More power, less influence: European union actors in media debates on fiscal policy after the eurocrisis

ABSTRACT Stronger European Union (EU) competencies come with greater media presence, according to existing cross-sectional comparisons. While presence comes with the power to influence public discourse, we know little about how it affects the overall tone of public discourse. This article investigates the effect of the empowerment of EU institutions on media presence and the tone of debate in the EU. It does this through an original claims analysis of newspaper articles in Germany, France, Italy, Spain, Denmark, and Poland during the height of the Euro crisis (2011–2012), the Greek bailout crisis (2015), and the Italian budget crisis (2018). EU media presence indeed increases following empowerment. However, this does not result in a more pro-European debate, because pro-European national voices are crowded out. While this does not affect the desirability of austerity, it does harden Eurosceptic public discourse.


Introduction
'Who inhabits the European public sphere?', is a classic question asked by scholars interested in the Europeanization of public spheres and the alleged communication deficit of the European Union (EU) (Koopmans 2007;Koopmans and Erbe 2004). The interest in this question is both normative and positive. From a normative perspective, presence of powerful actors in the news is often considered a prerequisite for democratic accountability (Habermas 1996;Steffek 2003). Part of the job of journalists is to hold those in power to account (e.g. Deuze 2005). Without such public scrutiny, well-organized special interests and the mobilized few may have excessive influence at the detriment of the less-invested majority. From a positive perspective, presence in the news is a prerequisite for shaping public discourse. For EU institutions, presence in the news provides the opportunity to bolster their own legitimacy in the public's eye and to shape the narrative. Absence means the debate is shaped by other actors, notably state actors. These actors frequently claim credit for collective decisions while blaming policy failures on the EU (Hobolt and Tilley 2014;Schlipphak and Treib 2017). This blaming may ultimately undermine EU legitimacy. The extent to which EU actors are present in the news is thus of interest to citizens and others interested in the democratic legitimacy of the EU, not in the least EU institutions themselves. In their key cross-sectional comparison of Europeanization of public spheres across seven policy areas, Koopmans and colleagues find that the more powers EU institutions have in a given policy field, the more present they are in the news coverage of those issues (Koopmans 2007;Koopmans and Statham 2010). Thus, EU actors have a much stronger voice in news articles on monetary policy and agriculture than they do on immigration, troop deployment, or education (Koopmans, Erbe, and Meyer 2010, 65).
But presence in the public sphere is not automatically a good thing (Gaber 2009). First of all, journalists tend to focus on scandals and conflict. Chances are thus that when EU institutions make it into the news, it is not because they have performed well. Reporting on scandals may undermine an institution's legitimacy, even if it comes with increased publicity. More interesting for our purposes here, is that the opportunity to present one's own arguments in the news often provokes criticism. Opposition may be triggered by the substantive arguments EU actors make, the way they frame it, the system they represent, or the personality, nationality, and ideology of the actor making the arguments (De Wilde and Trenz 2012, p. 541-542). This is part of the dialectic nature of public debate, where arguments provoke counter arguments (Trenz and Eder 2004). Additionally, journalists often try to provide a platform for 'both sides of the story' to uphold objectivity and present citizens with alternative perspectives on public policy (Deuze 2005, 456).
It is thus an open question how increased presence in the public sphere of EU actors affects the nature of debate. On the one hand, we expect that empowerment of EU actors through European integration and further pooling of sovereignty leads to greater presence in the news and thus a stronger presence of their pro-European narrative. On the other hand, greater presence could provoke a stronger presence of counter narratives through the dialectic nature of the public sphere and through mass media's attempts at balanced reporting. There are, in other words, different mechanisms in play that lead to different outcomes and their net effect on the nature of the debate remains uncertain. Additionally, we know very little about how these mechanisms and their net effects differ from country to country, depending on their involvement in particular EU policy fields. This paper thus asks: how does the empowerment of EU actors affect the nature of public debate about European issues? It investigates this through a comparative case study of debates on Economic Monetary Union (EMU) and fiscal policy. During the Euro crisis, a range of defacto and dejure measures that empowered EU institutions were implemented to counter the crisis. This includes the European Central Bank's (ECB) new policy of quantitative easing and the creation of the Fiscal Compact, for example. This article presents new content analysis data on EMU and fiscal policy debates, collected during three key episodes of contention: the height of the Euro crisis (2011)(2012), the referendum on the Greek bailout (2015), and the Italian budget crisis (2018). This case selection allows for the comparative analysis before (2012) and after (2015 & 2018) the effects of EU empowerment in a particular policy field. Newspaper debates in six EU member states -Germany, France, Italy, Spain, Poland, and Denmark -are analyzed. This includes two of the richer crediting Eurozone members (Germany, France), two of the heavily indebted nations that suffered particularly during the Euro crisis (Italy and Spain) and two countries outside EMU (Poland and Denmark). This design allows for analyzing the effect in differently involved member states. The article presents the results of claims analysis (Koopmans and Statham 1999) in 12 newspapers, two per country, with a total of 1.407 claims. It finds that the empowerment of EU institutions during the Euro crisis resulted in a stronger presence of EU actors in the news in the two subsequent fiscal crises. However, while EU actors continued their pro-European narrative, the overall tone in the news became more Eurosceptic. The increased presence of EU institutions came at the cost of pro-European national voices in the news and EU institutions functioned as a red flag for national Eurosceptic voices. Newspapers provided less space in their coverage for pro-European national and societal voices. While EU institutions may have gotten more power during the Euro crisis, they at the same time seem to have gotten less influence in shaping media debates.

Theory
In order to understand what kind of effects the empowerment of EU institutions might have on news coverage, we need to start with general media logic. That is because media logic to a large extent influences the daily decisions that editorial boards and journalists make on which stories to feature in the news, and how. Since there are many more events and stories every day than can be covered in any particular news outlet, journalists perform a function as gatekeepers (Shoemaker and Vos 2009). They select the few stories they have the resources to cover and deselect all others. This happens on a daily basis for individual stories and events, but it also unfolds on a more structural level, as news outlets make decisions on how many reporters to maintain on staff, where they should be based, and which type of events and stories they should be following or investigating. Whether to have a permanent reporter -or several -based in Brussels with the mandate to cover the EU has a major influence on the extent to which a particular news outlet covers EU politics, and the key actors involved in making it. In both those daily and those structural decisions, journalists and editors employ a range of news value criteria to decide what is news, and what is not (Altheide 2004;O'Neill and Harcup 2009).
Part of those practices and criteria evolve around identification, which leads journalists to primarily feature actors in the news that readers and viewers identify with. Mass media remain nationally organized in the EU, catering to the audience of any given member state. Given that media outlets cater to national audiences, they also predominantly feature national actors (Koopmans and Statham 2010). The identity criteria of media logic leads German journalists to provide a dominant spot for German actors in the news, French journalists focus on French actors, and so on (De Wilde 2019; Kleinen-von Königslöw 2012). But identity is not the only news value that structures the decisions of journalists and editors. Valence criteria lead journalists to feature the powerful in the news. Media are supposed to perform a watchdog function on those in power and the actions of powerful actors have most effect on the daily lives of readers and viewers. The combination of identity and valence criteria means that Europe in the news is dominated by national actors, particularly national executive actors (De Vreese 2001;Koopmans 2007). Yet, in the news of smaller countries, there is more room for foreign actors, because of valence criteria. Furthermore, Koopmans and Statham (2010) find that EU-level actors are more prominent in issue debates in which the EU has significant competencies -like monetary policy -than they do in issue debates in which the EU has few competencies, like migration. In other words, the valence criteria fuel an authority-coverage mechanism: the more powers EU institutions have in any particular policy field, the more they feature in the news, because their increased competencies increase their valence levels for journalists.
How does stronger presence as a result of empowerment affect the nature of the debate? A stronger presence of EU actors in the news likely leads to a more pro-European discourse. Yet, the creation of empowered EU institutions may feed a backlash too. Given that empowered EU institutions predominantly advocate pro-European and austerity policies influenced by German ordoliberalism, the backlash likely comes in the form of both anti-EU and pro-stimulus arguments. It is in the nature of public debate that argument provokes counter-argument. Those who either disagree with the substance of the argument or dislike its maker, will be motivated to make a forceful response (De Wilde 2016; Trenz and Eder 2004). Consider this logic the dialectic-debate mechanism that fuels debate and disagreement in the public sphere. It is not just that opponents are provoked to mobilize, journalists are also likely to provide more space for them or even to seek them out actively. Journalists' desire to maintain objectivity and to shed light on what is at stake in any particular policy question often leads them to feature representatives of two or more different points of views in their news. This balanced-reporting mechanism where journalists aim to provide a balanced platform to 'both sides of the story' may lead to a crowding out of pro-European allies of EU institutions. The pro-European arguments are now represented by powerful EU institutions, so journalists have no further need for pro-European national actors in their coverage.
To understand how these three different mechanisms ultimately shape the nature of public debate on EMU and fiscal affairs, we need to take a closer look at the different political issues and conflict dimensions at play in this policy field. First, political conflict in the area of EMU and fiscal policy revolves around the question of whether austerity or stimulus is the right answer to high public debt and stagnation of growth. On the one hand, advocates of austerity argue that states need to make cuts to public expenditure to lower deficits and public debt. Stimulus advocates, on the other hand, argue that economies need to 'grow' out of their debts. Rather than cutting budgets, states should spend more money to stimulate economic activity and accomplish growth. Second, the question is who should be in charge of making the decision on austerity versus stimulus. Should this be up to member state governments individually or should there be a common European response, orchestrated by EU institutions? Since the Eurozone has a united monetary policy, many economists have argued that the EU also needs strong coordination of fiscal policy (e.g. De Grauwe 2010). High national public debt could otherwise create negative externalities for other EMU members. Anticipating this, EMU was accompanied from the beginning by the Growth and Stability Pact, setting strict limits to public debt and deficit. This makes austerity a more or less constitutionalized policy within the Eurozone. The questions of stimulus vs austerity and of more or less Europe thus become partially intertwined, as creditor states argue that debtor states needed to implement austerity because of their European obligations under the rules structuring EMU.
At the very beginning of the 2008 financial crisis, the positions of the governments of advanced economies could be summarized as 'we are all Keynesians now'. Although the causes of the crisis were not completely clear -at least in the public debate -there was a strong consensus among decision-makers that stabilization policies with substantive injection of public money into the economy needed to be implemented to respond to the financial crisis (European Central Bank 2012). After this initial stage, positions became more diverse. Several factors started to be identified as potential causes of the crisis in the public debate and different policies were proposed as a response. These different approaches can be grouped into two main narratives: one based on austerity and another one based on solidarity and EU integration.
The austerity narrative divided EU member states into saints and sinners (Matthijs and McNamara 2015). The main argument was that mismanagement of public finances by southern European countries caused the crisis. Consequently, it was necessary to implement austerity measures, carry out structural reforms and impose strict rules and procedures on fiscal policy in these countries. This narrative was heavily influenced by German ordoliberalism and emphasized the need for fiscal consolidation, rules, and conditionality.
The solidarity narrative, in contrast, draws inspiration from the Keynesian or expansionist philosophy. It placed more emphasis on international factors and financial deregulation as the main causes of the crisis. Consequently, this narrative advocated more integration at EU level, more solidarity between countries and less stringent conditions. This narrative was less successful in creating a common discourse and rallying powerful actors to its cause. Thus, some scholars have argued that one of the reasons why the austerity narrative became dominant is that it succeeded better at organizing its arguments in a coherent and credible way (Helgadóttir 2016).
Unsurprisingly, these two narratives were not equally popular in all member states. Drewski (2015) analyzes newspaper editorials from Germany and Spain for the period 2010-2012 and finds that German media were more prone to emphasize fiscal discipline, free-riders` problems, and moral hazard issues, while the Spanish media tended to focus on systematic risks and advocated bailouts. The dangers and negative effects of financial speculation were a key element in the public debate of some southern European countries. Sarkozy, who famously called to reinvent capitalism in the G20 meeting of 2008, argued that 'Self-regulation as a way of solving all problems is finished. Laissez-faire is finished. The all-powerful market that always knows best is finished' (Sarkozy in Quaglia 2013, 70). During the negotiations of some of the EMU reforms, the role of speculation and the need to regulate financial markets were a significant part of the justification of the French position (Crespy and Schmidt 2014).
The austerity paradigm and the ordoliberal perspective ended up becoming dominant among decision-makers and dominated the public debate for a long time. It is worth noticing that, although this narrative identified some problems at EU level and accepted the idea that EU economic governance needed to be reformed, it emphasized that the main causes of the crisis lay at the national level. Hence, the solutions also needed to come from reforms implemented by the national governments. Many of the reforms proposed at the EU level were destined to increase the level of control of EU institutions over member states and to reinforce their ability to monitor and condition them, as well as to strengthen the existing fiscal policy framework. EU institutions created during the Euro crisis, such as the EFSF, the ESM, the Six Pack or the Two Pack, testify to the dominance of this pro-European, ordoliberal austerity paradigm.
Solutions and reforms implemented during the Euro crisis were thus heavily influenced by the German position (Matthijs and Blyth 2015). We can find several examples of how France lost the battle for more integration and solidarity and ended up negotiating or accepting more ordoliberal solutions (Crespy and Schmidt 2014;Gocaj and Meunier 2013;Schild 2020). In most of the cases, the compromise can be summarized as the northern countries agreeing to some (limited) form of solidarity and assistance, although heavily conditioned and in exchange for tighter reforms. In short, the austerity narrative was dominant and the solidarity narrative played a subordinate role.
However, the existence of a dominant narrative does not mean that alternative explanations were not present in public debates. First, alternative narratives still exist and may gain the support of some actors. Second, because even within the dominant narrative there can be some nuances in the way it is framed. Heft (2017) studies statements on the beginnings of the Euro crisis around 2009-2010 in the German and Spanish quality press. She finds both countries feature a combination of Europeanized and national debates. While there are unique national actors and frames in both countries, the debates in both countries also refer to the same contextual crisis and its dynamics and are not as polarized as one might expect based on the common image that the Euro crisis featured a conflict between Northern creditor member states in favor of austerity and Southern debtor states in favor of redistribution and investment. Yet, Joris, Puustinen, and d'Haenens (2018) show that war frames -in which the Euro crisis was discussed by journalists in terms of battles and zero-sum games -were most frequent in The Netherlands, Spain, and Belgium, while positive-sum game construction was most frequent in Germany and Finland. Bijsmans (2021) shows how much of the debate on the Euro crisis in Austria, The Netherlands and Ireland revolved around policy questions, rather than Eurosceptic or federalist demands about the EU polity. Visions of alternatives to the current EMU setup and the Growth and Stability Pact governing it dominated the quality press in those countries. Kaiser and Kleinen-von Königslöw (2017) examined the framing of the crisis in Germany and Spain during the period 2010-2014. In Germany, the crisis was largely presented as a sovereign debt crisis, focusing on the dangers posed to the German public sector due to the exposure to the financial problems of other EU members. In Spain, the crisis was at this stage largely presented as a banking or financialsector consequence. However, together, both cases point towards a Europeanization of the framing around the idea of 'conditional assistance' over time.
Ultimately, the austerity narrative lost its hegemony, it started to be contested, and alternative discourses gained support from various actors. The so-called Brussels-Frankfurt consensus (Woodruff 2016) started to fade way towards the end of the Euro crisis in 2012. We can start to see a change of heart in the policy response of the EU around this time. For instance, the ECB started to move away from the austerity and national responsibility approach with Draghi's famous 'whatever it takes' speech in 2012 and continued to do so with the implementation of quantitative easing.
Summarizing, the build-up of new EU institutions during the 2010-2012 Euro crisis to deal with the present and future fiscal crises may have several effects on how public debates after the Euro crisis about EMU unfold. First, in light of the existing literature on public debates about the Eurozone, it is likely that we find pro-European and anti-European, pro-austerity and pro-stimulus arguments in all member states and time periods. It is the balance between them that might shift, rather than that one argument can be expected to be completely dominant. This shift in balance may come to the fore in two ways. First, the creation and empowerment of EU institutions during the Euro crisis likely leads journalists to focus more on EU actors during post-crisis fiscal debates within EMU. Journalists' application of the valence criteria fuels this authority-coverage mechanism. Second, this increased presence of EU actors in the news likely functions as a red flag for opponents, who become more vocal in opposition. This dialectic-debate mechanism likely brings in more Eurosceptic voices. In addition, journalists trying to balance the news and provide a platform for 'both sides of the story' may amplify such Eurosceptic voices and dampen other pro-European voices through the balanced-reporting mechanism. How these three mechanisms balance each other out and what their net effect is on the overall tone of the debate, both on the austerity vs stimulus and on the pro-European vs Eurosceptic dimensions, is what this article tries to find out.

Data and method
In order to analyze the extent to which the empowerment of EU institutions to address the Euro crisis have affected the way EMU is debated in public discourse after 2012, this study focuses on three different episodes of contention (cf. Tilly and Tarrow 2007) in news media and conducts claims analysis (Koopmans and Statham 1999) to identify actors and their arguments.

Episodes of contention
As benchmark, we analyze the height of the Euro crisis: from 1 October 2011 to 31 December 2012. In this period, the Euro crisis escalated and several new EU institutions to overcome it were introduced. Most notably, in the summer of 2012, ECB President Mario Draghi made a now famous speech in which he promised the ECB would do 'whatever it takes' to protect the Euro, after which bond spreads started to narrow and the specter of immanent collapse of the Eurozone dissipated. Since so much has already been written on debates during the Euro crisis, as discussed in the theory section above, we will not dwell on it much further here.
The second episode we study is the Greek bailout crisis of 1 June 2015 to 31 July 2015. In this period, the Greek Syriza government led by Alexis Tsipras organized a referendum on whether Greece should accept the bailout package presented to Greece by the socalled Troika: the European Commission, the International Monetary Fund, and the ECB. The resulting 'no' vote subsequently startled markets and raised the prospect of a Greek default on its debt. Unlike in the previous crisis, however, the rest of the Eurozone appeared much better prepared to weather such a storm and the chance of contagionleading other Eurozone member states to default -was generally considered much lower. In a way, the Eurozone had inoculated itself against the threat of a Greek sovereign debt crisis by a combination of new EU institutions, policies to combat crises and a reshuffling of debt portfolios among major banks. The result was a much more national as opposed to European crisis. In the end, the Tsipras government had to fold to the pressure exerted by creditors and accepted a bailout package that was very similar to the one that had been voted down in the referendum.
The third episode of contention is the Italian budget crisis of 2018. In September 2018, the Italian government presented a government proposal to its European partners containing considerable spending plans. As the Italian debt-to-GDP ratio was very high at 138%, this proposal was in breach of European agreements documented in the Fiscal Compact. Italian Prime Minister Conte was pressured by the Lega and MS5 coalition partners. Lega leader Salvini especially played a Eurosceptic role, pushing the Italian government to ignore its European partners. The Italian budget crisis was in principle a government crisis, with Eurosceptic parties Lega and MS5 in conflict with pro-European Italian politicians, such as Conte, President Mattarella, and interim Prime Minister Cottarelli. The internal Italian conflict created strong uncertainties among European partners and investors about whether the Italian economy and government budget were sustainable, raising the specter of default.
Delimiting these three episodes of contention with clear starting and end dates is in several ways an artificial enterprise. All three crises are rooted in structural weaknesses and tensions built into the EMU and in the national fiscal situation of several of its member states. They have been built up over decades and will continue to influence the economy and politics for decades to come. In effect, they do not have a very clear start or end date. Second, the three episodes are not fully independent of each other. The Euro crisis was kick started by the revelation that Greek sovereign debt was much higher than officially reported in 2009 and bailouts happened in several stages. The sustainability of both the Italian and Greek government debt situation was already an issue when the Eurozone was created in the late 1990s. The fact that we pick full calendar months to facilitate the empirical analysis further adds to the artificiality of the studied episodes of contention. Given the longevity of the issues and problems inside EMU and the connection between the Greek, Italian, and overall Eurozone challenges, it is highly unlikely that the three episodes of contention feature completely different debates. At the same time, different locations, different players and different institutional settings are unlikely to generate exactly similar debates. Whether or not the similarities outweigh the differences thus becomes a question of interpretation. They will be discussed accordingly in the findings of this study. These caveats notwithstanding, a clear demarcation of episodes of contention is necessary in order to compare the nature of public debates over time and facilitate an empirical analysis of how and to what extent the empowerment of EU institutions during the Euro crisis altered the way EMU was debated in mass media afterwards.

Countries and newspapers
This article presents debates about EMU-related issues in six EU member states: Germany, France, Italy, Spain, Denmark, and Poland. These countries contain the biggest economies in the Eurozone, and thus the most important and influential EMU countries. Germany is the leading creditor nation, with the biggest contribution to the bailout funds. Its government and dominant public discourse emphasized the need for austerity and enforcement of the Growth and Stability Pact during the Euro crisis. Spain and Italy are debtor countries with high levels of government debt and budget deficits, thus particularly vulnerable to market volatility and speculation. France stands somewhat in between the debtor and creditor countries, though clearly more a creditor than a debtor, particularly since French banks were heavily invested in Greek public debt. Besides these four biggest EMU economies, we also include two outsiders: Denmark and Poland. Neither one of these countries were in the Eurozone during the period of investigation. Being less effected by EMU policies, the debates in these countries are likely to be less contentious than those within the Eurozone. Collectively, the debates in these six countries should provide a varied perspective on how EMU policies were contested.
In each of these six countries, we sample articles from two newspapers, one more conservative and one more liberal, using the Factiva database and key word search strings. Newspapers sampled in the analysis are Frankfurter Allgemeine Zeitung and Süddeutsche Zeitung in Germany, Figaro and Le Monde in France, Corriera della Sera and La Republica in Italy, ABC and El Pais in Spain, Rzeczpospolita and Gazeta Wyborcza in Poland and Jyllands-Posten and Politiken in Denmark. Through this sampling, we focus on newspapers of record. The reporting of news in such newspapers is generally considered to be leading and influential for other news media, both online and offline, as documented in earlier studies on the Europeanization of public spheres (De Wilde, Michailidou, and Trenz 2013;Koopmans and Statham 2010;Trenz 2004;Van de Steeg 2006). We sampled a total of 110 newspaper articles per country, equally divided over the three episodes of contention.

Claims analysis
Conducting claims analysis (Koopmans and Statham 1999), we focus first and foremost on who is making claims, coding whether contributions to public debate are made by state actors, such as Prime Ministers, Members of Parliament, Finance Ministers or agency representatives; representatives of EU institutions, such as the Commission or ECB President; or societal actors, such as sectoral interest representatives, experts, or ordinary citizens. We also analyze nationality and party affiliation of these actors. The composition of actors featured in the news provides a first indication of the extent of similarity and difference across episodes of contention. If we find a higher share of claims made by EU actors in the two later episodes of contention than in the first, this would be interpreted as support for the authority-coverage mechanism.
Second, we focus on the demands voiced in the claims. Claims are by definition an expression of political opinion. Mere statements of fact, such as on the nature of debt or bonds spread would not qualify as a claim and thus would not be coded in our analysis. We coded the policy direction of demands and evaluations in two dimensions. First, we code whether the expressed demand or evaluation reveals a preference for stronger engagement of EU institutions in EMU affairs vs weaker engagement or renationalization of competencies to the national level. We code integration demands if the claim reveals a preference for more EU-level competencies, engagement, or a defense of or positive evaluation of existing EU institutions, competencies, or policies and demarcation demands for the opposite (cf. Kriesi et al. 2008). The second dimension captures the desired level of investment. On the one hand, demands for austerity reflect the ordoliberal belief that the solution to the sovereign debt crisis lies in cutting expenditure in national budgets. On the other hand, demands for investment reflect the Keynesian belief that the right path out of economic crisis lies in stimulus packages and active state-led growth. These two dimensions are not necessarily unrelated, given that EU policy and institutionsparticularly the Fiscal Compact -in a way constitutionalize austerity as policy. A critique of austerity thus easily leads to criticism of the Fiscal Compact, and thus of established EU competencies and institutions. The lines between policy contestation and polity contestation are often blurry in the EU context, as several major policies have been written into the Treaties or have otherwise been institutionalized into EU institutions. To safeguard intercoder reliability, we gave preference in coding to the integration-demarcation dimension over the stimulus-austerity dimension. Statements in which both austerity and EU institutions were clearly criticized would be coded as two separate claims by the same claimant where ambiguous claims in which it is not clear whether the claimant objects to policy or polity features are coded as demands for demarcation. Percentage agreement in intercoder reliability results are at .989 for claimants and .853 for position. These are all above the standard .700 threshold (Lombard, Snyder-Duch, and Bracken 2002, 593). The full dataset, codebook and intercoder reliability test results are available online De Wilde 2021, 2022).

Findings
Did the empowerment of EU institutions during the Euro crisis change the composition of claimants in the news, with a higher share of EU actors in the news afterwards? Figure 1 below shows evidence that this is indeed the case. The share of claims made by EU actors increased from 13,54% during the height of the Euro crisis, to 17,17% during the Greek Bailout referendum, to 30,06% during the Italian budget crisis. This comes at a small cost to the presence of international actors, as the IMF became less relevant in solving European fiscal crises, but mostly at the cost of national actors. The share of claims made by societal actors like experts and civil society representatives remained roughly the same. We here see the authority-coverage mechanism in effect, leading to a stronger EU voice in the news after authority has been transferred to the supranational level.
How did the empowerment of EU institutions affect the discourse, in terms of support for European integration and more transfer of sovereignty to EU level? Figure 2 below shows the mean position on the integration-demarcation dimension. Claims advocating a stronger European role in fiscal policy, for example, through bail-outs, an increased EU budget, common lending tools or for more EU scrutiny of national budgets would load here on the integration side whilst demands criticizing the above, calling for leeway to exempt national budgets from EU rules or even calls to leave the Eurozone altogether would load on the demarcation side. We see here that the increased prominence of EU actors in the news did not lead to a more pro-European discourse. Quite the contrary, the tone of the debate becomes more critical of integration. Rather than adding more pro-European voices to the debates, the rising dominance of EU actors in the news seems to co-occur with a crowding out of national pro-European voices. This is an indication that the dialectic-debate mechanism and balanced-reporting mechanism outweigh the effects of the authority-coverage mechanism. This is further supported by data presented in Table 1 below, which shows mean positions on the integration-demarcation axis of EU, national and societal actors. EU actors remain predictably pro-European, whilst the voice of national actors -and to a lesser extent societal actors -become more critical of EU solutions and existing EU institutions and mechanisms. This is most likely the effect of pro-European national actors  getting less coverage in the news rather than that the same actors became less prointegration over time.
The increase in criticism about EU solutions was not accompanied by an increase of criticism in austerity. The discourse on EMU and fiscal politics in the Euro zone has not become more favorable of stimulus policy since the height of the Euro crisis. Where the Euro crisis featured the discursive conflict between austerity and stimulus narratives described above, the following debates in the context of the Greek bailout referendum in 2015 and the Italian budget crisis of 2018 were very much dominated by the austerity narrative, as shown in Figure 3.
In contrast to arguments about European integration, EU actors did change their tone on the desirability of austerity after the height of the Euro crisis. Table 2 shows that they became much more supportive of austerity, where they used to be more balanced. A similar move was made by societal actors, whilst national state actors remained overall mildly critical of austerity on balance. More detailed analyses of country-by-country variation in these trends is included in the online appendix. This shows the decreasing pro-European tone in discourse is mostly shaped by the German, Italian, and Danish debates, while the French, Spanish, and Polish debates remain more constant. Increasing  demand for austerity is notable in Germany, France, Spain, and Poland, while stable in Denmark and Italy. Since these variations come with substantial levels of uncertainty at country level and do not appear to follow theoretically meaningful patterns based on member states' involvement in EMU, this article focuses on overall trends and patterns rather than country differences. In short, the empowerment of EU institutions during the Euro crisis is connected to an increase in the share of claims made by EU actors in later fiscal crises. This increased presence did not make the discourse more pro-European, however, as the increased share of claims by EU actors with continued pro-European arguments was more than offset by more Eurosceptic national and societal actors. Yet, it did come with a stronger emphasis on austerity. Both EU actors and societal actors championed austerity more strongly during the Greek bailout and the Italian budget crises than during the height of the Euro crisis.

Conclusion
This article investigates how the empowerment of EU institutions affects the nature of media debates about EU issues. It does this in the case of Economic and Monetary Union, by studying newspaper debates on the 2015 Greek bailouts and the 2018 Italian budget crisis against the background of debates during the height of the Euro crisis, in national quality newspapers in Germany, France, Italy, Spain, Poland, and Denmark. During the Euro crisis, the EU was equipped with a range of new institutions that strengthened its capacity to deal with fiscal crises. These include notably the European Stability Mechanism, European Semester, Six Packs, Two packs and the European Central Bank's defacto accepted policy of quantitative easing. Whether these new institutions changed the nature of debate in subsequent episodes of contention can shed great light on the effect of authority transfer on media coverage. In light of previous research that showed that journalists and editors are quite capable of identifying powerful actors and institutions in European politics, we may expect EU actors to play a more prominent role in public debates during the Greek bailout and Italian budget crises than they did during the Euro crisis. It remains an open question how such an increased presence of EU actors would affect the nature of the debate, because a greater presence of pro-European EU actors in the news may simultaneously feed a crowding out of national pro-European actors and a more forceful anti-European and pro-stimulus backlash.
This study shows that -as expected -EU actors are much more prominent in the news during the Greek bailout and Italian budget crises than they were during the height of the Euro crisis. The share of claims made in national newspapers by EU actors more than doubles from around 13,5% during the Euro crisis to a little over 30% during the Italian budget crisis. Yet, the increasing prominence of EU actors in the debate does not make the overall discourse more pro-European. Quite the contrary. While EU actors remain consistently very pro-European in the solutions they advocate to fiscal crises, the contribution by state actors at national level -including representatives from government, parliament, and central banks -becomes increasingly critical of European solutions. We see the same among societal actors like trade union representatives, opinion makers, and economists. This is likely caused by a stronger presence of already Eurosceptic voices and a crowding out of pro-European actors, rather than a change of heart by the same national and societal actors. The net effect is a discourse more critical of EU solutions, despite a considerably stronger voice of pro-European EU-level actors in the debate. In other words, while EU actors gained more power in the course of the Euro crisis and this power translated into stronger media presence, they at the same time lost influence over the overall tone of media coverage.
This observation cannot be explained by a change in public opinion. After the Euro crisis and again after the refugee crisis, public opinion on European integration features a 'rebound' (De Wilde 2021). Had public opinion been the driver of media discourse, it would have likely become more pro-European. Rather, what we see at play here are three key mechanisms driven by media logic. The authority-coverage mechanism leads journalists applying valence criteria to provide more space in the news for more powerful actors. As authority is transferred to EU institutions, they become more prominent in the news. At the same time, the dialectic-debate mechanism triggers opponents to EU actors and their pro-European arguments, mobilizing them to make stronger counter arguments the more EU actors are in the news. Finally, the balanced-reporting mechanism leads journalists to leave pro-European national actors out of the news, because their arguments are increasingly represented by EU actors. The latter two mechanisms clearly outweigh the first in terms of their net effect on media discourse.
The same logic does not apply to substantive policy debates. There is more emphasis in media discourse on the need for austerity to solve budgetary crises during the Greek bailout and Italian budget crises than during the Euro crisis. Conventional knowledge has it that the debate during the Euro crisis was strongly influenced by German ordoliberal thought and that Keynesian critique of austerity swelled afterwards. This study shows no evidence of this. In contrast, public discourse in the mass media of EU member states became increasing supportive of austerity from the Euro crisis to the Greek bailout and Italian budget crises. This development is driven by both EU and societal actors, which demanded more austerity in the latter crises.
While this study contains an elaborate and in-depth analysis of debates in the news around major controversies in Economic and Monetary Union, it comes with some important limitations. First, claims analysis is a highly labour-intensive method, which means the analysis is restricted to debates in six member states around three episodes of contention. The inclusion of crises that were more financial rather than fiscal in nature, like the 2013 Cyprus crisis, might have generated different results. The inclusion of debates in more member states might in addition bring more variation in the debates. Particularly, the debates in Greece and other smaller Eurozone member states could bring added perspective. Yet, national differences among the six countries studied here did not amount to a need to qualify the general findings.
Michael Geary, Carine Germond, and Nadja Kratow for useful feedback on earlier versions. This article reflects only the author's view. The European Commission Research Executive Agency is not responsible for any use that may be made of the information it contains.

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

Funding
This work was supported by the European Union's Horizon 2020 research and innovation programme under grant agreement No 770142.