Negotiating Greece. Layering, insulation, and the design of adjustment programs in the Eurozone

ABSTRACT

The paper sets out to explain why the European Central Bank and the European Commission relaxed their opposition to debt restructuring and fiscal accommodation for Greece in the shift from the first to second adjustment program. Using the findings of the empirical analysis, the paper shows that EU institutions’ repositioning cannot easily be ascribed to the mechanisms that are typically at play in international negotiations, namely exogenous pressures and internalization of new beliefs through persuasion. Instead, the paper argues that a more nuanced and complete explanation of the relaxation of opposition to the change in the program strategy requires taking into account the institutional and temporal dimensions of the Troika negotiations. Specifically, the paper shows that the evolution of European actors’ preferences was shaped by the layering of new crisis management rules into the machinery of the monetary union. Layering allowed political actors in favor of the status quo achieving their preferences under changed external circumstances.

KEYWORDS:

• Greece
• IMF
• European Central Bank
• Troika
• sovereign debt crisis

1. INTRODUCTION

In May 2010, Greece became the first Eurozone country under an international financial assistance program and the test case for a newly established institutional arrangement: the Troika, or the bailout ‘Institutions’ as the latest version of Eurospeak has it.¹ The Troika brought together two institutions responsible for economic policy in the European Monetary Union (EMU) – the European Central Bank (ECB) and the European Commission (EC) – and an international institution – the International Monetary Fund (IMF) – with the aim of coordinating financial support for troubled EU economies. In addition to Greece, Troika adjustment programs were developed for Portugal (2011), Ireland (2011), and Cyprus (2013).

Although each of these programs has been not without difficulties,² the Greek program has been particularly problematic as attested by the negotiation of three different programs in five years. These negotiations have often tested the resilience of the monetary union by exposing significant power asymmetries among Eurozone countries (Moschella 2016; Schimmelfennig 2015) as well as repeated political failures in dealing promptly with the crisis and its contagion (Featherstone 2011, Tsebelis 2016). The Greek negotiations have also tested inter-institutional collaboration within the Troika. As of 2011, the IMF started becoming increasingly skeptical about the sustainability of the debt dynamic in Greece.³ Its calls for debt restructuring and for relaxation in the path to fiscal balance, however, met with stiff resistance from the two European Institutions.⁴

This paper undertakes to explain the negotiating dynamics among the three Troika Institutions. In particular, the paper sets out to explain why the ECB and the EC eventually relaxed their staunch opposition to debt restructuring and fiscal accommodation in the shift from the first to the second adjustment program in 2012 (Table 1). Both the ECB and the EC had opposed these measures lest they would have heightened pressures for monetary financing of public debts and set a precedent in favor of fiscal indiscipline. That is to say, the ECB and the EC opposed changing the basic principles that underpin the European Monetary Union (EMU).⁵

Table 1 Troika programs for Greece, 2010 and 2012

	Euro area contribution	IMF contribution	Debt relief	Fiscal policy
1st program	€80 bn	€30 bn	No	Fiscal adjustment of 11% of GDP in cumulative terms from 2010 through 2013
2nd program	€144.7 bn	€28 bn	Combination of PSI and official support	Fiscal adjustment of 5.5% of GDP from 2012 to 2014, with efforts back loaded to 2013–2014 (the primary balance was allowed to run a deficit of 1% of GDP in 2012)

Using the findings of the empirical analysis, the paper shows that European Institutions’ repositioning cannot easily be ascribed to the mechanisms that are typically at play in international negotiations, namely exogenous pressures and internalization of new beliefs through persuasion. On the one hand, as will be discussed at greater length below, there was no neat correspondence between market pressures and policy responses. For instance, the write-down of Greek government bonds held by the private sector was agreed in spite of indicators of markets’ uneasiness at the prospect of debt restructuring. On the other hand, in spite of the new IMF's advocacy for gradual fiscal consolidation since the start of the sovereign debt crisis (Ban 2015), the empirical record does not provide indications that the European Institutions were persuaded by the arguments articulated in Washington.

Building from the insights of the historical institutionalist (HI) tradition, the paper argues that a more nuanced and complete explanation of the relaxation of opposition to the change in the program strategy requires taking into account the institutional and temporal dimensions of the Troika negotiations. Specifically, the paper shows that European actors’ preferences were largely shaped by the institutional mosaic that governs political and economic relations in the Eurozone. This institutional mosaic was transformed between 2010 and 2012 as new rules on crisis management were layered onto existing ones, culminating with the creation of the European Stability Mechanism (ESM). These rules, which link the provision of finance to fiscal discipline, were pivotal for the Greek negotiations because they offered the ECB and the EC the tools to preserve the status quo in the Eurozone in spite of the changes in the political and market environments. In other words, the layering of new rules into the Eurozone machinery was the mechanism through which political actors endeavored to maintain the status quo by insulating it from changed external circumstances.

Advancing this argument, the paper contributes to scholarly works that emphasize the endogenous sources of actors’ preferences at the supranational and international levels (Farrell and Newmann 2010; Fioretos 2011; Posner 2010). This institutional perspective helps filling in the limitations of alternative rationalist and constructivist explanations that are here represented in the arguments focusing on market pressures and IMF persuasion. The paper also expands the literature on regime complexes. Whereas much of this literature is focused on explaining the origins or the consequences that stem from the interaction of multiple institutions that operate in the same issue-area (for instance, Alter and Meunier 2009; Gehring and Oberthür 2009; Keohane and Victor 2011; Raustiala and Victor 2004), the paper starts shedding light on how conflict is managed within the regime itself. Finally, the paper contributes to the research agenda sets out in this special issue by lending support to one of the key propositions of the new interdependence approach (NIA) (Farrell and Newman, this issue). Specifically, the paper showcases that ‘rule overlap’ and layering give some actors a unique opportunity to advance their preferences. In the case under investigation, for instance, the development of new crisis management rules provided the ECB and the EC with the tools to secure their preferences on how the monetary union should work.

The arguments here advanced are illustrated based on the official documents published by the IMF and the EC. Both organizations publish, on their websites, their program reports and supporting documentation containing the Troika diagnosis and associated economic projections and the list of policy measures subscribed to by the Greek government – i.e., the Memorandum of Economic and Financial Policy (MEFP) and the Memorandum of Understanding (MoU). In addition, both organizations make available the content of the regular program reviews. The analysis of primary documents is complemented by interviews with officials involved in the Greek negotiations at the IMF and the EC.⁶ Because ECB officials declined to be interviewed, the analysis of ECB preferences relies exclusively on secondary sources, including financial press reports and analyses.

Before proceeding, some clarifications are in order. To start with, the paper focuses on changes in the program conditions aimed at restoring a sustainable debt dynamic in the short-to-medium term. Long-term measures of debt sustainability – such as structural reforms to boost gross domestic product (GDP) growth – will be only briefly analyzed. Second, the purpose of this paper is neither to provide an effectiveness analysis of the Troika programs nor to suggest that the changes adopted in 2012 enhanced the effectiveness of the adjustment program. As the third round of negotiations has dramatically revealed, the debt restructuring agreed to in 2012 was far from sufficient to ensure debt sustainability, and the required fiscal efforts have worsened the depression and contributed to the erosion of social and political cohesion.

Finally, as any public bureaucracy, it is plausible to expect that differences of views on specific policy tools were present in each of the Troika Institutions at different points in time. However, while I acknowledge that the ECB, the EC, and IMF policy preferences can well be disaggregated, I build from a unitary actor's assumption for the purposes of the analysis. As applied to the EC, which negotiated as the agent of the Eurogroup, this assumption implies that its preferences can be treated as a proxy (albeit an imperfect one) of the outcomes of the negotiations among Euro area member states.

The paper develops its arguments as follows. First, the paper theorizes on the endogenous sources of actors’ preferences. Second, the paper reviews the terms of the original financial assistance program as well as the problems that led Greece to negotiate a second adjustment program. Third, the paper identifies the policy preferences of each of the three Troika Institutions at the time the second adjustment program started being formulated. Fourth, the paper shows how the layering of new crisis management rules into the Eurozone machinery was pivotal for explaining the changes to the adjustment strategy. The last section concludes by linking the findings of the paper to the emerging literature on new interdependence.

2. PUTTING PREFERENCES IN THE ‘INSTITUTIONALIST’ PERSPECTIVE

On 25 March 2010, the Heads of state and government of the Euro area stated ‘their willingness to take determined and coordinated action, if needed, to safeguard financial stability in the euro area as a whole’. To achieve this goal, they declared themselves ready ‘to contribute to coordinated bilateral loans’ as part of ‘a package involving substantial International Monetary Fund financing and a majority of European financing’.⁷ These statements marked the creation of the so-called Troika. The first Troika mission started consultations in Athens in April and the program was concluded in May. Similar missions were created to design the adjustment programs in Ireland and Portugal in 2011 and Cyprus in 2013.

Under the Troika arrangement, the ECB, the EC, and the IMF coordinate their financial assistance for crisis-stricken Eurozone member countries. This peculiar institutional arrangement was officially designed with the intention of improving the EU crisis management capacity by capitalizing on the longstanding IMF expertise in this area. Furthermore, the Fund's participation to the Troika was a device to lend credibility to the monitoring of financial assistance programs (Hodson 2015), particularly in the context of a monetary union in which members have direct vested interests in borrowers paying creditors (see also Rogers 2012).

Collaboration with other institutions is not a completely new arrangement for the Washington-based organization. The IMF has been involved with the development of the EMU since the signing of the Maastricht Treaty (Broome 2013) and the Fund had recently joined forces with the EC in extending financial assistance to EU members outside the Euro area (Hungary, Latvia, and Romania) (Lütz and Kranke 2014). Furthermore, although the IMF traditionally provides financial assistance on its own terms and conditions, it had coordinated its lending with the G7 and the World Bank in previous crises. In spite of these past collaborative experiences, the Troika framework was peculiar in a number of respects. For one thing, it was the first time that the central bank of the program countries had a seat next to the IMF on the creditors’ side. Furthermore, what was distinctive about the IMF's collaboration with the other Troika Institutions was ‘the degree of institutionalization’ and, specifically, the fact that ‘each participant was bound by the decisions of the group’ (Bernes 2014, 12).

Such a binding relationship is evident in operational modalities through which the three Institutions negotiate among themselves the policy position that is then presented to domestic authorities.⁸ In particular, the three Institutions negotiate the content of the MEFP, which encapsulates the overarching strategy for the key macroeconomic policies to be pursued in the program countries. The ECB, the EC, and the IMF also negotiate the MoU, which specifies the detailed economic policy measures that domestic authorities are required to implement within the overarching strategy agreed in the MEFP.⁹ Furthermore, the three Institutions collaborate during the monitoring stage to determine whether the conditions are in place for the disbursement of the successive tranches of financial assistance. This is done via a joint EC/ECB/IMF mission that visits the borrowing country with the aim of assessing compliance by national authorities with the terms and conditions of the economic adjustment program.¹⁰

In spite of its peculiarities, the Troika is somehow emblematic of a phenomenon that has become increasingly common in several areas of international relations, namely the phenomenon of regime complex (see, for instance, Keohane and Victor 2011, and the contributions in Alter and Meunier 2009). This term indicates ‘structures’ within which multiple, overlapping, and non-hierarchical institutions contribute governing a particular issue-area. Such fragmentation in governance responsibilities is important because is associated with specific behavioral outcomes. The literature on regime complexes has drawn attention to the political consequences that derives from institutional fragmentation. Among the others, institutional fragmentation provides states and non-state actors with the tools to game existing commitments most notably by opening up forum shopping opportunities (Alter and Meunier 2009a; Gomez-Mera and Molinari 2014). This scholarship, however, has thus far provided only limited insights on the relationship among the various institutions that make up the regime and, in particular, on how potential conflicts among them are solved (for an exception, see Breen et al. 2014). The same limitation applies to much of the scholarship on international organizations where questions related to the relationship with Member States (including questions of delegation and compliance) have overshadowed research on inter-institutional collaboration.

The presence of multiple institutions that interplay with each other is not an international phenomenon only. As such, it has not escaped attention in comparative politics and especially in the HI tradition. Orren and Skowronek (1994) developed the notion of ‘intercurrence’ to indicate the tensions that are routinely associated with the workings of numerous institutions and policies that operate simultaneously in a polity. In comparative political economy, the notion of 'complementarity' is used to indicate the configurations of domestic institutions that underpin specific varieties of capitalism (for instance Amable 2000; Hall and Soskice 2001).

The coexistence of sets of institutions is important because of the potential influence on political and policy developments, as change in one segment affects change in another (for a review. Fioretos et al. 2015, 12–13). This dynamic is well exemplified in the literature on gradual institutional change, where the introduction of new rules and institutions on top, alongside or as a replacement of existing ones figure prominently in the list of mechanisms of domestic change (Mahoney and Thelen 2010, Thelen 2004; Streeck and Thelen 2005). The implications of layering for institutional stability have similarly been identified in a number of studies that attempt explaining change in international financial regulation and economic cooperation (Farrell and Newman 2010; Moschella 2011; Rixen et al. 2016). One of the common themes of this scholarship is that layering is often a mechanism that allows disadvantaged political actors to bring about change. As Mahoney and Thelen (2010, 17) put it, ‘While defenders of the status quo may be able to preserve the original rules, they are unable to prevent the introduction of amendments and modifications.’

Having cast layering (as well as other mechanisms of gradual institutional change) as the mechanism through which advocates of change achieve their goals risks obscuring the possibility that the same mechanism(s) are used to perpetuate the status quo, especially when faced with environmental change. Just like change occurs when actors choose not to respond to environmental changes thus altering the impact of the institution (Hacker 2005), so layering can stymie change when actors decide to take actions to respond to environmental changes. In other words, institutional adaptation is a means to entrench existing arrangements (Thelen 2004). Actors that favor the status quo may adapt the institutional framework by adding pieces that help secure their policy preferences in spite of changed external circumstances. Layering can thus be thought of as an insulation strategy: actors that seek to preserve the status quo from outside pressures may achieve this goal through adapting existing rules and institutions to promote resistance (Farrell and Newman 2014a, 349).

In what follows, the paper applies these theoretical insights to the political negotiations among the Troika Institutions regarding the second adjustment program. In doing so, the paper argues that the evolution of the European Institutions’ preferences cannot easily be traced back to market pressures or the internalization of new beliefs as those advocated by the IMF on the speed of fiscal adjustment – as the application of the rationalist and constructivist lenses would have it respectively. Rather, the argument is that the evolution of preferences can only be thoroughly explained through an institutionalist lens: political actors ‘pursue projects based on their preferences within institutions, just as institutions delineate the scope of possible projects and help bring preferences forth’ (Katznelson and Weingast 2007, 15) In short, preferences are shaped endogenously by determinate situations regarding which choices are in fact on offer.

3. RESCUING GREECE: PART I

When in October 2009, the newly elected Papandreou government revealed that the budget deficit would amount to 12.5% of GDP, more than twice as high as previously reported, Greece faced severe fiscal and competitiveness challenges. These challenges can be traced back to the period following the country's accession to the Eurozone club (also Jones 2013). On the back of declining interest rates following the adoption of the Euro in 2001,¹¹ the country's fiscal position deteriorated as domestic governments adopted procyclical fiscal policies with tax cuts and increased spending on wages and entitlements.¹² Next to the fiscal problems, Greece competitiveness position had deteriorated in the run-up to the crisis following an upward trend in relative wages and prices. The loss in competitiveness is reflected in the large current account deficit, which hovered around 15% of GDP already in 2007 and 2008 (Gibson et al. 2012, 502).

From a systemic perspective, Greece's problems were apparently small; Greece represents approximately 2% of the Eurozone's GDP. However, the instability triggered by Greece's problems was greater than originally perceived and required substantial international support to be contained.

This support materialized in the conditional financial assistance agreed with the Troika in May 2010. Specifically, the Eurogroup agreed to provide bilateral loans pooled by the EC through an ad hoc facility (the Greek Loan Facility) for a total amount of €80 billion to be disbursed over the period 2010–2013.¹³ The financial assistance agreed by Euro area Member States was part of a joint package that included an IMF €30 billion loan, the largest Fund program to that date, relative to quota (IMF 2013, 9).

In addition to financial assistance provided by Euro area Member States and the IMF, Greece was supported by the ECB. When Greek debt lost its investment grade, the ECB relaxed its collateral rules so that Greek government debt instruments remained eligible as collateral for central bank financing. Furthermore, the ECB purchased Greek government bonds within the framework of the Securities Markets Program (see Xafa 2014).

As for the assistance provided under the Troika program, the adjustment strategy was built on two main legs: fiscal policy and structural reforms.¹⁴ Other policy levers, which have been traditionally applied in other financial crises, were excluded because of legal and political considerations. For instance, monetary policy was off the table because, in the EMU, it is the exclusive remit of the ECB and was thus considered a non-negotiable issue with the other members of the Troika. Devaluation was also ruled out because the political objective of the program was to keep Greece in the Eurozone. The restructuring of private claims on the Greek sovereign was also rejected as an option primarily because of its potential contagion risks across Euro area countries and out of fear that it would have created moral hazard by setting a precedent (in favor) of highly indebted sovereigns. Although these arguments were particularly well entrenched in the European members of the Troika as will be discussed in the next section, the IMF initially seemed joining the camp of those advocating against debt restructuring.¹⁵ For instance, in September 2010, the IMF's Fiscal Affairs department published a policy note entitled, ‘Default in Today's Advanced Economies’. The note argued that debt restructuring was ‘unnecessary and undesirable’ – as the subtitle reads – because default would not reduce the need for fiscal adjustment and that its political and economic costs would not be lower than those incurred under a strategy based solely on fiscal adjustment (Cottarelli et al. 2010)

With restructuring and devaluation off the table, and with no freedom to loosen monetary policy, the burden of adjustment fell on fiscal policy and structural reforms.¹⁶ The program frontloaded fiscal consolidation with a package of measures totaling to 11% of GDP during 2010–2013, in addition to the 5% of GDP in measures already adopted in 2010 (IMF 2010, 10). Among the specific measures, the program envisaged substantial reduction in government spending, particularly in pensions and wages, and an increase in government revenues to be attained by raising the value-added tax, taxes on luxury items, and taxes on tobacco and alcohol. The program also supported the creation of a revenue administration and expenditure control system to strengthen the country's tax collection system. Although the risks associated with such a large share of upfront measures were acknowledged, this ‘exceptionally strong frontloaded fiscal effort’ was regarded as necessary ‘to bolster confidence, regain market access, and put the debt-to-GDP ratio on a declining path from 2013’ (IMF 2010, 8). In particular, the fiscal effort was expected to bring the fiscal deficit under the 3% level by 2014, placing the debt–GDP ratio on a declining path from 2013 onwards.

Concerning the structural reforms, the focus was on measures to liberalize product, service, and labor markets, boosting competitiveness and exports. Furthermore, the program focused on safeguarding financial sector stability. In particular, the program focused on increasing banks’ safety net for addressing solvency pressures in a context of deflation. This was necessary as real GDP growth was expected to contract sharply in 2010–2011 and the frontloaded fiscal adjustment was expected to suppress domestic demand in the short run with negative effect on banks’ profitability. However, program projections assumed a rebound in growth from 2012 onwards.

Contrary to these rosy expectations, the adjustment strategy went unmistakably off-track. For instance, the turning point in GDP contraction, which was expected to materialize by late 2011, did not arrive. Instead, by the summer of 2011, it became clear that GDP was projected to contract by 3.75% instead of 3% as originally envisaged (IMF 2011b, 10). Furthermore, under the baseline scenario, Greece's public debt as a share of GDP was expected to peak at 149% in 2013 and gradually decline to 120% by 2020. However, by the end of 2011, it was acknowledged that, in spite of the fiscal adjustment, debt would have peaked ‘at a very high level of 187 percent of GDP in 2013’ (IMF 2011a, 66). Several factors contributed to the dismal performance of the first program, including external economic and political developments and the failings of Greek economic management.¹⁷ However, as several observers have noted (for instance, Manasse 2011; Summers 2015; Wyplosz 2010), it is not very surprising that a substantial fiscal contraction, in an economy that cannot loosen its monetary policy or devalue its currency, eventually leads to depression and political instability – although the program designers pretended, and often continue pretending, otherwise.

As the economic and financial situation in Greece continued deteriorating in the summer of 2011, the need to negotiate a new financial assistance program became public.¹⁸ The problem was that the three Troika Institutions held different views on how to bring the Greek debt back to a sustainable path.

4. POLICY PREFERENCES INSIDE THE TROIKA

In 2013, the IMF published a scathing report on the design of the 2010 Greek adjustment program that made public the disagreements harboring inside the Troika. At the root of the controversy lied different understandings of how to restore debt sustainability. Although the Fund had initially shared the European adjustment strategy for Greece (as discussed in the previous section), important changes had taken place within the organization since the start of the global financial crisis in 2008. In particular, intellectual and personnel shake-ups had led the organization to reconsider the tenets of its fiscal doctrine by paying more attention to the negative effect on domestic demand of too-quick fiscal adjustment (Ban 2015; Clift 2014). This thinking was applied to the Euro area and is clearly illustrated in the Fund's assessment of the design of fiscal adjustment in the Greek program. Specifically, the IMF acknowledged that the fiscal multipliers were set too low for the 2010 program, ‘despite staff's recognition that Greece's relatively closed economy and lack of an exchange rate tool would concentrate the fiscal shock’ (IMF 2013, 21) The Fund also started questioning the ‘political’ sustainability of the adjustment strategy: based on its long-standing experience in crisis-fighting, the IMF noted that ‘few countries have [ever] been able to sustain a 4 percent primary surplus’, as was required of Greece (IMF 2011a, 65). Given these limitations in program design and implementation, the Fund became skeptical of the optimistic assumptions on the fiscal efforts. As its 2011 debt sustainability analysis reads, even assuming ‘sustained and unwavering commitment to fiscal prudence by the Greek authorities, the overall fiscal balance would not drop below 3 percent of GDP until 2020’ (IMF 2011c, 2).

The IMF also became increasingly concerned about Greece's financing needs. As the Fund put it in the fourth review of the program published in early July 2011, in contrast to prior expectations, ‘Greece will take longer to regain capital market access’ (IMF 2011b, 32). In this scenario, the Fund recommendation was for Euro area Member States to restructure Greek debt by involving the private sector and accepting losses on official bailout loans. In the words of the Fund's report, ‘comprehensive private sector involvement [PSI] is appropriate, given the scale of financing needs and the desirability of burden sharing. And Greece's debt service capacity may also need to be bolstered by combining appropriate PSI and official support going forward’ (IMF 2011b, 33). The Fund's preferences for upfront restructuring are clearly worded in the already mentioned 2013 report, where it is also acknowledged that this solution ‘was not acceptable to the euro partners’ in 2010 (IMF 2013, 28)

Indeed, both the ECB and the EC, in its capacity of representative for Euro area Member States, had opposed an upfront debt restructuring as well as a loosening of fiscal discipline. These measures were opposed because of the risks they raised to the institutional stability of the monetary union. Specifically, debt restructuring and fiscal accommodation called into question the principle according to which each member country is responsible for its finances with the consequence that no monetary financing nor bailout can be enacted to make up for failures in putting the fiscal house in order.¹⁹

The ECB, in particular, had been the staunchest opponent of private involvement in Greek debt restructuring when the idea was initially floated at Deauville in October 2010 (Bastasin 2012; Bini Smaghi 2013). The ECB opposition was largely rooted in the concern about contagion risks across the Euro area, a contagion that would have been amplified by the bank–sovereign feedback loop (Angeloni and Wolff 2012; Howarth and Quaglia 2014). As Bini Smaghi (2011), then member of the ECB Governing Council, clearly articulated the central bank's stance on this issue: debt restructuring ‘entails taking a risk that no responsible policy maker can afford’.

Contagion was particularly perilous from the ECB's perspective because it raised the pressure on the monetary authority to push the limits of its institutional mandate. As the sole institution able to provide sufficient liquidity to sovereign bond markets to remove fears of default, the ECB would have been under pressure to take remedial action, pushing the central bank into the boundaries of the no monetary financing prohibition (see also Yiangou, O‘Keeffe, and Glockler 2013, 229). This risk was particularly acute because the ECB had bought Greek bonds at below face value as part of the already mentioned Securities Market Program. According to Open Europe, a London-based think tank, as of June 2011, the ECB faced €444 billion in potential losses from struggling Eurozone countries, with €140 billion from Greece alone (The Economist, Europe‘s debt crisis Trichet's long game, 13 June 2011). Losses in the ECB holdings of Greek government debt would have lent support to the view that the ECB market interventions had been undertaken in clear contravention to the Treaty prohibition on monetary financing and would have eventually required a forced recapitalization of the central bank by the Member States.

In addition to the reservations on debt restructuring, the ECB was also lukewarm at the prospect of diluting the pace and extent of fiscal adjustment because it did not want to set a precedent that could endanger the principle of ‘sound money’, which is one of the cornerstones upon which the EMU was built (Dyson and Featherstone 1999; Howarth and Loedel 2003; McNamara 1998).²⁰ Allowing domestic authorities to deviate from fiscal discipline would have raised further pressures on monetary policy and, in particular, on the principle of ‘monetary dominance’ (Lombardi and Moschella 2016).²¹ As Mario Draghi publicly commented, just some months before taking the helm of the ECB, ‘The response to the debt crisis is first and foremost in national policies, in the complete implementation of the adjustment plans that have been agreed’ (Financial Times, ‘Eurozone inflation shows surprise drop’, 31 May 2011).

Turning to the other EU member of the Troika, the EC shared many of the concerns raised by the ECB. In particular, the EC shared with the ECB the view that PSI would unleash severe contagion risks across the Euro area. From the perspective of the EC, these risks were particularly pernicious because they entailed a threat to the political project that the EMU embodies. Specifically, the EC feared that market contagion would push Greece out of the Euro area, with destructive consequences for the Euro project. As one interviewee in Brussels stated, ‘If Greece would have left, then the markets would have concluded that the Euro is not a permanent institution.’²²

The EC was also skeptical about the prospect of debt restructuring because of the potential implications for the no-bailout clause. This position reflected the political preferences of its principals, namely those of the Euro area Member States. Although creditor countries – most notably Germany with the support of France – led the way for PSI, they were much more wary of reducing the debt burden by taking losses on their outstanding loans to Greece. Doing so would have made the breach of the no-bailout clause public thus raising the prospects of a political backlash in several creditor countries. As a result, although the EC ultimately endorsed the IMF calls for deeper PSI (following the agreement reached in the Euro group in July and October 2011), it maintained a much more ambiguous stance on the contribution of the official sector. For instance, whereas the IMF fourth program review in July 2011 explicitly called on Euro area countries to consider providing official support in combination with PSI (IMF 2011b, 33), the analogous report issued by the Commission acknowledged the need for additional financing but fell short of recommending Euro area countries to provide the required financial support (European Commission 2011b, 102).

Concerning the design of fiscal policy, until late in 2011, the EC resisted proposals to slow down the fiscal effort. This position largely reflected the balance of power among Euro area countries, as the most powerful creditor states led by Germany were not ready to forego fiscal discipline. As the Bulgarian Finance Minister recalled the debate that was occurring in the ECOFIN in the spring of 2011 in his memoirs, ‘the common … point made was that it was better to address initially the short-term deficit issue and then worry about how to resolve the debt issue’ (Djankov 2014, 93). Reflecting the mood among Member States, the Commission rejected a more accommodative fiscal stance that risked creating a financing gap – which Eurozone partners were unwilling to fill in with further assistance.²³ The recalibration of fiscal adjustment was also opposed in light of the moral hazard effects that this choice would have unleashed (Matthijs and McNamara 2015). As the EC put it in its 2011 debt sustainability analysis:

If Greece can expect that its official creditors will eventually substitute for a lack of policy performance through additional financial support, the painful and politically difficult adjustment efforts needed to achieve sound fiscal position and healthy economic growth are unlikely to materialise. By contrast, if confronted with a hard budget constraint and the corresponding incentives to decisively break with past fiscal profligacy and weak policy implementation, debt sustainability prospects could be markedly improved. (European Commission 2011a, 30)

The EC was also skeptical at the prospect of alleviating the pressure for fiscal adjustment out of a concern for the rules that guide budgetary policy in the Euro area. In particular, the Commission perceived fiscal accommodation as a breach of the fiscal rules that govern the Eurozone, namely the Stability and Growth Pact, and thus as a negative precedent that had to be avoided to ensure the respect of existing norms.²⁴ As the IMF observed in its ex post assessment of the 2010 program, the Commission arrived at the Troika negotiations with ‘its own fiscal targets from Maastricht’ and was keen on ensuring compliance with EU norms (IMF 2013, 31).

As this overview of the policy preferences of the three Troika institutions reveals, the negotiations on the second adjustment program started off with significantly different views on whether to grant relief on the Greek debt stock and on what fiscal policy stance to recommend to stabilize debt. In particular, the ECB and the EC were more adamant than the IMF to not change the original adjustment strategy because they were concerned that debt restructuring and fiscal accommodation would have endangered the key principles upon which the EMU had been built. And yet these concerns were ultimately overcome with the signing of the second program in March 2012. How such an outcome materialized requires considering how the new rules for crisis management developed between 2010 and 2012 led the ECB and the Commission to reassess the political implications of a changed strategy in Greece.

5. RESCUING GREECE: PART II

The program agreed in 2012 was supported with financing from Euro area Member States, which committed €144.7 billion through end-2014, and from the IMF, which contributed a four-year €28 billion loan. Although the adjustment strategy was not fundamentally different from the one that had guided the 2010 program, the new program deviated from the original one in a number of important respects. To start with, the pace of adjustment was slowed to account for the potential deflationary effects of the required structural reforms, especially those in the labor market (IMF 2012, 18).²⁵ In particular, the program allowed a primary deficit of 1% of GDP in 2012 and back-loaded the bulk of fiscal adjustment to 2013–2014. Furthermore, debt reduction entered the program through two channels. First, the Greek debt burden was reduced, with the offer made by Greece to its private creditors for a reduction in the nominal value of government bonds by 53.5%. Second, Euro area countries agreed to a retroactive lowering of the interest rates on the 2010 loans²⁶ and to pass on to Greece the income accruing on Greek bonds held by the national central banks until 2020.²⁷

The economic strategy that underpins the second program incorporated most of the IMF policy preferences, from more lenient fiscal policy to debt restructuring. It would thus be tempting to attribute the change in the policy strategy to the influence that the IMF exerted on the negotiations. However, this argument is not fully supported by the empirical record. While it is true that the IMF made its concerns known to the Troika partners on the prospect of debt sustainability and on the pace of fiscal adjustment, its advocacy was not sufficiently candid and its assumptions on Greek growth, fiscal effort, and social cohesion often too optimistic (Moghadam 2015). Furthermore, the IMF did not always get what it wanted. For instance, and in spite of the recalibration of the fiscal strategy, IMF reports hint that the Fund would have preferred ‘a more accommodative fiscal policy in the near term’ to offset possible short-term costs of structural reforms (IMF 2012, 18). However, ‘resistance from their European partners’, weakened the Fund's negotiating stance because Greek authorities ‘worried that [a more a accommodative fiscal stance] would be seen as a lack of commitment to Stability and Growth Pact targets’ (IMF 2012, 18). Furthermore, in spite of IMF calls for official sector involvement in reducing the debt burden, the contribution of Euro area countries fell short of accepting losses on outstanding loans.

If the IMF arguments failed to influence European partners, why did the ECB and the EC eventually accept some adjustment measures that they had forcefully resisted until some months before? Market pressures could provide a cut-and-dried answer to this question. As financial turmoil spilled from Greece to other peripheral countries, driving yields up to record levels in the second half of 2011, EU Institutions were simply forced to change their adjustment preferences.²⁸ While this explanation certainly captures an important driver of policy-makers’ motivation to revise the terms of the original program, the empirical record does not neatly fit with its theoretical expectations. For instance, market sentiment significantly deteriorated in the fall of 2011 at the prospect of a deep reduction in the value of Greek government bonds held by the private sector.²⁹ However, the evolution in financial markets did not prevent debt restructuring from being included in the second program. Furthermore, discussions for a new financing package were drawn out until early 2012 – well after markets had started indicating high sensitivity to restructuring moves.In short, it is difficult to map the content of the new program onto market pressures.

The limitations of the arguments that emphasize the persuasive impact of IMF's advocacy and the power of market pressures lead us to explore an alternative explanation of why the European Institutions eventually relaxed their original opposition to debt restructuring and fiscal accommodation. Specifically, a more complete explanation requires taking into consideration the changing institutional framework within which Greek negotiations took place and, in particular, the creation of new rules on the provision of financial assistance.

The starting point is that the Euro area lacked a crisis management capacity when the sovereign debt crisis started.³⁰ In particular, the Euro area had not a framework for providing bridge funding to member countries to correct imbalances and no mechanism for dealing with contagion effects. The Greek crisis and those that followed in Ireland, Portugal, Cyprus and Spain, vividly showcased the perils of such an institutional gap in the EMU architecture. Developing a crisis management capacity thus became one of the major items in the EU reform agenda and one that occupied policy-makers from 2010 to early 2012. Specifically, Eurozone policy-makers created the temporary European Financial Stability Facility (EFSF) which was later replaced by the ESM.³¹

The EFSF was designed as a three-year tenured institution subject to an intergovernmental agreement. Its aim was to safeguard financial stability in Europe by providing financial assistance to Euro area members within the framework of a macroeconomic adjustment program. To fulfill its mission, the EFSF was allowed to issue bonds or other debt instruments on the capital markets and on-lend the proceeds to countries under a program. To provide the facility with a sufficient capital base to raise funds in financial markets, Euro area governments agreed to equip it with €700 billion in pro-rata guarantees, able to support up to €400 billion of lending.

Because the crisis in the Eurozone did not lose its steam, Euro area authorities were forced to step up the profile of their crisis management framework. In particular, from 2010 to early 2012, the Euro area gradually refined the workings of the EFSF with the aim of developing a full-fledged firewall for the currency union. To achieve this goal, the lending capacity was raised to an effective €500 billion, and financial assistance widened to include not only loans to sovereigns but also interventions in primary and secondary sovereign markets of Euro area countries, support to banks and precautionary interventions.

The build-up of the Euro area crisis management system culminated in the decision to replace the EFSF with the permanent ESM. The ESM is responsible for providing financial assistance packages to Euro area Member States under strict conditionality. Furthermore, following the agreement reached at the European Council on 30 January 2012, the ESM Treaty stipulates that the granting of financial assistance is conditional on the ratification of the Treaty on Stability, Coordination, and Governance in the Economic and Monetary Union (the ‘Fiscal Compact’) by the borrowing country. In other words, financial assistance is reserved for those members that commit to the principle of fiscal discipline by including balanced budget rules in national constitutions (or equivalent legislation).

The creation of the EU crisis management capacity reflects the influence of many factors, including political and legal constraints (Yiangou, O‘Keeffe, and Glockler 2013), sunk political costs (Gocaj and Meunier 2013), learning (Schwarzer 2015), and even mere necessity as financial conditions deteriorated. The European Troika Institutions also contributed to the shape of the new institutional setup, by advising member countries and articulating specific proposals on the design of financial assistance in the Eurozone.

For instance, it is no secret that Jean-Claude Trichet had forcefully asked Eurozone members to step up efforts to build a credible crisis management system after the start of the crisis as the necessary precondition for an orderly debt restructuring in Greece and for long-term financial stability in the Eurozone (Bastasin 2012, 251).³² These calls were reiterated even after the creation of the EFSF with the ECB President inviting domestic leaders to improve it ‘quantitatively and qualitatively’ (Reuters, ECB Trichet urges flexibility in size, scope of EFSF, 14 December 2010). In particular, the ECB made clear that it would support expanding the size of the newly created Facility and giving it the power to buy government bonds (Financial Times, ‘March deadline for stability accord’, 13 January 2011). Although couched in the diplomatic language of central banks, these calls offer a thin veil for the preoccupation that, absent an appropriate institution charged with providing financial assistance, the burden of the task would have fallen on the ECB.

The Commission led by José Manuel Barroso was similarly forthcoming in its support for speeding up the process of creating a full-fledged Euro area firewall (Financial Times, ‘March deadline for stability accord’, 13 January 2011). Furthermore, the Commission proposals to strengthen the lending capacity of the EFSF were presented as inextricably linked to the reinforcement of fiscal governance in the Euro area. In the words of Barroso, ‘We need stronger governance; we need more discipline. We need integration and discipline’ (as reported in The Guardian, Barroso ready with plan to prop up European banks, 12 October 2011).

The ECB was also vocal about the need to design a crisis management framework that did not undermine the principle of fiscal discipline by inducing moral hazard. As the central bank articulates this point in the legal opinion on the creation of the new stability mechanism, ‘Safeguards such as … non-concessional terms … and regular and strict surveillance on compliance by the assisted Member States with the program of fiscal and macroeconomic adjustment on which financial assistance is conditional, are indispensable for providing strong and lasting incentives for sound fiscal and economic policies in the Member States whose currency is the euro’ (ECB 2011). This position was echoed in several creditor countries where the gradual revisions to the Eurozone crisis management framework were adopted with the view to force crisis stricken countries to put their budgets on a sustainable basis (Financial Times, ‘Merkel's intransigence is a question of time and tactics’, 27 February 2012).

In short, the ECB and the EC put their weight on the creation of a Eurozone crisis management framework with specific institutional features. While the relative influence of these political actors on the actual design of the new framework is open to empirical research, what is relevant for the purposes of this study is that the existence of the new framework was pivotal in shaping agreement on the second adjustment program in Greece. In particular, the creation of a crisis management system in which assistance is provided under strict conditionality led the ECB and the Commission to reassess their concerns about the implications of a changed strategy in Greece for the Euro area.

The ECB, for instance, started hinting that the conditions were now in place to relieve the central bank off the pressures for intervention, should debt restructuring unleash nasty contagion effects amplified by the financial sector. As the vice chairman of the ECB, Vítor Constâncio, commented on the second Greek bailout, ‘What is important to underline is that there is a huge amount of money in the second program approved for Greece to address the recapitalization of Greek banks and also all the operations to restructure the sector’.³³ Striking a similar note, Draghi emphasized the safeguards created via the EFSF that allowed the Eurosystem to continue accepting Greek bonds as collateral even were they to be downgraded to ‘selective default’, thus ensuring that the central bank was not violating its mandate (Draghi 2012). In other words, the ECB concerns receded only after EFSF/ESM resources were increased and their operations refined to ensure an effective firewall against financial contagion and a shield from political risks for the central bank.

Likewise, concerns that a looser fiscal stance would be a prelude to more financing from Euro area partners (and thus a breach to the no-bailout clause) were dispelled following the establishment of a crisis management framework that anchors financial assistance to a firm fiscal adjustment path. In this spirit, the Commissioner for economic and monetary affairs, Olli Rehn (2012), presented the agreement on the second program for Greece as having ‘a strong emphasis on sustainable growth and jobs’ but within the overall framework of ‘putting Greece's public finances on a sustainable path’. In other words, fiscal conditionality ensured that Greece could ‘only draw on a fixed envelope for official support’, as one member of the ECB Executive Board put it (Asmussen 2012), implying that neither bailout nor further assistance from Member States were in the cards.

In short, the EU crisis management framework provided an overarching institutional layer through which it was possible to decouple the Euro area from the repercussions of a changed strategy in the Greek adjustment program. Because a new set of rules was in place that preserved the EMU's most basic principles, namely the one according to which each Member State is responsible for the sustainability of its own public finances without recourse to bailout or monetary financing of public debts, the ECB and the EC's opposition to debt restructuring and fiscal accommodation was eventually relaxed. It is no coincidence that the Troika started negotiations with the Greek authorities on 17 January 2012, following the 2011 December EU Council when the final touches on the ESM were agreed upon, and announced the second program immediately following the signing of the ESM treaty.

6. CONCLUSIONS

It is fair to say that the Greek adjustment program has been the most controversial among those agreed among Euro area countries since the start of the sovereign debt crisis. Program design failures and political and economic mismanagement, both in Greece and at the European level, have combined to make Greece a textbook case of tense international economic negotiations. The negotiations for the third bailout program in the summer of 2015 are but the latest example of the acrimonious relationships between Greece and its creditors as well as among creditors (Moschella 2016).³⁴

This paper addressed the controversies surrounding the Greek program by shedding light on the political dynamics among the three Troika Institutions in the shift from the first to the second bailout negotiations. Specifically, the paper illuminated the institutional logic that drove the evolution of the preferences of the European Institutions, which were initially opposed to debt restructuring and to the slowdown in the path of fiscal adjustment. Rather than being driven by exogenous market pressures or by the internalization of new beliefs, the preferences of the ECB and the EC were reshaped through the layering of new rules that specify how financial assistance can be disbursed to member countries. This institutional change altered the way in which the ECB and the EC assessed the risk of debt restructuring and fiscal accommodation for Greece. Since the two Institutions had rejected these measures out of concerns that they would have weakened the principles of monetary independence and fiscal discipline that underpin the EMU, the creation of rules that insulated the EMU from such potential negative effects led the ECB and the Commission to relax their original opposition.

In advancing these arguments, the paper contributes to the growing research agenda on the application of historical institutionalism in international relations and international political economy (Farrell and Newmann 2010; Fioretos 2011; Moschella and Tsingou 2013; Posner 2010; Rixen et al. 2016). Specifically, the paper shows that actors’ preferences are endogenous to the institutional system in which these actors operate – and as such they are alterable over time due to factors internal to the system. Acknowledging the institutional dimension of preference formation and evolution helps overcoming the limitations of well-established rationalist and constructivist explanations, which are here exemplified in the arguments positing the influence of market pressures and IMF persuasion.

The paper also exemplifies some of the themes in the research agenda of the NIA. To start with, the conflict among the Troika Institutions speaks to the problems that arise under conditions of increased interdependence whose management requires the collaboration of several political actors who act beyond traditional domestic/international divides (see also Johnson in this issue). Furthermore, the paper confirms one of the key propositions of the NIA, namely the proposition according to which rule overlap offers some political actors more power than others. In the case under investigation, the ECB and the EC’s access to the development of new crisis management rules provided them with a source of asymmetric power as they layered new rules aimed at preserving fiscal discipline in the monetary union.

The paper also contributes to the development of the NIA by drawing attention to an understudied phenomenon of globalization, namely the proliferation of cases of interaction among international institutions. Although the creation of complex systems in international cooperation has certainly not escaped scholarly attention (for instance, Alter and Meunier 2009; Gehring and Oberthür 2009; Keohane and Victor 2011; Raustiala and Victor 2004), the paper goes beyond the ‘regime complexes’ debate on the origins of these regimes and their consequences for the prospects of compliance with the regime rules. The analysis instead focused on examining the factors that shape the balance of power among the institutions that are part of the same regime as well as the modalities through which inter-institutional conflict is managed over time. This paper has only scratched the surface of this power relationship; additional empirical investigation is warranted in future research.

ACKNOWLEDGEMENTS

The author is grateful to Cornel Ban, Ben Clift, Eliot Posner, and the journal's editors and anonymous reviewers for thoughtful and incisive comments on earlier drafts. Special thanks go to Abe Newman and Henry Farrell for close reading of several versions of the manuscript as well as for their incredibly helpful comments. An earlier version of this article was presented at a 2015 Workshop at the Mortara Center of the Georgetown University and to the annual meeting of the International Studies Association, and the author thanks all who participated in those sessions for their useful feedback.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the author.

Additional information

Notes on contributors

Manuela Moschella

Manuela Moschella is an associate professor in International Political Economy at the Scuola Normale Superiore and senior fellow at the Centre for International Governance Innovation. She is the author of Governing Risk: The IMF and Global Finanicial Crises (Palgrave 2010) and a number of articles on the political economy of international finance. Her current research explores central banking in the advanced economies and international macroeconomic cooperation.

Notes

1. In early 2015, Eurozone finance ministers agreed to extend the Greek second bailout by four months. In the Eurogroup statement of 20 February that announced the decision, the ‘Troika’ was relabeled as the ‘Institutions’.

2. For an assessment of the Greek program and the problems that have bedeviled its implementation, see, for instance, IMF (2013) and Pisani-Ferry, Sapir, and Wolff (2013).

3. See the Fund's retrospect in IMF (2013).

4. Throughout the article I refer to the ECB and EC as the ‘European’ Troika Institutions for terminological convenience. However, it is worth noting the ECB is an EMU institution only. That is to say, its monetary policy decisions apply only to those European countries whose currency is the euro.

5. The policy preferences of the ECB and the EC are discussed in detail in the following sections.

6. Interviews were conducted on a ‘deep background’ basis, with interviewees assured of confidentiality to encourage candor about sensitive matters. Interview subjects are thus referenced here anonymously.

7. Statement by the heads of state and government of the Euro area, 25 March 2010.

8. This paragraph draws from interviews with IMF and European officials that participated to Troika negotiations.

9. The MEFP is initially drafted by IMF staff and then circulated among the other Troika members. The EC takes the lead in preparing the MoU because the document covers the structural issues that the Commission routinely addresses in application of EU law and regulations.

10. The three Institutions release separate monitoring reports, however. Although the IMF has its own report, the Commission prepares its program review in liaison with the ECB.

11. The nominal interest rate on 10-year Greek government bonds declined from about 20% in 1994, when the goal of bringing Greece into the Euro area in 2001 was announced, to less than 3.5% in early 2005. See Gibson et al. (2012).

12. The analysis draws from the IMF's assessment of Greek economic problems, in particular, IMF (2010, 4).

13. For further details, European Commission, Financial Assistance to Greece; accessed at http://ec.europa.eu/economy_finance/assistance_eu_ms/greek_loan_facility/index_en.htm, 22 January 2015.

14. The rationale of the program is reconstructed based on both interviews and program reports issued from 2010 to 2015. Further insights are gleaned from the transcripts of an IMF conference on the Euro crisis held on 20 April 2012. The transcripts are available at https://www.imf.org/external/np/tr/2012/tr042112b.htm

15. Blustein (2015, 1) argues that ‘debt restructuring was being actively explored at the IMF, though in the most sotto voce way imaginable’, since the first adjustement program.

16. To allow time for the fiscal effort to occur, the program removed Greece from the bond markets for 18 months, programming a gradual return thereafter.

17. Among the external developments that most affected Greek performance, two stand out. First, the issue of private sector involvement was brought into the political agenda by the Franco-German Deauville agreement. The most immediate effect of the announcement to involve the private sector in Euro area debt restructuring was a plunge in investors’ confidence that pushed Greek government interest rates upwards. Second, a nascent recession in Europe started weighing on Greek exports, further depressing growth. See IMF conference on the Euro crisis, 20 April 2012. The transcripts are available at https://www.imf.org/external/np/tr/2012/tr042112b.htm

18. At the July and October 2011 EU Summits, European leaders called for a new financing package for Greece and mapped out the tenets for the involvement of the private sector in the restructuring of Greek debt.

19. The prohibition of monetary financing and the ‘no-bailout’ clause are spelled in the Treaty of the Functioning of the European Union (TFEU), Article 123 and 125, respectively.

20. Author's interviews with European and IMF officials involved in the Greek negotiations (Brussels and Washington, DC, May, June, and August 2014).

21. ‘Monetary dominance’ means that domestic authorities should be ready and willing to use fiscal levers (revenues and primary spending) to stabilize debt at any level of the interest rate that the central bank chooses – the opposite would imply fiscal dominance.

22. Author's interviews with a European official involved in the Greek negotiations (Brussels, May 2014).

23. Author's interviews with officials at the EC (Brussels, May and June 2014).

24. That the EC was keen on ensuring the respect of the Maastricht criteria was emphasized in interviews in both Brussels and Washington, DC.

25. The package of labor measures included reforms to the collective bargaining system, reductions in the minimum wage, and in non-wage labor costs.

26. As noted above, in 2010, Euro area pooled their bilateral support for Greece in ad hoc facility named Greek Loan Facility, whose interest rates were now lowered.

27. Statement by the Eurogroup, 21 February 2012.

28. See, for instance, ECB (2012), Financial Stability Review, June.

29. By mid-November, spreads on 2-year and 10-year Greek debt over German bunds exceeded 11,350 bps and 2600 basis points, respectively.

30. The crisis vividly exposed several other gaps in the institutional architecture of the monetary union, including the lack of a fiscal backstop or common banking supervision (e.g., De Grauwe 2013)

31. In addition, Member States extended the balance assistance support that the EC can provide to non-Euro area members to all EU member states under the EFSM.

32. See also Jean-Claude Trichet's remarks on the EFSF in the introductory statement to the press conference (with Q&A), Frankfurt am Main, 13 January 2011, available at https://www.ecb.europa.eu/press/pressconf/2011/html/is110113.en.html

33. Introductory statement to the press conference (with Q&A) Mario Draghi, President of the ECB, Vítor Constâncio, Vice-President of the ECB, Frankfurt am Main, 4 April 2012. https://www.ecb.europa.eu/press/pressconf/2012/html/is120404.en.html

34. For instance, Financial Times, ‘Greek politicians accuse EU hardliners of blocking deal’, 12 July 2015, and Financial Times, IMF signals it could walk away from Greek bailout deal, 15 July 2015.