU, however, held that monetary policy is not clearly defined in the Treaties, but that rather regard should be had of the instruments available and particularly the objective they serve. 124 Subsequently, the court considered that the primary objective of the programme is still to maintain price stability through the maintenance of the monetary policy transmission tools; the fact that the program would also contribute to the stability of the eurozone -indeed a matter of economic policy -does not change this analysis. 125 Thus, the ECB's 'stability paradigm' was still in place and its mandate broad enough to encompass the OMT programme. 126 After the CJEU's decision, the German constitutional court, although admittedly grudgingly, accepted the conclusion from Luxembourg.
The argumentation of the CJEU in Weiss was much in line with (and further elaborated) 127 the decision in Gauweiler and resulted in a confirmation of the legality of the PSPP program. 128 For the CJEU, the independence of the ECB (Article 282(3) TFEU) resulted in a high degree of discretion in designing monetary policy. 129 Consequently, the Court relied strongly on the statements and justifications provided for by the ECB and refrained from carrying out a comprehensive assessment on the merits itself. 130 The Court concluded that in the absence of a specific and concrete definition of price stability (and thereby of the main objective of EU monetary policy), the ECB had discretion in specifying that objective to mean inflation rates below but close to 2% over the medium term. It agreed with the ECB that the PSPP would indeed contribute to this objective and the fact that the program may have a myriad of economic consequences as well did not alter that conclusion. It is precisely along this line that, as we mentioned, the Court equally considered the PSPP a proportionate measure in relation to its -monetary -objective. Furthermore, the PSPP did not infringe the prohibition on monetary financing, as the assets at issue were to be bought on the secondary market and sufficient guarantees accompanied the program.
The FCC's rejection of the CJEU's decision has various constitutional dimensions. 131 First, the FCC's critical stance marks a departure from or an exception to its regular approach to matters of EU law. Normally, the legality of EU measures is primarily to be decided at the EU level itself and the FCC exercises restraint but, as we have seen, this hierarchy is far from being carved in stone. The FCC justifies this departure by observing that a 'fundamental interest' of the Member States is at stake here. This shows how the constitutional elements of the whole EMU architecture can be inherently weak. Second, the FCC interpretation of proportionality, while somewhat uncertain in its economic implications, surely serves as a mechanism to balance competences between the EU and the Member States, so that its constitutional character is stronger than in the CJEU understanding. 132 It is precisely for this reason that, in the FCC's view, the economic effects of the PSPP program should be balanced against its monetary purpose. For the FCC the conclusion would apparently be obvious that the PSPP would need to be brought under the economic policy domain. Thirdly, the FCC endorsed the ECB's independent position within the EU law framework, but -in line with prior decisions -observed that its decisions should remain subject to full judicial review.
EU law scholars and others have been critical of the FCC's decision in Weiss and with more than sufficient reason, one may add. However, even if one accepts the CJEU's reasoning in Gauweiler and Weiss -and thus denies the illegality of instruments such as the OMT, the PSPP and the PEPP -the constitutional vulnerability of these programs remains. Future programs will inevitably be equally contested and subject to legal claims and so may the position of the ECB itself, the scope of its powers and the ways it may be held to account, both judicially and politically.

Democratic legitimacy
The FCC's critical assessment of the PSPP program is arguably rooted in a concern for democratic legitimacy. The FCC's reasoning is intricate, it is rooted in a series of earlier decisions, but it also -as we will see later -reflects a rather limited understanding of how democratic requirements could be fulfilled in this context. The FCC has derived from the Basic Law an individual right to democracy which has several implications for the PSPP program. First, the German right to democracy demands that every act of public authority should ultimately be based on the will of the German people. This requires EU authority to be founded on an explicit act of approval by the democratically elected state institutions. This is complemented by the FCC's jurisdiction to review whether the EU has indeed acted within the limits of the approval given by these institutions. Ultra vires control thus forms a cornerstone of the FCC scrutiny of EU acts, albeit that the Court applies a qualified test. The second element of the German constitutional right to democracy is the notion that the core of democracy should remain national in any case and that certain powers may therefore not be transferred to the EU, even if democratic institutions would choose to do so. The authority of the Bundestag to decide on budgetary matters is part of that core. Thirdly, with the ratification of the Treaty of Maastricht, the FCC has accepted the ECB's independent position within EU law as a justified derogation from the right to democracy. This derogation is built on the 'proven and scientifically supported particularity of monetary policy' which imposes that an independent central bank is a better guarantor for monetary stability than organs which rely on the approval of political actors. However, as this constitutes an exception to normal democratic decision-making the German democratic institutions have a qualified responsibility (Integrationsverantwortung) to monitor the execution of the European integration agenda (Integrationsprogramm) with regard to ECB decision-making.
This approach is obviously typical to the FCC. Questioning the democratic credentials of ECB decision-making and their relation to ECB independence has, however, become a key aspect of EMU and EU institutional law scholarship. 133 This has included ways to increase transparency, 134 to move from procedural to more substantive forms of accountability, 135 to assess existing mechanisms of accountability (such as the Monetary dialogue with the European Parliament) 136 and to zoom in on particular (new) tasks of the ECB. 137 Although the trade-off between democratic accountability and independence (the latter sustained by a narrow mandate) has been challenged, 138 this still marks the discourse on ECB accountability. Thus, the expansion of ECB activities to include programs such as the OMT, the PSPP and the PEPP will only increase the calls to improve the democratic legitimacy of ECB decision-making.

Outline of the model
The third and final model of solidarity would signify a much higher level of integration. It would entail setting up substantial EU budgetary policies with a significant degree of autonomy.
-Revenue: fiscal capacity and fiscal autonomy. The EU would have greater autonomy over its revenues through specific taxation powers and the ability to borrow funds on the capital markets. The latter would entail a certain level of risk sharing and mutualization of debts. -Expenditure: the EU would have significant powers to allocate funds, resulting in the redistribution of wealth within and among Member States. -Governance: as such budgetary powers entail key political decisions on the revenue and expenditure side, they require strong democratic legitimacy.
It is true that the current EU budget meets these requirements to some extent. It has some redistributive effects (through the EU's Structural and Investment Funds) and it includes the European Parliament in its adoption. 139 However, the European Parliament is only consulted on the Own Resources Decision (ORD) which establishes the basis for the EU's revenues. 140 The EU's autonomy is limited, however, as Member States contributions constitute the main source of revenue. Moreover, the EU's budget is quite modest in size, thereby limiting its redistributive effects. 141 Especially since the start of the economic crisis, it has been argued that a genuine Fiscal Union, based on risk-sharing, would be necessary to complete EMU. 142 Effectively addressing macro-economic imbalances and maintaining the stability of its currency in times of crises would not be possible without further steps towards fiscal union. 143 According to Hinarejos, fiscal federalism based on a central fiscal authority would be one of the ways to achieve this. 144 It is questionable, however (as Hinarejos already acknowledged) whether this would be politically feasible and whether the EU has the appropriate political system to deal with such key questions as the redistribution of wealth between richer and poorer Member States (the so-called 'Transfer Union'), decisions on mutualization of debts and the identification of priorities for economic reforms at the EU level.

Next Generation EU
The plans that have been developed to address the consequences of the COVID-19 pandemic contain, however, several elements that fit this model of greater integration. Under the umbrella of Next Generation EU, the EU's political institutions have been designing measures to recover from the pandemic in combination with the new Multiannual Financial Framework (MFF). Some of these measures may be qualified as conventional (e.g., the investments in the Horizon 2020 program), but the core of the approach concerns an ambitious investment plan backed by a substantial recovery fund. The Corona recovery fund is based on a regulation. 145 Member States may apply for loans but also for grants. After the Commission has assessed national recovery plans the Council may decide to grant funding. Apart from this solidarity in granting financial assistance, the corona recovery fund includes solidarity on the liability side of the fund's balance sheet as well. Part of the funding may be attracted through borrowing on the capital markets, the liability for which is assumed under the EU budget. Thus, the Member States bear a common responsibility, and the Corona recovery fund entails a mutualization of debt. The legality of this new tool may be derived from Article 311(3), which allows the EU to introduce 'new categories of own resources.' Still, as the Commission Legal Service has pointed out, introducing a borrowing competence is ' quite foreign' to the current system of own resources of the Union. 146 As an 'incidental component' of the ORD (the Corona Recovery fund is set up as a temporary instrument), however, the Legal Service deems the inclusion justified. 147 Indeed, the corona recovery fund is an emergency measure -which is thereby of a temporary nature -but its size and the way the Regulation breaks with hitherto orthodoxies of EMU law marks the crossing of a Rubicon. 148 Yet another sign is that the recovery plan includes the option to finance the repayment partly through EU taxes. Partly as a revival of earlier plans, the 'EU plastic packaging waste tax' (which would in reality be a levy to be paid by Member States into the EU budget); the plan for a levy on Member States' emissions trading system (ETS) revenues; the 'Digital Services Tax' and the proposal for an own resource based on the Common Consolidated Corporate Tax Base (CCCTB) have been mentioned. 149 This model involves a strong Europeanization of economic policies, and has been promoted by both the Commission and the European Parliament. 150 If this model were to be adopted outside emergency situations and a certain level and form of fiscal union were to be created, these institutions would gain significant control and political preferences of the Member States would determine revenue and spending to a much lesser extent. 151 This has been indeed the ambition of the Five Presidents. In their 2015 Report they suggested to create a eurozone fiscal stabilization mechanism to address macroeconomic shocks. 152 A euro treasury would be locus for collective decision-making in this regard and would ensure the democratic quality thereof but this plan has not been further elaborated.

Economic consequences and context
As we have seen, monetary unions can bring substantial benefits in terms of market integration, as they erase the exchange rate risks for transactions that take place within their geographical scope. Exchange rate is, however, a macroeconomic stabiliser, so that participants to monetary unions might be prone to higher instability compared to their previous situation.
The magnitude of this problem is subject to debate, however, as there are other mechanisms that can have a stabilising effect and that -if and when they exist -a monetary union does not remove. Economists mention, in particular, factor mobility (and job mobility more specifically), real wage adjustments, and fiscal transfers. 153 On top of this, monetary unions may also make these shock absorbers relatively more efficient, so that the net impact of removing exchange rates may not be as negative as it might appear at first sight. 154 However, the EMU can enjoy the benefits of these stabilising factors to a limited extent. Real wage adjustments depend on the organization of the labour relationships within each country, and on the relative strength of labour unions. 155 All these conditions are still reflecting idiosyncratic elements of EMU Member States. They are also very sensitive from a societal and political point of view, particularly in the European tradition where job markets are often less flexible than, for instance, in the United States. 156 But there is a second reason why the United States display more features of an OCA, compared to the EMU. This is job mobility, which in the European Union is constrained by persistent language barriers -another relevant difference with the United States. This fragmentation of the job market prevents job mobility from becoming a truly cross-border stabilising factor. 157 It comes therefore as no surprise that the third model of solidarity, the most recent one, relies on fiscal transfers. These are also subject to constraints, however, as long as individuals perceive themselves as members of their national community rather than European citizens, and as long as European governance is perceived as incapable of avoiding moral hazard by the countries that are net receivers of fiscal transfers. In this sense, Member States' individual responsibility is a prerequisite not only for the proper functioning of a common monetary policy, but also for a viable system of fiscal solidarity.
It was intended to cushion macroeconomic shocks and was to be designed to fund specific projects and draw funding from equally specific resources. An important condition for granting funding would be that the regular incentives for sound fiscal policy making and addressing national structural weaknesses would remain intact.
In this light, the main principles of the EU's economic constitution would remain largely unchanged. Especially in light of the CJEU's flexible approach to earlier measures to address economic problems, it is unlikely that the current Next Generation EU measures would be declared illegal. This does not mean to say that the changes to the economic constitution should not be reflected in the TFEU, however. Indeed, it has become widely acknowledged that economic policy based on coordination alone is inadequate and needs to be supplemented by a certain level of fiscal solidarity. A limited and targeted form of fiscal solidarity indeed, but still one that would certainly mark a departure from one of the most fundamental principles on which EMU is founded, the principle of exclusive Member State fiscal responsibility as we are witnessing what perhaps may be seen as a first step towards an independent common EU budget.

Democratic legitimacy
The fiscal union model poses probing questions on how to ensure parliamentary involvement. The main reason is that a model based on fiscal union introduces a hybrid system of economic decision-making -or it will at least significantly complicate the existing system based on the European Semester. The core thereof is that the decision-making is shared between EU and national levels. The experience has shown that executive authorities usually handle this better than parliaments. How parliamentary control would be designed in a situation of greater fiscal union remains unclear. The debates in relation to the adoption of the Next Generation EU measures indicate, however, that the issue is not easily resolved or indeed even addressed. That has been demonstrated in relation to the adoption of the SURE instrument and the RFF (the corona recovery fund). Even though these measures directly affect 'the power of the purse' (in terms of both spending and revenue) the role of national parliaments in the adoption of these measures has been limited and, in any case, highly asymmetrical. In case of the SURE instrument, it was mostly the parliaments from the so-called Frugal Four countries that actively scrutinized the proposal. But even their impact on the substance of the proposal has been almost non-existent. 164 Even the RFF proposal, which entails a measure of a significantly greater magnitude, attracted less parliamentary attention than could have been expected. 165 The European Parliament has been more actively involved, especially in the adoption of the RFF. Nonetheless, despite its efforts to ensure its involvement in the ex-ante democratic scrutiny and 'in the expost verification that funding under the RRF is well spent, it has not been able to acquire substantive powers in the implementation of the RFF'. 166 Fasone argues how this is similar to the European Semester but that the situation is more problematic in this context. Indeed, decisions on national recovery plans may directly affect EU spending. Thus, the EP as one of the budget authorities of the EU should not be side-lined. The position of national parliaments in the implementation of the RFF is equally unclear, although the normal scrutiny and accountability mechanisms may be applied.
The decision-making on the RFF and SURE thus demonstrates how difficult it is in practice to design proper mechanisms to ensure democratic legitimacy. This is a risk for any step towards fiscal union, a risk which is further exacerbated by the lack of a clear model to elaborate the EU's two-tier democracy (Article 10 TEU) in this field.

Solidarity models in EMU: Fierce competition or peaceful co-existence?
The key provisions of EMU -on fiscal rules, the no-bail out clause, the prohibition of monetary financing and the independence of the ECB -are by no means isolated provisions. Together they shape a coherent model of fiscal solidarity based on individual Member States' responsibility. In the last decade, these provisions have all been challenged. Nevertheless, no outright violations of these provisions have thus far been established by the CJEU. This observation should not, however, obscure the fundamental tensions with newer solidarity models that these challenges to the fundamental EMU provisions demonstrate. The economic crisis has demonstrated that the original EMU model is incomplete and simply unfit to address a-symmetric shocks. Economic stability but also convergence may only be effectively achieved by collective rather than individual Member State action. 167 Still, the original solidarity model is built on firm ground. That is true even if in the current COVID-19 crisis the public deficits and public debts rules have been partly disabled. This can legally only be a temporary and emergency measure. Whilst the current situation may reduce the sense of inevitability of these rules, it is unlikely that the Member States would agree to a fundamental redesign of EMU in which the original model would be relinquished. Furthermore, the importance of sound national budgetary policies has only been highlighted by the COVID-19 crisis as Member States in better budgetary situations have proved to be much more resilient in coping with the crisis. Even in the most far reaching scenarios of prospective EMU development, the suggested level of fiscal union is at most partial. EU fiscal capacity would even in those scenarios remain only supplementary to national fiscal policies. Thus, the second and third models of solidarity we have highlighted above are not to be seen as the logical next stages in the development model of EMU and/or as a sign that the EU economic governance has reached a level of maturity that allows relinquishing individual Member State responsibility. The original solidarity model will not disappear and constitutional limits, as well as economic and democratic imperatives, dictate that it should not. The recent judicial and political developments in the wake of Covid-19 cast a bright light on the flaws of that model, but they also make clear Member State responsibility will not disappear with the adoption of alternative solidarity models.
The introduction of elements of the other models has, however, provoked new tensions. The CJEU has been able to avoid declaring EMU measures illegal, but this has required quite some legal acrobatics. In the case of the PSPP, it involved a narrow reading of the prohibition of monetary financing combined with a broad interpretation of the ECB's monetary mandate and its independence. There are limits, however, to such interpretative stretching of Treaty provisions, if only in terms of the willingness of national actors to accept it (as the Weiss saga has demonstrated).
But there is a deeper tension between possibly conflicting models of EMU solidarity that goes beyond the legality of specific measures in the light of equally specific provisions of EMU law. Up until now this tension has been addressed by declaring the second and third model as the exceptions to the standard model based on individual fiscal responsibility. The corona recovery fund will be designed as an emergency and temporary measure. The ESM Treaty provides that financial assistance under the ESM shall only be provided if this is 'indispensable to safeguard the financial stability of the euro area as a whole and of its Member States' and only under 'strict conditionality.' In the same vein, the PSPP program has been justified to meet the objective of price stability in an unprecedented economic and financial environment (which caused the monetary transmission mechanism to cease functioning).
But exceptionalism as an argument to justify divergences from the original EMU model has lost much of its strength. The fundamental inability to absorb economic shocks and to support economic convergence of the first model is not the only reason. The magnitude and scope of measures such as the PSPP, the PEPP and the Corona recovery fund have created an economic and political reality that is neither reflected in the formal EMU constitution nor fortified by adequate democratic safeguards. Calling on the exceptionalism argument, furthermore, turns a blind eye to the snowball effect crisis measures are subject to. The EU's governance system is full of elements that started off as incidental and exceptional initiatives but which have organically evolved into more permanent and structured parts of the EU. Such incremental evolvement risks that fundamental questions of design remain unaddressed. The re-design of EMU has now become unavoidable. If we accept that EU economic governance is today build on not one, but three distinct models the main issue is to assess how these should be combined and how they should interrelate. Existing and possible future tensions between them should be effectively addressed rather than being concealed by invoking the exceptionalism argument. This requires a coherent approach to the economic, constitutional and democratic dimensions at stake. Herein lies an important task for the Conference of the Future of Europe, even though its mandate contains no explicit reference to this.