When do crises centralise decision-making? The core executive in the Greek economic crisis

ABSTRACT

When can crises overcome the fragmentation of a core executive and facilitate a centralised management response? Here, we identify the latter by reference to the concept of a ‘crisis response network’ (CRN). We draw on several literatures that refer to crisis centralisation and develop hypotheses that comprise likely contributing factors. We explore these hypotheses in the setting of a core executive normally identified as being ‘segmentary’, but which has exhibited centralised management of past crises, and in the context of an acute economic crisis: that is, Greece in its debt crisis in two seminal periods. Based on extensive interviews with all the senior personnel involved, including both Prime Ministers, as well as documentary evidence, we find that the strength of the CRN in leading the crisis response varied and that this was consistent with our hypotheses. We consider the conceptual and empirical implications of our findings.

KEYWORDS:

• Core executive; crisis management; leadership; prime ministers; Greece; Eurozone

Introduction

Across different literatures, crises are frequently seen as prompting a centralisation of the management response. This is evident in crisis management studies, analyses of foreign policy crises, studies of US presidents, and in investigations of the eurozone crisis. Yet, while crises may give rise to such expectations, the generic effect is by no means certain, but conditional.

What, then, may be the crisis impact, not in a presidential system, but in that comprising a prime minister and a cabinet?¹ The latter is one of complex and mutual dependencies, as captured by the frame of a ‘core executive’. Here we test the centralisation proposition in a particular core executive identified as often being fragmentary and inimical to a strong central direction.

To assess the crisis impact, we choose two governments facing an economic crisis in the context of the Eurozone. We focus on the case of Greece, where the crisis was at its most severe. Our research question is: when do economic crises prompt centralised decision-making? Such centralisation may occur around the PM or a designated senior minister, but in our case, the centralisation would occur despite the wider and longer-term fragmentation of the core executive. We select two seminal periods: the initial eruption of the country’s debt crisis in 2009 (when George Papandreou, leader of PASOK, was prime minister) and its biggest conflict, in 2015 (under Alexis Tsipras, leader of SYRIZA) and we follow these cases until the signing of each of their ‘bail-out’ agreements: in May 2010 and July 2015, respectively. For both, the crisis involved immense external pressure, acute time constraints and uncertainty with the risk of the ‘money running out’ in their first months in office. We choose these two cases – rather than the intervening government of Adonis Samaras or the short-lived administration of Lucas Papademos – as ones where the force of the crisis was at its sharpest and where the interpretation of the crisis was more in question. These conditions highlight the importance of the strategic lead given by the PM.

Greece is an intriguing context in which to assess the crisis effects. The norms and practices of the core executive are deeply embedded. Since 1974, the core executive has evolved within a ‘party-state’, exhibiting patronage and a culture of short-termism (Featherstone, 2015; Sotiropoulos, 2000; 2004; Spanou, 2008). Indeed, the senior posts across government are political appointees, subject to change between governments and ministers, sustaining little continuity of institutional knowledge or personnel. Ministries have established de facto much operational autonomy (or ‘fiefdoms’) in their respective policy domains. This has been compounded by an opacity of organisational lines of responsibility. And, yet such fragmentation has existed alongside the historical importation of a quasi-Napoleonic administrative tradition providing hierarchy, legalism, rigidity and impersonalism akin to Crozier’s (1964) classic depiction of the bureaucratic state in France (see Spanou, 2008). Fragmentation and hierarchy have impeded innovation or reform. Crozier suggested that, in such a model, meaningful change was very difficult and was likely to involve a crisis.

In recent Greek history, some crises do appear to have centralised decision-making around the PM, however. Constantinos Karamanlis is widely recognised to have managed the transition to democracy in 1974 on the basis of a tightly-knit circle of trusted staff, distanced from a wider sphere that might pose risks to the PM’s own security (Featherstone & Papadimitriou, 2015; Woodhouse, 1982). In the Seismik crisis with Turkey over oil exploration rights in 1987, Andreas Papandreou as PM clearly took charge of Greece’s response, with a disposition to assert national independence (Karamouzi & Chourchoulis, 2019), personally ordering that Turkey’s survey ship be sunk if it entered Greek waters. More recently, in the recent COVID pandemic, Zahariadis et al. (2023, p. 61) concluded that ‘the Greek response was a top-down, highly centralized process where decision-making power was concentrated, the message was centrally crafted and delivered, and the adopted nationwide stringent measures were enforced by the state’ Similarly, Exadaktylos and Chatzopoulou (2023, p. 53) confirm its ‘high centralisation' and note that this ‘was not an expected outcome’. But, ‘this strategy worked well to demonstrate that the government was in control of the situation: clear message, parade of expertise, enforcement of measures’ (Zahariadis et al., 2023, p. 61).

The above cases are of centralisation amidst different types of crises. Studies of the eurozone crisis impact on Greece give only limited evidence of change. With respect to public administration, bail-out conditionality does not appear to have led to paradigmatic shifts and, indeed, the conditions set seem to have been problematic (Featherstone, 2015; Kougias, 2020; Spanou, 2016). Lampropoulou and Oikonomou (2020) examined how conditionality had promoted delegation to semi-autonomous agencies, ostensibly to circumvent established constraints. None of these studies were concerned with how the crisis impacted on the position of the PM, however.

Our task, then, is both conceptual and empirical: to ascertain the conditions that might enable centralised decision-making in a crisis and to assess how far our two cases met them. By ‘core executive’, we follow the now-classic conception of Dunleavy and Rhodes (1990, p. 4) to refer to ‘all those organisations and structures which primarily serve to pull together and integrate central government policies or act as final arbiters within the executive of conflicts between different elements of the government machine’. To identify centralised decision-making, we adopt the concept of a ‘crisis response network’ (CRN) (Boin et al., 2016, p. 68). The term has often been applied in the crisis management of other types of case such as an environmental disaster, involving emergency services (e.g., Moynihan, 2009). The CRN frame allows us to differentiate the actors directly involved in the crisis response (executive decisions, policy planning and advice) from the more general core executive pattern.

We should note that our focus on the Greek government’s crisis management differs from previous studies. It does not attempt to account for the bargaining process between Greece and its lenders or the outcome of the bail-out negotiations (see, for example: Hennessy, 2017; Lim et al., 2019; Pitsoulis & Schwuchow, 2017; Tsebelis, 2016; Zahariadis, 2017). Instead, our contribution is to provide a conceptual frame by which to test the centralisation proposition within a core executive; to examine this in a challenging setting; and to explore how the Greek crisis was managed in Athens.

The article proceeds as follows. Section Two develops the conceptual frame and outlines the hypotheses for when decision-making may be effectively centralised via the CRN. Sections Three and Four apply this to the empirical examination of the two crisis governments in Greece. Section Five concludes with a discussion of the results and implications.

Conceptual frame

Since Hermann’s (1963) classic study of the impact of crises on organisations, the most widely identified and supported structural response in crisis management studies has been the centralisation of decision making (t’Hart et al., 1993; see also Boin et al., 2016, p. 49). These, together with leadership studies, emphasise the importance of leaders directing the response and of those decisions being pivotal (Boin & t’Hart, 2012). Traditionally, crises in foreign policy, in particular, are seen as requiring the ‘centralization of decision-making authority around the leader of the executive’ (Lehmann, 2011, p. 27; Young, 2013). This ‘doctrine of centralization’ has been followed in crises, large and small, and a ‘substantial literature exists showing that in a crisis the number of actors involved reduces dramatically’ (Lehmann, 2011, p. 32). Further, in studies of US presidents, crises are often seen as a test of the incumbent’s ability to exercise the powers of the office (Burke & Greenstein, 1988; Young, 2013). The ‘most notable aspect’ of the response to the 9/11 attacks, for example, ‘was its centralization’ (Lehmann, 2011, p. 34).

Studies to date give some reason to expect centralisation effects from the eurozone crisis, though the pattern is not consistent. Kickert (2012) found the centralisation impact varied between countries and over the crisis cycle. In his survey, Parrado (2012) suggests the crisis strengthened the PM’s Office (Spain); created a dual centre of power between the PM and the finance minister (Germany; see also Schoeller, 2019); and empowered governments over parliaments (Spain). Raudia et al. (2015, p. 842) found survey evidence from 17 EU member states of a ‘centralization cascade’ in which there was a spill-over of centralisation from one part of the decision-making process to another. Of relevance to Greece, they note the tougher the sanctions for failure to hit targets, the greater the centralisation effect.

While crises may prompt these outcomes, the generic effect is by no means certain across cases (Boin et al., 2016). There is a contingent nature to crisis management responses (t’Hart et al., 1993). The task, then, is to elaborate what factors are more likely (or not) to result in centralised decision-making. By ‘centralisation’ here, we primarily refer to the strategic decision-making that set the key parameters and preferences and would be expected to emanate from the ‘top’ of government (t’Hart et al., 1993).

Leadership and crisis management studies place the leader between the crisis and the institutional setting. Indeed, our focus is on leadership confronting a crisis external to a government, rather than a leader setting out to be transformational of the organisation itself. ‘Leadership Trait Analysis’ and ‘Operational Code Analysis’ have argued the significance of leadership personality to foreign policy decision making (Çuhadara et al., 2021; Dyson, 2023; Kaarbo, 2021). In comparative politics, the ‘Leadership Capital Index’ differentiates leadership strengths (Bennister et al., 2015). A longer and wider literature similarly asserts the importance of the agency of the leader in a crisis. The leader narrates and makes ‘sense’ of the external crisis (Boin & t’Hart, 2012, p. 181). The narration may be in complex conceptual terms (Hermann, 1980; Siniver & Featherstone, 2020) or part of a grand narrative (Dyson, 2006). Who owns the narration – the definition of the crisis – helps to determine the scope of actions that might be agreed. The PM’s approach may encourage ‘groupthink’, blocking out alternative perspectives (Janis, 1982; Turner & Pratkanis, 1998).

Those acting for the PM need to know his/her strategic preferences and priorities and how to interpret them as events unfold. A clear strategic narrative has a predictability in what is expected and is one nuanced to possible new circumstances. Fall-back preferences should be evident: what plan B is, if A fails. An inadequately informed or too rigid a strategic narrative (George, 1980) can be dysfunctional for those acting on the PM’s behalf: undermining the effectiveness of the response; creating confusion or disagreement; and prompting internal fragmentation.

The leader’s task orientation also impacts on the strength of the lead given. Is the PM at the forefront of the management of the crisis, giving strategic direction and responding to new challenges, or is the PM distanced or not sufficiently communicative? The PM’s task orientation may be affected by his personal knowledge and/or experience of government (Kaarbo, 1997), asserting leadership or deferring to others. Conditions may favour assertion or not: a PM’s standing within the government and/or party offers political capital that can be deployed. The PM is expected to make the tough decisions and to resolve internal conflicts (Hermann & Preston, 1994, p. 81; Kaarbo, 1997, p. 553). But there may be political self-interest to calculate for the PM: to be seen shouldering the responsibility or to cede the space for others to bear the cost.

In the instances in which the Greek core executive, in the past, proved able to overcome its fragmentation and centralise decision-making in a crisis, certain conditions appear to have prevailed. In each case – Karamanlis in 1974; Papandreou and Seismik in 1987; Mitsotakis in 2020–2021 – the PM’s crisis narrative was seen as accurately representing the conditions being faced; the strategic objectives were clear and understood; and the PM was fully engaged in the crisis management. Moreover, appropriate mechanisms were in place to carry out the PM’s strategy. In each case, the PM could also rely on a supportive Cabinet.

From the above, we take the agency of the PM as critical in this process. Our first independent variable is therefore the clarity of the PM’s strategic lead. We define this broadly, to cover the PM’s crisis narrative; strategic objectives; and, task engagement, as follows:

Crisis narrative: how far is the PM,

1. informed and understanding of the nature of the crisis and the challenges to be faced;

2. open and responsive to new information and contending ideas, as well as changing conditions.

Strategic objectives: how far does the PM,

1. set clear objectives to be followed, which are readily understandable by those acting on the PM’s behalf;

2. sanction actions that can reasonably be expected to achieve the desired objectives;

3. indicate an order of preferences when confronted with choices, providing a degree of predictability.

Task engagement: the extent to which the PM,

1. Is at the forefront of the strategic management of the crisis, with a strong task engagement, ready to respond to new developments and act consistent with the above.

On this basis, our first hypothesis is:

H1: The greater the clarity of the PM’s strategic lead, the greater the strength of the CRN in providing a centralised response to the crisis.

In other words, the greater the clarity of the PM’s strategic lead, the more centralised crisis decision-making will be by proxy of a CRN operating on the PM’s behalf (our dependent variable). A strong CRN will be empowered by the PM’s strategic lead, clear in what it is expected to do and with the ability to act accordingly. A weak CRN will be uncertain of the strategic objectives, unclear or divided in how to respond to changing events, and will lack the capacity to act on the PM’s behalf. With the former, decision-making will be centralised, while with the latter it will be diffuse.

As already noted, our research question – when do economic crises prompt centralised decision making? – is not necessarily restricted to an outcome where the PM is central: decision-making may instead centre around the finance minister (Parrado, 2012). But, in this case, the PM needs to empower the minister with a direction similar to the above and the latter needs to be confident that they have a full and clear agreement.

The centralisation of crisis decision making, however, is not contingent on prime ministerial agency alone. The setting of the core executive places PM leadership within a complex set of relationships. Relationships within the core executive display mutual dependencies: crucially, the PM lead rests on his/her support amongst senior ministers and the Cabinet collectively.

Thus, centralised management around the CRN may prompt challenge from within the Cabinet and its reaction can be consequential for the legitimation of its role. A Cabinet can involve political divisions limiting the scope for a shared crisis understanding or action. A crisis may disturb the levels of trust amongst these actors and/or affect team-building (Boin & t’Hart, 2012, p. 186). The Cabinet may contain powerful rivals to the PM’s leadership; at least, ministers have their own channels to the party beyond government. Moreover, individual ministers may avoid the political cost of taking responsibility for emergency actions (Zahariadis, 2017).

Thus, separate to the clarity of the PM’s strategic lead is the response of Cabinet members to the management of the crisis, a second independent variable. This can be summarised as the legitimation of the role of the CRN within the Cabinet and it prompts a second hypothesis for our investigation:

H2: The more legitimate the CRN is viewed within the Cabinet, the more likely it is that the CRN will provide a centra lised response to the crisis.

While H1 covered the crisis lead (of the PM), H2 introduces legitimation as key to the effectiveness of the CRN. Building on the literature cited above, we see legitimation as being evidenced by:

1. The absence (or relative weakness) of an alternative crisis narrative or strategy to the one articulated by the CRN supported by senior Cabinet ministers

2. The absence (or relative weakness) of vocal criticism by ministers over the leadership or key personnel of the CRN

3. The absence of credible threats by Cabinet colleagues ministers to reject key agreements struck between the CRN and its international interlocutors

The degree to which these conditions do not hold will imperil the legitimacy of the CRN and its ability to centralise crisis decision-making.

Our two hypotheses cover distinct stimuli – the PM’s lead and the legitimation offered to the CRN by the Cabinet. It is possible to have the former, without the latter: where the PM gives a lead that is not endorsed by the Cabinet. The reverse appears less likely: though it may emerge, for example, if the Cabinet endorses a position adopted by the CRN (e.g., in the person of the finance minister) that contradicts that of the PM.

For both hypotheses, our dependent variable is the strength of the CRN in providing a centralised response to the crisis. We can gauge this strength by reference to:

1. Operational unity and stability within the CRN: lack of change in key personnel, effective and cohesive working methods, and a clarity in its line-management.

2. Embeddedness: information sharing and working partnership with the wider administration so that key parameters of its strategy can be uploaded/downloaded effectively

3. External acceptance: the CRN is recognised as a credible interlocutor with international negotiating partners

These agential and structural attributes allow us to gauge the role of the CRN. They denote a CRN that is providing strong, centralised decision-making; alternatively, their relative absence indicates a CRN that is in a weak position within the core executive.

The empirical investigation that follows is exploratory. The selection of our two cases intentionally follows a most similar comparative design. The scope conditions of both cases were similar: both prime ministers had to navigate an existential economic threat during the first six months of their tenure, before striking a deal with Greece’s creditors. Yet, the dependent variable (the strength of the CRN in leading the crisis response) differs across the two cases. We investigate how and why.

With a prime focus on leadership in the crisis, we draw upon lengthy semi-structured interviews with both prime ministers, their senior ministers and officials to examine the inner workings of their crisis management, as well as with senior EU and IMF officials for an ‘outsider’ view. This represents an exceptional quality of first-hand testimony: over seventy hours of interview material informs the analysis. Where appropriate our interviews are referenced in the text but we maintain their confidentiality by anonymising the contributors. In addition, we utilise published sources on Greece’s strategy in the crisis (Dendrinou & Varvitsioti, 2019; Papaconstantinou, 2016; Varoufakis, 2018), relevant government data and newspaper reports. In each instance, we base our conclusions on cross-referrals, triangulating our sources.

The crisis erupts: the government of George Papandreou

The clarity of the PM’s strategic lead

Papandreou was obliged to give a strategic lead for a crisis that he had not fully foreseen and that raised sensitive issues for him about Greece’s place in Europe. As such, his stance was reactive, but also adaptive to EU pressures. Critics would assert that his narrative lacked clarity and ceded too much to those inclined to punish Greece. Papandreou sought to be open and consensual to those around him, responding to events, probably exacerbating a sense that his lead lacked some coherence. Yet, there is no doubt that Papandreou led from the front and that he ultimately made and ‘owned’ all the key strategic decisions.

Papandreou’s crisis narrative stemmed from important personality traits. This reflected his core instincts of internationalism and cosmopolitanism and these militated against accepting that Greece should be left isolated or that it could not converge. Since the collapse of Lehman Brothers in September 2008, he had argued that the West needed a Keynesian-style stimulus, one coordinated internationally. Now, the specific Greek crisis was a ‘shared’ problem, requiring a collaborative international solution (Reuters, 11.2.2010). Indeed, as he put it more strongly later, ‘Greece has a problem’, but it is ‘not the problem’. Yet, Papandreou’s narrative was challenged, both domestically and internationally. His frequent references to corruption and clientelism as the key drivers of Greece’s troubles, left him vulnerable to domestic criticism that he was undermining the country’s credibility during an escalating crisis. Indeed, internationally, Greece was seen as the exception, rather than as a symptom of a wider eurozone malaise (Papadimitriou et al., 2019).

Domestic critics came to question Papandreou’s strategic objectives in dealing with the crisis. He was giving in too quickly to demands for ‘blood on the carpet’ as a means of placating the markets and Greece’s would-be creditors. His government did not produce a blueprint of its own on how to overcome the crisis – one in which the ordering of priorities and the sequencing of reforms could be presented to its international partners and sold to an increasingly restless domestic audience (Simitis, 2014, pp. 24–52). Was Greece facing a debt or a liquidity crisis? Was the IMF a desirable partner in the rescue efforts? Papandreou’s own ambivalence on these questions became a source of frustration by some of his Cabinet rivals (Interview #2 and #6).

Notwithstanding such misgivings, Papandreou’s task engagement in the complex negotiations over the Greek bailout was intense from the outset. He had earlier served as minister for over ten years, but his portfolios had never involved economics. On the domestic management of the crisis his close allies recognised a strength of mind at variance with his image of seemingly avoiding confrontation, though his critics accused him of lacking the managerial skills to coordinate the government’s response effectively (Interview #3 and #6). As PM, Papandreou forged a very close partnership with his finance minister, George Papakonstantinou, who shouldered the responsibility for the detailed policy management and negotiations with the EU and the IMF. Papakonstantinou became a very important conduit for how Papandreou understood the crisis and the PM reciprocated by providing a strong backing to his finance minister’s unprecedented spending cuts. As the crisis grew so did their personal and political rapport, which acquired a ‘groupthink’ character, detaching them gradually from the wider government (Interview #2).

There is no doubt that throughout the first six months of this premiership Papandreou made all the ‘big calls’. Faced with what allies saw as the opportunistic response of Opposition leader, Antonis Samaras, not to support the austerity measures enshrined in the first bailout deal, Papandreou chose not to gain party advantage by putting them to a parliamentary vote requiring a three-fifths majority (Interview #5). This would have required the Opposition to take a highly consequential stance. Instead, Papandreou let Samaras off-the-hook and enabled a counter-narrative to his strategy to grow. This choice, premised on Papandreou’s own conviction of putting ‘country before party’, would later on prove fatal for his ability to control the political fallout of the first bailout programme.

Cabinet legitimisation

The scale of PASOK’s electoral victory in 2009 (the largest of any party since 1981) afforded Papandreou significant freedom in choosing his Cabinet. Papaconstantinou’s arrival at the Finance Ministry was one of many ministerial appointments that were meant to signify a generational shift within PASOK. Indeed Papaconstantinou’s professional credentials were immaculate, but he had no prior government experience and his political capital relied heavily on the PM. Papandreou’s main rival, Evangelos Venizelos, who had challenged him for the leadership of the party in 2007, was appointed minister of defence, but his clout within the government in the early stages of the crisis remained limited (Interview #2 and #3). The party’s traditionalist (socialist) wing was represented by a relatively small group of ministers, headed by Louka Katseli, who became the minister of economy and Competitiveness. Despite her powerful portfolio, Katseli had good reasons to regard this appointment as a demotion. She had been the party’s economic supremo in opposition and had been widely tipped to take over the Finance Ministry and become Greece’s voice in the Eurogroup. Papandreou’s decision to appoint Papaconstantinou to the top job, left Katseli resentful (Interview #3 and 6). Yet, her misgivings never led her, at the time, to articulate a coherent alternative narrative on the handling of the crisis. Neither was her political standing high enough to rally a rival fraction of Cabinet colleagues against Papandreou’s charted course (Interview #5; Katseli, 2020).

For his part, Papandreou sought to increase buy-in amongst his colleagues through the frequent convening of the Cabinet, where open discussion was encouraged and Papaconstantinou was called upon to report on the progress of the rescue talks. A large number of ministerial ‘informals’ also took place in Papandreou’s office in Parliament in order to placate internal opposition and improve ministerial coordination (interview #1 and #3). Although the defence of ministerial ‘fiefdoms’ from spending cuts in these meetings was fierce, criticism against the overall strategy of the government and the senior leadership of the CRN was rather muted and rarely made public. Neither was there a credible threat by ministers to vote down key milestones of the government's rescue strategy. Hence, the Cabinet approved unanimously both the establishment of the EFSF (the funding vehicle of the Greek bailout) on 25 March and the terms of the MoU (the conditions attached to the bailout) between Greece and its creditors on 8 May 2010.

PASOK’s lead in the opinion polls throughout the first 6 months of 2010 might have preserved the unity of the government and PASOK’s parliamentary group, but in the background, the legitimacy of the CRN continued to erode. Famously, senior Cabinet ministers claimed that there was no time to read the provisions of the Memorandum they were called to implement (Ta Nea, 24.1.12). Indeed, many of Greece’s international commitments (even before the Memorandum was signed) were hastily legislated with little or no time for proper parliamentary scrutiny. The required secondary legislation meant a level of urgency and inter-ministerial coordination that was alien to the system. This was never going to be Papandreou’s forte or an area in which his inexperienced team was likely to succeed.

Problems of policy implementation both reflected and intensified underlying divisions within the Cabinet. Faced with defensive ministerial bureaucracies, Papaconstantinou’s lack of allies in the Cabinet became all the more evident and his political capital depleted fast (Interview #3 and #6). In just over a year after the signing of the first bailout, he was out of his job as finance minister. Soon afterwards, Papandreou would meet his own political demise, under unceremonious circumstances, as he faced rebuttal from his EU partners, and colleagues at home, over his proposal to put Greece’s imminent second bailout to a referendum.

The strength of the crisis response network

The CRN assembled by Papandreou to handle the early stages of the crisis harboured some contradictions in terms of its operational unity and stability. The key personnel in Athens was not much larger than a football squad, but its membership remained stable over time. It involved three small clusters of officials: straddling the Ministry of Finance, the General Accounting Office, and the Council of Economic Advisers (SOE). The academic credentials of its members were impressive, although their experience in working in government varied. Operating under extreme pressure, the team remained cohesive and was bound together by a regular pattern of meetings under Papaconstantinou and his deputy minister, Phillipos Sahinidis. Yet, Papandreou’s CRN also suffered notable absences. Relations between the government and the Bank of Greece were tense, with senior ministers scathing about its Governor, George Provopoulos, for not disclosing earlier the level of Greece’s fiscal derailment (Interview #3 and #4). Surprisingly, the country had also been without representation to the IMF since the autumn of 2009. By the time a new representative, Panagiotis Roumeliotis, was appointed in March 2010, discussions on the Greek bailout were well underway and relations with Roumeliotis became strained (Roumeliotis, 2012).

Beyond the key team members in Athens, Papandreou maintained regular contact with a wide range of influential experts and policy makers around the world. Some of them were formally appointed as his pro-bono advisers, including the former Italian finance minister, Tommaso Padoa-Schioppa and the recently-retired Vice-President of the ECB, Loukas Papademos. Others, like prominent US economists Larry Summers, Jeff Sachs, Joseph Stiglitz, as well as Richard Parker, also provided input into Papandreou’s thinking (Interview #1). Yet, the Greek PM’s impressive network of personal contacts was never integrated into a pattern of regular meetings or embedded meaningfully into the CRN’s business. When Papandreou inserted his foreign experts into Cabinet and party meetings it was met with resistance, with party stalwarts critical of US ‘tutors’ lacking knowledge of the EU (Interview #2 and #3). This pattern of informal and ad hoc interactions might have suited Papandreou’s apparent diffused management style and ‘big picture’ cognitive orientation, but it undermined the clarity of direction given to the CRN (Interview #6).

The issue of the CRN’s embeddedness into the wider administration went much deeper than the relative detachment of Papandreou’s foreign advisers. In fact, from the outset the capacity of the CRN to mobilise and discipline different government departments was limited (Interview #3 and #7). Papandreou did not maintain a functioning economic team within his PM Office (Interview #2). Senior appointments in the administration were delayed by the government’s new personnel selection process (opengov) and confusion over the delineation of portfolios. As the crisis raged, the Ministry of Finance was deprived of its top political echelons for weeks. Even when fully deployed (by early December 2009), the core team spearheading Greece’s CRN struggled to tame the wider bureaucracy (Interview #3 and #7). Papaconstantinou’s own political inexperience also made it harder to navigate foot-ragging by senior Cabinet colleagues. Although these tensions simmered in the background during the very early stages of the crisis, they would later have devastating consequences once the implementation of Greece’s commitments to its creditors was put under the microscope.

Yet, despite its shortcomings, Papandreou’s CRN enjoyed a remarkable unity of purpose that withstood the tremendous challenges it confronted (Papaconstantinou, 2016). The possibility of Greece breaking ranks with its EU partners was explored in early 2010, but it was quickly and unanimously dismissed, leaving the government totally focused on delivering a solution from ‘within the system’ (Interview #2 and #7). The strong political partnership between Papandreou and Papaconstantinou also enhanced the external acceptance of the CRN in the eyes of Greece's creditors (Interview #19). Such an acceptance complemented Papandreou’s own narrative in his famous ‘Kastelorizo address’ when he presented the first bailout as a mutually agreed plan to allow Greece to return to the markets within two years (Financial Times 23.4.10). The truth, however, was very different. The fiscal targets and associated actions were put together hurriedly and the Greek government’s ability to shape the agenda was rather minimal – the result of evident power asymmetries in Greece’s relationship with its creditors, but also its own inability to feed into the process alternatives drawn from within the government machine.

Summary of results

In sum, for H1 on the clarity of the PM’s strategic lead, Papandreou’s record was mixed. Although his task engagement remained high throughout this period, his crisis narrative of Greece’s predicament as a ‘shared problem’ was ambivalent and was challenged both at home and abroad. His strategic objectives also suffered major weaknesses. Although his immediate team was united, it remained rather reactive to a rapidly changing external environment. No ‘local’ plan for exiting the crisis was ever produced and the operationalisation of international advice received by Papandreou was erratic.

For H2 on the legitimisation of the CRN, members of Papandreou’s close team faced some (mostly private) criticism by adversaries in Cabinet, particularly with regard to their apparent ‘groupthink’ on how the roots and remedies of the crisis were conceived. However, such opposition never materialised into an alternative crisis narrative or credible threats to veto key agreements with the creditors. Yet, the buy-in from Cabinet colleagues remained muted and conditional. This detachment was later to define Papandreou’s fate as PM.

Papandreou’s CRN had the remarkable potential to lead a highly centralised crisis response (our dependent variable). It was united and stable in its senior membership. It also enjoyed high levels of external acceptance. Yet, over time, its operational stability was impacted by shifts in Papandreou’s crisis narrative and a weakening legitimation within the Cabinet, while it also failed to sustain its embeddedness within the wider administration. In this context, its centralising effects were sketchy, conditional and ultimately short lived.

The crisis redefined: the government of Alexis Tsipras

The clarity of the PM’s strategic lead

Alexis Tsipras rose to power on a promise to unilaterally end (‘with a single law’ as he put it) Greece’s bailout programme and reverse austerity in the face of a domestic ‘humanitarian crisis’. His crisis narrative was simplistic, but electorally highly effective. He was an expert purveyor of the need for Greece to fight back against the ‘neo-liberalism’ that had taken hold at the EU level and relieve the suffering of the people. The rhetoric of fighting back appealed to the domestic sense of victimhood, whose seeds had been planted with the criticisms of Papandreou’s government. Tsipras lambasted the old political class for being a corrupt ‘kleptocracy’: a catharsis was needed. It is against this background that his party, SYRIZA, had seen its vote rise from just 4.6 per cent in 2009 to 36.3 per cent in January 2015.

With no prior government experience, little international exposure and a very modest grasp of economics (Interview #17), Tsipras had to look elsewhere for deeper economic thinking. In a ‘high risk, high return’ political investment (Interview #10), the new PM opted to make a popular, independent economist – Yanis Varoufakis – his minister of finance, having travelled personally to Texas to persuade him to join his team (Interview #8, #17). Varoufakis’ appointment had important repercussions for the internal politics of the new government (see below), but more importantly introduced much uncertainty about the PM’s own strategic objectives (Interview #10 and #13). On the surface, the two men shared a belief that, if properly challenged, Greece’s creditors would eventually back down. Yet, the fall-back position remained unclear: the strategic objectives had not been shared.

Varoufakis refers to Tsipras and himself establishing a ‘covenant’ in November 2014, laying down the terms of the SYRIZA crisis strategy (2018, pp. 100–103). The terms of this personal understanding are now mired in post-hoc political conflict. No-one recalls the term ‘covenant’ being used. In Varoufakis’ version, the objective was a new EU deal; if offered a bad deal, Greece should leave the eurozone and re-introduce the drachma currency – in part, to show that Greece would not be cowed. Thus, it was not a ‘game of chicken’: Greece should show it was willing to go over the edge. In government, however, Varoufakis’ position seemed not so clearly defined and nor did he present a plan for such a follow-through (Interview #12, #15 and #10).

Yet, for Tsipras, and most of his senior SYRIZA colleagues, while it was useful to bluff, the underlying commitment was to remain in the eurozone (Interview #8 and #13). Indeed, given the very high public support for doing so, this had been part of SYRIZA’s appeal: to have its ‘cake and eat it’, according to its critics. What Tsipras had seemingly not prepared for was the EU contemplating ‘GREXIT’, an outcome he rejected. There is some ambiguity then about Tsipras’ understanding: in what he had expected to happen and the conditions under which he would shift position. His reluctant acceptance of a ‘poor’ deal with his Eurozone partners in July 2015 – keeping Greece in the ‘euro’ – may be seen in this light. It became known in Greece as his ‘somersault’ (kolotoumba).

Alongside this, Tsipras’ task engagement sustained a number of contradictions, combining a sense of detachment with ruthless calculation. Despite his very high political capital, the PM ceded much ground to Varoufakis and to his deputy prime minister and party stalwart, Giannis Dragasakis. The former became a key interface with Greece’s creditors whilst the latter retained significant influence over the government’s economic strategy and senior appointments. Such choices reflected Tsipras’ own self-awareness as the country’s youngest ever PM, but also a deeper conviction to collective decision-making and a ‘left-wing-way’ of doing things (Interview #8 and #12). SYRIZA’s internal party culture was not conducive to centralised leadership, but Tsipras had proven to be a shrewd tactician who had risen to power by out-manoeuvring party opponents. He was also a quick learner.

In the early stages of his premiership Tsipras let Varoufakis make the running, but he soon realised that his bet on Yanis had come unstuck (Interview #8, #9 and #15). In April 2015 he moved to draw back his finance minister from the EU negotiations when the latter stood in the way of a deal. This was a ‘big call’ that led later to another – the acceptance of the third bailout which he had so vehemently opposed at the beginning of his tenure.

Cabinet legitimisation

In the aftermath of his electoral victory in January 2015, Tsipras had to navigate a delicate political terrain. His party lacked an overall majority in Parliament and his junior coalition partner, the populist Independent Greeks (ANEL), came from the opposite end of the political spectrum. The internal politics of SYRIZA were also a constraining factor. For years the party had operated as a loose coalition of different fractions, ranging from the edges of social democracy, to radical left-wing internationalism, all the way to neo-communist EU rejectionism. The latter (the so-called ‘Left Platform’) commanded significant power within the party and controlled key ministries in the government. The common denominator of this diverse coalition of ministers was a strong anti-bailout sentiment alongside a recognition that Tsipras’ own popularity had been instrumental in the party’s meteoric rise to power. Yet, their views on how far the new government should push the confrontation with its creditors varied significantly.

Tsipras’ own ambiguity on this issue made it difficult for any faction within the Cabinet to develop an alternative crisis narrative, be it one of an immediate rupture with the creditors or a speedy compromise. The PM’s management style deterred direct confrontation. Tsipras was approachable inclusive, and informal (Interviews #10, #13, #15). Although the mobilisation of his full Cabinet was much less than that under Papandreou, the PM did hold twice-weekly meetings with a group of senior ministers and party officials to discuss the progress of negotiations (Interview #9 and #12). He also played off his key allies: he gradually brought in Euclid Tsakalotos, to balance against Varoufakis; a significant supervisory role was maintained for Dragasakis, to favour the party’s traditionalists; and, to secure the coalition government, Panos Kamenos (the ANEL party leader) was kept close to the centre of discussions.

With its own internal divisions (see below), the CRN's currency in the Cabinet fluctuated widely. During the early stages of the negotiations with the creditors, Varoufakis’ antics alienated moderate ministers, particularly those close to Dragasakis (Interview #10, #12, #13). At the same time, Varoufakis, as an ‘outsider’, was never fully trusted by Left Platform ministers, most notably Panagiotis Lafazanis, the powerful minister of productive reconstruction (Interview #17). Despite the intermittently vocal criticism against senior members of the CRN, Tsipras and Varoufakis were able to push through Cabinet a short-term extension of the second bailout in February 2015. Later on, the replacement of Varoufakis by Tsakalotos as chief negotiator improved the standing of the CRN among moderate ministers, but the prospect of signing up to a new long-term austerity programme was met with explicit threats of resignation by a number of Left Platform ministers who also encouraged their ministerial bureaucracies not to cooperate with Troika officials.

Faced with raging disagreements within his Cabinet, Tsipras’ instinct was to ‘manage’, rather than lead from the front. Such ‘creative ambiguity’ over his own preferences might have afforded him more room for short-term manoeuvring, but it also came at a cost as the PM simply pushed conflict further down the road. When, in July 2015, Tsipras and his reconstituted CRN made up their minds to strike a deal with Greece’s creditors, both his government and his party were too fragile to survive the u-turn, without further political steps being taken. Tsipras called elections in September 2015 and despite a high-profile split within the party, he was successful in renewing his governing mandate, demonstrating remarkable resilience in the face of the crisis. Tsipras could feel vindicated, but in the process, his party and government had changed, as had his own lead.

The strength of the crisis response network

The appointment of Varoufakis at the helm of the Finance Ministry might have helped the PM to internationalise his case against austerity, but it also brought serious challenges to the make-up and nodality of the CRN at home. Varoufakis had no prior experience of government business and was a political outsider who had lived away from Greece for the preceding three years (Interview #17). His perceived expertise on ‘strategy’ was not matched with a network of collaborators with EU policy know-how. His relations with key insiders of the previous order, including with Bank of Greece Governor Yiannis Stournaras, were openly hostile. Moreover, Varoufakis’s influence over appointments within the Greek bureaucracy was very limited (Varoufakis, 2018, pp. 118–123, Interview #17). His own network of personal advisors was rather small and suffered an almost complete lack of embeddedness in the administrative apparatus they meant to mobilise. The majority of them were foreign (banker Glenn Kim and Lazzard-based consultants, Daniel Cohen and Matthieu Pigasse), except Elena Panaritis, a former IMF official, who had spent most of her working career outside Greece. Notably, the person who was entrusted with the top-secret plan to devise a parallel payment system in case Greece crushed out of the Eurozone was James Galbraith, a professor at the University of Texas, with a rudimentary knowledge of the Greek administration (Interview #16).

The operational unity and stability of CRN also suffered from intense infighting and frequent turnover of senior personnel. Of all political appointees in the Ministry of Finance, only Nicos Theoharakis, the Secretary General of Fiscal Policy, was a personal choice of Varoufakis. Crucially, the finance minister had no say over the appointment of the President of SOE, George Chouliarakis, whose remit included attending Euro Working Group (EWG) meetings where he managed to establish a good working relationship with his EU counterparts. Chouliarakis maintained direct lines of communication to both the PM and Dragasakis, but his relations with Varoufakis were fractious from the start, with the two men hardly on speaking terms (interview #11 and #17). In early March 2015, Varoufakis succeeded in removing Chouliarakis as the Greek representative to the EWG, replacing him with his trusted ally, Theoharakis, but this proved short-lived.

Behind the dysfunctionalities of the Greek negotiating team laid broader issues of approach and strategy. Varoufakis and Theocharakis were determined not to discuss further austerity until the creditors committed to debt restructuring measures (Interview #14 and #17). Their attitude was often elusive and abrasive, creating an impression to their counterparts that they were poorly prepared (Dendrinou & Varvitsioti, 2019, pp. 105–107, Interview #19). The external acceptance of the Greek CRN under Varoufakis was further undermined by the complex political landscape in Athens. Varoufakis’ relations with Dragasakis and Spyros Sagias (the Secretary General of the Government who coordinated technical discussions with mid-level Troika officials in Athens) grew increasingly strained (Interview #13 and #16). Others within the Greek team, with Chouliarakis most prominent among them, argued for a more accommodating approach towards the creditors, fearing that Greece was sleepwalking towards a disastrous Eurozone Grexit. With no direct access to Tsipras and his inner-circle at Maximou, senior Troika officials struggled to ascertain who was the ‘true voice’ of the Greek government (Interview #19). They looked to the PM for answers but Tsipras was oscillating between competing demands without making his mind up (Interview #11).

In spring 2015, the replacement of Varoufakis by Tsakalotos as chief negotiator and the reinstatement of Chouliarakis as Greece’s representative at the EWG improved the external interface of the Greek CRN with the creditors. However, ambiguity over strategy persisted in Athens. Despite his apparent demotion, Varoufakis retained his ministerial post and continued to attend Eurogroup meetings, but his ‘cohabitation’ with the government’s newly appointed negotiators was both uncomfortable and dysfunctional (Interview #10, #11 and #17). Neither was there consensus within the Greek CRN on how to respond to what they perceived as increasing intransigence on behalf of the creditors. The decision to put a proposed third bailout package to a referendum in late June 2015 was seen by Tsipras and his closest supporters as a win-win situation (Interview #8, #9 and #15). Popular endorsement of the new deal would help detoxify the domestic political scene, whereas a rejection would have strengthened the hand of the Greek government in a subsequent re-negotiation. Others within the CRN, however, feared that the referendum risked destabilising further the economy with no tangible benefit (Interview #11 and #13).

Tsipras’ own campaigning during the referendum also raised doubts about his preferred outcome. Either way, the meaning of the overwhelming ‘No’ vote of the referendum received starkly contrasting responses within SYRIZA circles. Although Tsipras’ reformulated CRN remained sufficiently united to strike a third bailout deal with the creditors in July 2015, the fallout of this agreement was to have severe consequences for unity of the party and the government, as noted earlier.

Summary of results

In sum, for H1, on the clarity of the PM’s strategic lead, Tsipras’ record sustained a number of weaknesses. While his crisis narrative was stark and stable, it lacked an informed understanding of the international context and the constraints on exit options. His strategic objectives were unclear and incomplete. Crucially, Tsipras failed to give clarity on his ‘Plan B’. Even having side-lined Varoufakis, Tsipras’ direction to his new CRN remained under-elaborated until the eleventh hour. His ambiguity reflected a pattern of task engagement which, although activist, was ultimately driven by political calculation, keeping him one step back, rather than leading from the front.

These failings were compounded by the legitimation afforded within the Cabinet (H2). Tsipras’ ambiguity had limited the scope for alternative crisis narratives to emerge within the Cabinet, although the CRN was often confronted with open criticism and explicit threats of defection by senior ministers if it overstepped the mark during the negotiations. Eventually, such threats materialised in an open split within SYRIZA when Tsipras agreed the terms of a third bailout programme.

The potential of the CRN to provide a centralised response – our dependent variable – during the 2015 crisis was severely hampered from the outset. The operational unity and stability of the CRN became victim of competing fractions of the government, with neither Tsipras nor Varoufakis being able to provide it with a clear sense of direction. Political inexperience, personnel changes and ministerial infighting also undermined the CRN’s embeddedness within the wider administration. Externally, the CRN acceptance as a credible interlocutor was minimal, certainly under Varoufakis’ tenure. Of greatest consequence for the CRN‘s weakness was undoubtedly Tsipras’ strategic ambiguity. Given his electoral standing, he had political capital to deploy within the government but chose not to give a more clearly elaborated lead. This meant the CRN was soon a centrifugal force in government processes before it became part of intra-government contestation.

Conclusions and implications

In this paper, we explored whether a crisis may centralise decision-making around the PM by virtue of a crisis response network. Our cases concerned a core executive (Greece) marked by long-term fragmentation, but one that appears to have achieved centralised crisis management in certain past crises. We noted evidence in the comparative literature of the euro-crisis having centralising effects in other EU member states. Drawing on leadership studies and the literature on crisis management, our first hypothesis (H1) was that the clarity of the PM’s strategic lead was a crucial factor in enabling the CRN to lead the response to the crisis, and hence centralise the relevant decision-making. Our second hypothesis (H2) was that this was necessary, but not sufficient in a core executive setting: the strength of the CRN would also depend on its legitimation within the Cabinet.

We found variation on our independent variables (the PM’s lead and legitimation within the Cabinet) as well as our dependent variable (the strength of the CRN in providing a centralised response to the crisis). In our first case, under Papandreou, the CRN began with a relatively strong centralised response, but this became eroded with a more inconsistent PM crisis narrative and opposition emerging within the Cabinet. In the second, under Tsipras, the CRN fared poorly throughout: it was divided and divisive as a result of the PM’s incomplete strategic objectives and an increasingly restless Cabinet. Within the Cabinet frame, the relationship between the PM and the finance minister offered little help to centralisation (cf. Parrado, 2012; Schoeller, 2019): that between Tsipras and Varoufakis proved highly dysfunctional to coordination, while Papandreou’s closeness to Papaconstantinou isolated them from their colleagues.

The consequential nature of the lack of clarity in the PM’s lead (shown in different ways in both cases) highlights not only the importance of agency in crisis management, but also the potential mutability of arrangements within the core executive. Crucially, what was missing in these cases was evident in others in the same Greek setting: the clarity of lead given by Karamanlis in 1974; Papandreou in 1987; and Mitsotakis in 2019. All crises are distinct – that of 1974 may be termed ‘political’, that of 2019 – ‘social’ in nature – whereas our cases here were ‘economic’. It is arguable whether government actors find it more difficult to share a common mission in an economic crisis, sensitive to issues of personal political cost (Zahariadis, 2017). Young found that foreign policy crises could strengthen the ability of a US president to act but found no evidence of the same from economic crises, possibly due to the lack of what she terms a ‘rally round the flag’ effect (2013, p. 350).

This paper has made a number of contributions. Firstly, it has elaborated a conceptual frame, drawing on diverse literatures, by which to examine the ‘centralisation in a crisis’ proposition within a core executive setting. To do so, it has introduced the notion of a CRN and elaborated likely conditions for its empowerment (the PM’s strategic lead and the legitimation of the CRN within the Cabinet). We tested two relevant hypotheses to differentiate the outcomes. This is a frame that may be utilised in different contexts and via different methodologies and sources. The addition of discourse analyses of the (public) communication of crisis narratives or strategies, for example, could be a useful complement to evidence of internal government interactions.

Secondly, in gathering extensive personal testimonies by each of the key actors involved, we have utilised first-hand accounts of how the Greek crisis was managed within the government, extending the literatures on the external negotiations of the crisis. The focus on a single country has allowed a greater depth of investigation in an institutional context in which there is a frequent generic sensitivity about researcher access.

Thirdly, as more comparative research becomes available on how ‘bailout’ countries managed the pressures of their respective programmes (see, for example, Moury et al., 2021), a fertile research agenda opens up for the systematic study of both how different core executive types have developed their crisis management, with consequences for the content and ownership of their ‘rescue’ programmes. Greece was probably unique in having a (second) bailout stipulation concerning the greater coordination, within government, of the adjustment measures: a result, no doubt, of how external actors viewed its compliance and delivery thus far.

We recognise a number of limitations in our study. Given our focus on ‘crisis peaks’, we were not able to assess patterns of (de)centralisation in the later stages of the respective premierships – when the focus of the crisis switched to reform implementation. Here, the crisis pressures of time, threat and uncertainty are reconfigured, under the bail-out monitoring, and may have distinct outcomes. t’Hart et al. (1993) suggested that in ‘routine-oriented’ (rather than ‘pragmatic’) bureaucracies, when the time-pressure is seen as high, the resultant decision-making mode will be paralysis; when the time-pressure abates, there will be centralisation of decision-making. In our focus, on two critical periods, we did not find paralysis, but neither substantive or enduring centralisation. We were not able to engage with evidence of policy learning or lasting institutional change in core executive governance across the four different governments of the Greek crisis, however: a longer-time frame could assess these effects.

Understanding what crisis conditions enable a centralisation and coordination of responses will continue to be salient to both international and domestic policy actors, as well as academic literatures.

Disclosure statement

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

Additional information

Notes on contributors

Kevin Featherstone

Kevin Featherstone is Professorial Fellow at the London School of Economics.

Dimitris Papadimitriou

Dimitris Papadimitriou is Professor of Political Science at the University of Manchester.

Notes

1 The term ‘Cabinet’ is used to describe the Ministerial Council (Υπουργικό Συμβούλιο).

Appendix

List of interviews

Table

Number	Position	Location	Date
1	Prime Minister	Athens	20.2.20
2	Senior Minister	Athens	20.2.20
3	Senior Minister	Athens	24.2.20
4	Senior Adviser	Athens	21.2.20
5	Senior Minister	Athens	21.2.20
6	Senior Minister	Athens	23.2.20
7	Vice Minister	Athens	21.2.20
8	Prime Minister	Zoom	11.11.20
9	Senior member of the PM Office	Athens	4.11.20
10	Senior Minister	Athens	2.11.20
11	Senior Adviser	Zoom	17.9.20
12	Senior Minister	Zoom	2.11.20
13	Senior Minister	Athens	3.11.20
14	Senior government official	Zoom	16.11.20
15	Senior Minister	Athens	5.11.20
16	Senior government official	Athens	4.11.20
17	Senior Minister	Zoom	7.12.20
18	Senior Troika Official	Zoom	9.12.22
19	Senior Troika Official	Zoom	5.10.22