Between a Rock and a Hard Place: Preference Formation in France During the Eurozone Crisis

Taking Other States into Account: New Intergovernmentalism and Nested Games

The management of the Eurozone crisis has been considered as an intergovernmental process, excluding de facto the European Parliament and the Commission, but empowering the European Central Bank (ECB). The weakness or even absence of the community method in the management of the euro crisis left member states with unanimity decision-making rules. New intergovernmentalism stresses that in this context, consensus-seeking and iterative negotiations have become the guiding norms in EU intergovernmental politics (Bickerton et al., 2015; Puetter, 2012, 2014). In this context, preferences of other member states are scrutinised in advance by their partners and influence domestic preference formation since their inception, and not only during the bargaining phase as argued by the LI framework.

This understanding of preference formation not only as a vertical channel (from interest groups to the government) but also as a horizontal channel is also consistent with Tsebelis’ (1990) nested games. Nested games are considered to be a way of transplanting context into game theory. Instead of assuming that actors play games in a vacuum, it shows that these games are embedded in some higher order network. In Tsebelis’ approach, this higher order game determines the payoffs of the players: deviation from the optimal outcome of the negotiations on a given policy subject might be compensated by gains on another topic. In fact, the iteration of nested games was crucial during the creation of the single currency as ‘it opened opportunities for political players to redefine what constituted appropriate strategy and optimal outcomes in EMU negotiations’ (Dyson and Featherstone, 1999: 46–47).

According to new intergovernmentalism, the institutional environment of the European Council and its preparatory committees align the preferences of member state administrations, including senior bureaucrats and political decision-makers, within a transnational policy-making environment. Preference formation, as a process which involves governmental elites, thus, has become transnational. An example of such transnationalisation of preference formation is the Franco-German couple. In the context of EMU reforms, the Franco-German couple has indeed not only strongly influenced the intergovernmental bargaining process as Degner and Leuffen (2017) have underlined, but also domestic preference formation. France and Germany forge their preferences through bilateral contacts between governments and the subordinated bureaucracies (Degner and Leuffen 2017). Although German and French political economy and economic cultures diverge (Clift, 2013; Clift and Woll, 2012), their pre-negotiation exchanges lead to a rapprochement on a large number of points.

Feedback Loops: EU Decision-Making Under Domestic Public Scrutiny

As pointed out in the ‘Introduction’ to this special issue, policy approaches allow us to model the circular relationship between preference formation, decision-making and, again, preference formation. Decision-making takes place in complex environments (Jones, 2001; Simon, 1985, 1997). The multilevel character of the EU adds further complexity to this understanding of preference formation: preference formation is a more complex process than the result of a series of linear ‘push factors’. It is a circular process through which political elites and experts, interest groups and social movements as well as the media frame a problem and influence state actors in defining national preferences. Decisions taken on the basis of these preferences at the European level then influence again domestic actors who adjust and adapt their positions, and ultimately transform national polities.

According to the public policy literature, feedback loops understood here as the interplay between the domestic public opinion and decision-makers (Campbell 2004; Pierson 1993), should play an increased role in times of economic and financial crises since the salience of economic issues should increase. In other words, when the salience of an issue, understood as the perception of the importance of the issue for the public, increases, even complex and very technical issues become more resistance-prone (Wlezien, 2005). Indeed, a publicly perceived threat and a high degree of uncertainty increase the salience of an issue and might lead to an increased probability for actors to oppose governmental decisions. In turn, a higher degree of politicisation, understood as the process of making citizens and concerned actors more involved in politics, is often perceived as having influenced policy positions of member states, such as France, which searched to renegotiate policies at the EU level (Egan et al., 2010; Hooghe and Marks, 2009). Domestic opposition is taken up by national representatives and translated again at the EU level: member states reopen debates on specific policy instruments, insisting on the difficulty of implementing them at the domestic level and ask for a reinterpretation of these issues.

Reducing Uncertainty: Policy Networks and the State-Finance Nexus

While the new intergovernementalism approach and the public policy literature postulate that feedback loops increase politicisation, deliberation and the number of players involved within policy-making processes in times of crisis, another stream of literature focused on crises shows that opposite dynamics are at play. In times of crisis, uncertainty increases and standard operating procedures can rarely be applied. Crises trigger a perception of urgency, which usually leads to a decrease in the number of actors involved in finding solutions to the problem (Boin et al., 2005; Degner, 2018; Saurugger and Terpan, 2016). In fact, these unexpected critical junctures reveal the balance of power within political systems, since it is only the most powerful players who are participating in emergency meetings (Dobry, 1992). How to reconcile the contradictory crisis-induced effects observed by the public policy and the political economy literature? Braun (2015) offers a solution to this conundrum by distinguishing two distinct public policy phases: a phase of emergency crisis management and a subsequent institution-building phase. In an emergency, decision-making is usually restricted to a small number of actors, whereas in the follow-up process leading to policy change, these focusing events (Birkland, 1997; Kingdon, 1984) often lead to a broadening of the policy community, including the media and a large number of groups.

In a context of emergency, tight and small policy networks between civil servants and private banks play a preeminent role to reduce the high level of uncertainty caused by financial markets destabilisation in a short period of time (Fontan, 2013; Woll, 2014). To grasp why these networks are so powerful, we rely on the political economy literature on the ‘state-finance nexus’, which explains how the financialisation of French economy led to an intertwining of states’ and banks’ interests. Since the 1970s, political leaders have been increasingly governing through financial markets, that is, they relied on financial markets and instruments as a form of statecraft to achieve economic goals, which they could not attain anymore with fiscal and budgetary policies (Braun et al., 2018; Epstein, 2017: 104).

With the support of the State, French banks’ balance sheets expanded up to the point that they became ‘too-big to bail’: in 2008, the size of the banking system was equivalent to 350% of the French GDP (gross domestic product; Blyth, 2013). Moreover, since these banks became the main dealers of the French sovereign debt, the French state has become dependent on them to refinance its debt at low cost (Lemoine, 2016). Finally, since the creation of the Eurozone, the links between French banks and the balance sheets of peripheral Eurozone member states dramatically increased following a wave of transnational mergers and acquisitions and the purchase and recycling of peripheral sovereign debt on repo markets (Gabor and Ban, 2016). In other words, although French banks were indecisive about their interests during the creation of the Eurozone (Grossman, 2004), they quickly learned to recognise the opportunities offered by this new context and they engaged in a cross-border expansion.

We mobilise this political economy literature on the state-finance nexus to have a more fine-grained view on the formation of domestic preferences. While the LI framework distinguishes between states’ and banks’ interests, this political economy approach underlines that the financialisation of our economies led to an entanglement of their interests. Hence, the policy networks located at the heart of the state-finance nexus play a preeminent role in the crisis resolution process, because they have an insider knowledge of states’ and banks’ respective exposures and they are used to form coalitions to defend their intertwined interests. In other words, these policy networks were the main promoters and defenders of French ‘economic patriotism’ during the Eurozone crisis (Clift and Woll, 2012).

Policy Networks and Uncertainty

Since at least the 1980s, networks of high-level civil servants and private bankers defend and promote French interests within the EMU policy-making process (Bezes and Le Lidec, 2015: 502; Cerny, 1989; Clift, 2012; Dyson and Featherstone, 1999; Hardie and Howarth, 2009; Jabko and Massoc, 2012). These networks unite key sectors of the French administration in charge of European Economic governance (the head of the Treasury and the economic advisers of the French presidency) and the chairpersons of the six largest banks (Crédit Agricole, BNP Paribas, Société Générale, Crédit Mutuel, Caisse d’épargne and Banque Populaire). Close to this small core of civil servants and private bankers, the governor of the Banque de France and the Finance minister are also involved in the formation of French preferences, albeit on secondary and non-urgent issues.

These networks are characterised by a strong homogeneity, which is caused by the similar social backgrounds of its members and revolving doors between high-level French administration and private banks. This homogeneity was a powerful strategic asset for the French governments when it had to provide financial support to its banks after the collapse of Lehman Brothers. Compared with other national arrangements, the French bail-out was costless for the taxpayer ⁶ because of the swift agreement between the State and its banks (Woll, 2014). Unsurprisingly, the homogeneity of French policy networks allowed to bridge potentially different interests between the different actors involved in the crisis management process (Jabko and Massoc, 2012). In fact, scholars argue that the French economic governance is post-dirigist because these networks align the interests of private banks and the state (Clift, 2012; Clift and Woll, 2012). Hence, French economic policy-making is characterised by a close cooperation between banks and the State, a situation that fits well within the liberal intergovernmentalist approach because their interests are aligned and not confrontational (Grossman and Woll, 2014).

While the academic literature has studied these networks in the early stages of the subprime crisis (2007–2010), it has not yet thoroughly scrutinised their role in later stages of the crisis, when the locus of policy-making moved from the national stage to the EMU circles (2010–2017). Our interviews show that these small French policy networks have been crucial in reducing the high level of financial uncertainty caused by the crisis. Two players were central in the formation of French preferences during the negotiations on the first Greek bail-out (February–June 2010): the top civil servants at the Treasury and the cabinet of the French presidency. More precisely, these players were at the heart of policy networks at both national and supranational levels.

First, information sharing within French policy networks helped civil servants to determine precisely the extent of banks’ exposure to Greek sovereign debt (and later on, on the sovereign debt of other peripheral countries) and, hence, to design policy solutions accordingly. More precisely, French civil servants made a deal with French banks – based on the Vienna initiative template ⁷ – as early as in February 2010: banks committed to keep Greek debt on their balance sheets in order to stabilise its value, while the French government ensured that Greek debt would not be restructured and that it would provide liquidity to the Greek government in order to ensure the payment of coupons to its creditors (Interview 2). Indeed, in late 2009, French banks and insurance groups held 26% of the Greek severing debt purchased by foreign investors, well above German financial institutions (15%). ⁸ If the Greek debt would have been restructured, BNP Paribas, the largest French bank whose assets are worth 120% of the French GDP, should have been recapitalised (Interview 3). As Figure 1 shows, the evolution of French banks’ exposure to the Greek sovereign debt is in line with this bargaining: French banks held their level of exposure to the Greek debt from the first half of 2010 until they started to fire-sale Greek bonds just before their restructuring in mid-2012.

OPEN IN VIEWER

Figure 1.

Exposure of French Banks to Greek Debt (in €).

In parallel, the Presidency’s cabinet played a key role in persuading Nicolas Sarkozy that a common rescue fund was compulsory to stabilise the French banking system and alleviate systemic risks within the Eurozone (Interview 2). In short, the first Greek rescue was aligned with the French preferences (financial assistance without debt restructuring). Insider reports underline that French officials were at the forefront of the resistance against the option of debt restructuring because of French bank exposure during the Spring 2010 meetings involving agents from the International Monetary Fund (IMF), the Commission, the ECB and European member state governments (Blustein, 2015: 9–11). ⁹

Second, members of the Treasury sit at the most important meetings in the Eurozone governance: the Eurogroup and, even more importantly, its preparatory councils: the EFC and the Eurogroup Working Group subcommittee. For example, the European Financial Stability Fund, the predecessor of the ESM, was set up behind the closed doors of a Eurogroup working group meeting during the weekend of 9 May 2010 (Fontan, 2013). Treasury members relied on their participation to these meetings to gather information about the crisis resolution process before other French civil servants. This ensures a strong leverage in favour of the Treasury when interministerial struggles occur. For example, the Treasury and the Foreign Office hold opposite views on the future of EU integration. On the one hand, the Treasury supports an integrationist perspective because the topics on which it is competent are already negotiated at the EU level. From this perspective, more transfer of competences at the EU level gives more power to the Treasury, which is in favour of a new EU treaty (Interview 1). On the other hand, the Foreign Office is more focused on the defence of French sovereignty and is opposed to a new Treaty. This was especially true when Laurent Fabius, a proponent of the 2005 ‘No’ campaign was the Foreign Affairs minister (2012–2014). This asymmetry of power might explain why France has been advocating for further integration of the Eurozone during the crisis (Schimmelfennig, 2015).

In short, the French Eurozone governance system is strongly dominated by small policy networks of civil servants and private bankers. At the administrative level, the policy positions of Treasury and the Presidency are homogeneous because of the strong personal links between their members, which are fostered by similar education and career paths (Interview 4). In the case of the first Greek rescue, the positions of these small policy circles were unambiguously influenced by a strong degree of economic patriotism: the priority of the French government was to protect its banking system and avoid a second recapitalisation. In other words, the exposure of French banks to peripheral sovereign debt heavily influenced national preferences. More generally, they are supporting further integration of the Eurozone, which follows a historical French stance on EU economic integration. The homogeneity of positions within the French administration stands in strong contrast with Germany, which was riven by administrative division between the minister of Finance and the minister of Economy, which mirrors the political struggles between the members of the governmental coalition (see Degner and Leuffen in this SI).

The Horizontal Dimension of Eurozone Policy-Making

According to the LI framework, the formation of national preferences is independent from international bargaining (see section ‘Accounting for Preferences’). Here, however, our research results strongly undermine this assumption: French preferences were influenced from their inception by German preferences and by the iterative nested games, which characterise Eurozone policy-making.

First, the formation of French preferences is historically intertwined with German preferences. Academics who study the central role played by the Franco-German couple in the Eurozone answer to the financial crisis disagree, however, on the power balance between the two countries and their impact on the Eurozone crisis regulation process. On the one hand, some scholars argue that Germany is a ‘reluctant hegemon’ since its representatives rely on its economic power to play a ‘waiting game’ during the Eurozone crisis negotiations (Bulmer, 2014). From this perspective, France is merely another Eurozone country affected by German strategies without being able to modify them (Clift and Ryner, 2014; Lux, 2018). On the other hand, a small number of researchers claim that Germany’s preferences were strongly undermined by French ones’ (Degner and Leuffen, 2017; Rothacher, 2015; Schild, 2013). According to these scholars, post-crisis Eurozone policies were mainly influenced by the French governments, which managed to ‘invert the rules of the Maastricht Treaty’ (Rothacher, 2015). Between these two opposite poles, a third group of scholars recognise that post-crisis policies stem from a compromise struck between the two largest Eurozone countries whose preferences gradually converged (Crespy and Schmidt, 2014; Van Esch, 2014).

Our research findings show that while French policy space was constrained by German preferences, the outcome of Eurozone negotiations did not contradict French ‘red lines’. When Eurozone countries set up the ESM, French negotiators recognised that they would have preferred a higher capitalisation (which was set at €500 billion), but they did not ask for it since German negotiators would not agree with it. In the words of a Treasury agent: ‘French negotiators internalise German preferences when building their policy positions. We would not propose a policy blueprint that would go against German red lines’ (Interview 3). French negotiators are updated about the evolution of German red lines because they are in constant contact through direct phone lines with their German counterparts sitting at the Eurogroup preparatory committees (Interviews 1, 2 and 4). These open lines of communication between EFC members occur because they meet each other on a very regular basis and they are sharing the willingness to strike policy compromises, which are constrained by the red lines set up in Berlin and Paris. In the words of one member of the French permanent representation, ‘these channels of communication are open in order to find a proper balance between the logic of conditionality and debt mutualisation’ (Interview 4). These research results are consistent with the main tenets of the new intergovernementalist framework, which emphasises that transnational policy-making environment increases consensus-seeking behaviour (cf. section ‘French Preferences During the Eurozone Crisis’). While LI accounts would argue here that the process of preference formation still remains distinct from the intergovernmental bargaining phase, we see here empirically that these two are interlinked.

Second, even though Germany constrains its policy space, France is still one nodal point of Eurozone negotiations. Indeed, Luxembourgish, Dutch and Belgian EFC members underline that the Eurozone policy space is shaped by three groups of players: Germany and other ordoliberal countries (Finland and Netherlands), France and Southern Europe countries (Spain and Italy) and supranational institutions (European Commission and ECB). Other countries usually follow the positions defined within one of the nodal points or try to build bridges between them (Interviews 5–7). This is because smaller Eurozone countries do not have the necessary expertise or the structural power to define their policy positions autonomously for each EMU negotiation. France, like Germany or the European Commission, does.

In addition of having the ability to build its policy position autonomously, France relied on iterative Eurozone nested games to ensure that policy compromises did not deviate too far away from its preferences. For example, French negotiators underlined that the capitalisation of the ESM was less important than its instrumental use, which was enlarged to bank bail-outs following French pressure (Interviews 2 and 3). In other words, French negotiators ‘caved in’ on the capitalisation of the ESM but they put a constant pressure on German negotiators in order to increase the flexibility in the use of its instruments. In a similar vein, France agreed to implement the ‘debt-brake’ enshrined within the TSCG for two reasons. First, French negotiators did not cross their ‘red lines’: the debt-brake was not written down within the Constitution because the government wanted to avoid a constitutional revision. Instead, France built an alliance with Italy to push forward the idea that the implementation of an organic law was a sufficient commitment (Interview 2). Moreover, French negotiators obtained implicit counterparts from their partners. More precisely, the TSCG was part of a Franco-German compromise in which Germany accepted the creation of a Banking Union and France agreed on stricter fiscal rules (Interview 3). The ECB was also part of the compromise according to one EFC member:

The decisive moment was the ECB measures. When the ECB announced its Long –Term Refinancing Operations in late 2011, the implicit counterpart for these measures was the implementation of the TSCG. This is how it has been ‘sold’ to French Members of the Parliament. (Interview 2)

Moreover, even when a policy contrary to French interests is negotiated at the Eurozone level, French negotiators ensure that the final agreement offers enough caveats before accepting it. For example, French negotiators agreed on the RQMV ¹⁰ reform, when they were convinced that the proposal was blurry enough to avoid automatic sanctions and that the European Commission would not have enough political power to start the adequate procedures (Interviews 1 and 2).

In sum, the formation of French domestic preferences is strongly influenced by the dynamics of Eurozone negotiations, which represent a ‘multilevel and iterative nested game’. On the one hand, Germany and France are engaged in an asymmetrical power game in which French negotiators ‘internalise’ German preferences, while the opposite might not be true. On the other hand, France is sufficiently powerful to exploit the policy spaces offered by the Eurozone iterative nested games and it manages to strike policy compromises that do not infringe its red lines. Finally, the open communication lines and the high level of trust between the EFC members help member states to formulate initial policy positions that are sufficiently close to each other for policy compromises to be struck.

Feedback Loops

As already underlined, the French Fifth Republic gives a high level of autonomy to the government and its administrative branches in European negotiations. Moreover, this autonomy was strengthened in 2002 when the French political system became unambiguously presidential. Finally, the temporal pressures induced by financial markets pressures in times of crisis reduce the number of players directly involved in the negotiations (cf. section ‘Accounting for Preferences’).

These factors explain why French negotiators were isolated from the pressures of political parties and other administrative bodies during the crisis management stages. For example, during the weekend preceding the set-up of a €125 billion rescue fund to recapitalise the Spanish banking system in 2012, two members of the Treasury confirmed the financial commitment of the French government by phone to their EFC partners without consulting other domestic players (Interviews 1 and 2). In fact, all our interviewees agree that, outside the Treasury and the Presidency’s office, other administrative and political bodies, including the Parliament, the governing and opposition parties, did not play a role during the crisis management phase (Interviews 1–4). They point out the temporal pressures induced by financial instability and the high level of expertise needed to deal with financial rescue schemes to explain this outcome. In addition to the institutional and contextual factors outlined above, French political history provides other explanatory elements. Since the 2005 referendum on the EU constitutional treaty, the Socialist party dodges European issues because they proved to be divisive (Interview 1). Interestingly, the EMU Choices dataset shows that this high level of power concentration within the hands of a few players during the early crisis management phase is only found in those Eurozone countries that were granted financial loans or were under severe financial pressures, such as Spain, Italy and Greece.

Our research results therefore question the public policy literature on feedback loops and the previous studies on French preferences about financial rescue schemes, which are supposed to shift in 2012 because of increasing popular discontentment. Indeed, Rothacher (2015: 272) relies on five non-longitudinal polls to state that growing popular aversion against the ESM led to a shift in governmental preferences. In a similar vein, Crespy and Schmidt (2014) rely on polls and discursive analysis to point out that French leaders’ discourse was influenced by shifting stances of their constituency on this issue. This divergence of research results may be explained by different methodologies. Whereas we rely on interviews with policy-makers at the highest level to analyse the impact of public opinion on the formation of French preferences, other studies rely on polls (whose significance on complex economic matter should be seriously questioned ¹¹ ) and the analysis of political discourses (which are not necessarily aligned with policy preferences).

The French situation stands in high contrast with other countries, such as Germany and Netherlands, where EFC participants must obtain a negotiation mandate from their parliaments and report on the final agreements in front of them. Sometimes, these auditions were broadcasted on national TV (Interview 5). Counterintuitively, these strong controls gave some leverage to the Dutch and German EFC representatives. Indeed, tying its own hands proved to be a winning strategy for these countries during the Eurozone negotiations since it narrows down the set of policy options in their favour, a strategy in line with Putnam’s two-level games and win set theory (Putnam, 1988). A French EFC member acknowledges that they did not discuss policy options such as Eurobonds in early 2010 because its German counterpart repeated that he did not have the negotiating mandate allowing him to participate in such talks (Interview 2). This can be considered to be the reason why Finance ministers asked the Assemblée Nationale to hold ex-post hearings after the negotiations. This option was, however, declined by the MPs who argue they were lacking time and technical skills (Interview 4).

While the autonomy of French negotiators is the general rule of thumb in crisis emergency times, the salience of policy issues during institution-building phases brings back into play other administrative bodies and political institutions. The ratification of the TSCG is a point in case. During our interviews, French negotiators framed the TSCG as a political issue while the Greek rescue, the set-up of the ESM and the RQMV were considered as technical issues (Interviews 1 and 2). Why was the TSCG the most politicised topic of our four case studies?

First, MPs return in the policy-making process because the TSCG was designed as an intergovernmental treaty, which needs parliamentary approval for ratification. In addition, there was more interministerial coordination because the TSCG impacted more public policies than the set-up of financial rescue schemes. An EFC member underlines that when the TSCG was negotiated, they were in constant liaison with MPs in order to ensure that they would accept to ratify the Treaty (Interview 1). Moreover, the Prime Minister and its administrative body tasked with interministerial coordination for EU issues (Secrétariat Général des Affaires Européeennes) also went back into play. Second, fiscal issues are more salient than other economic policies. ¹² In the words of an EFC member, ‘parliamentary hearings were tough and MPs worked on their issues more intensively than in usual times to corner the government’ (Interview 1). Because the TSCG was more politicised, public opinion, political parties and media put French negotiators under more pressure than with other Eurozone issues. In other EU countries such as Netherlands and Belgium, the TSCG negotiations were also a window of opportunity for the political parties and administrative bodies to increase their participation in the EU negotiations, from which they had been excluded in the earlier stages of the crisis (Interviews 5 and 6).

Because the TSCG was the most politicised topic, we would expect feedback loops to play a more important role in the policy-making process: governmental positions should be affected by the outcomes of public debates and/or partisan dynamics. The campaign of the 2012 French presidential election hinted that this would be the case. Indeed, European issues were at the centre of the two main political parties’ manifestos and speeches (Crespy and Schmidt, 2014; Grossman and Sauger, 2014). More precisely, Francois Hollande pledged that, if he would be elected, he would renegotiate the TSCG in order to put more emphasis on growth issues. Our interviews show, however, that the Treasury played an important role in ‘reminding previous French engagements (made under Sarkozy presidency) to its EU partners’ to the newly elected French President (Interview 1). Despite intense pressures from its own political party, which would eventually lead to the secession of a group of MPs nicknamed Les frondeurs, ¹³ President Hollande caved in on its campaign promise and did not renegotiate the TSCG. This outcome led one of our interviewees to state, ‘On Eurozone issues, the Treasury matters more than the Socialist party’ (Interview 1). In a more nuanced light, while the Treasury and its networks played a role in the continuity of French positions, other factors were at play, such as the refusal of Germany to reopen the negotiation on the TSCG (Interview 3).

Grossman and Sauger (2014) underline that the rapid fall of popular support towards F. Hollande after his election was not different from secular Fifth Republic trends. ¹⁴ Yet, it would be short-sighted to conclude that the U-turn on the TSCG did not impact the Hollande presidency. Indeed, the secession of Les Frondeurs 5 months after the election of Francois Hollande weakened the internal cohesion of the Socialist Party during the whole term. Moreover, like other European social-democracy parties, the implementation of fiscal austerity policies enshrined within the TSCG weakens considerably the ideological cohesiveness of the Socialist party, which traditionally promote Keynesian-inspired economic policies (Escalona et al., 2013). In turn, the lack of internal and ideological cohesiveness was conducive to the historical collapse of the Socialist party in the 2017 presidential elections (Lefebvre, 2017). Beyond partisan consequences, our observations are also consistent with the ‘post-democracy’ theoretical framework, which emphasises that political change between centre-left and centre-right parties does not lead to different economic policies (Crouch, 2004; Glencross, 2018; Streeck and Schäfer, 2013). These authors also claim that this lack of change weakens voter turn-out, which was observed in the 2017 elections (Braconnier et al., 2017).

In sum, French policy networks enjoy a very high level of autonomy when they set up the French policy positions on Eurozone issues, especially during the crisis emergency phase. On matters that are deemed as ‘technical’, they are virtually the only players involved in the negotiations, with the cabinet of the French Presidency. While this high level of autonomy was a strategic asset when bailing out the banking system in the early stages of the financial crisis, the ‘tying hands’ strategy of Dutch and German negotiators proved to be more successful in the Eurozone negotiations.

On more politicised topics, these policy networks are challenged by other administrative bodies and political parties, but they manage to retain control over the build-up of French preferences. Indeed, on the TSCG, which was a highly politicised topic, French preferences remained surprisingly stable because the Treasury managed to circumvent political resistance. In other words, the centrality of autonomous policy networks within the French governance circumvent the feedback loops that should nourish policy-making within liberal democracies, even when topics are highly politicised.