Un-solvable crises? Differential implementation and transboundary crisis management in the EU

Abstract

Much has been said about how crises in the EU create disintegration or differentiation pressures. Considerable attention has been paid to EU crisis governance mechanisms. Yet, less attention has been paid to the anticipation of effects of differentiated implementation on transboundary crisis management regimes. This article asks how differential policy integration accommodates the anticipation of differential implementation through institutional choices in transboundary crisis management regimes. Concerns about the consequences of national customisation influence the way in which transboundary crisis management regimes develop in terms of allocation of authority and constraints on member state discretion. The paper compares EU transboundary crisis regimes in four sectors: banking, electricity, youth unemployment, and invasive alien species. Concerns with ongoing differential implementation of transboundary crisis management generate further inevitable tensions in governance systems, leading to continued contestation over institutional arrangements.

Keywords:

• Crisis
• multilevel governance
• EU
• implementation
• differentiation
• customisation

Over the past two decades, transboundary crises such as the financial, sovereign debt, refugee and Covid-19 crises have generated intensive debates about the shape and the effectiveness of the European Union (EU)’s multilevel governance structure. This has led many observers to declare the EU in crisis, whether because of the combination of transboundary crises or because of the challenges they raise to its institutions and governance system(s) (Börzel and Risse 2018; Dinan et al. 2017; Falkner 2016; Laffan 2016; Roos and Schade 2023; Saurugger 2014; Tosun et al. 2014). Some member states’ recurrent backsliding on liberal democratic constitutional norms has created institutional crises and threats to EU values (Bakke and Sitter 2022). These dynamics have also pointed to perceived flaws in the institutional architecture of the EU (Chalmers et al. 2016), where neither central authority nor member states can agree on joint crisis resolution, sometimes leading to further differentiation (Leruth 2020).

In response, the EU has developed notable crisis management capacities since the start of the twenty first century. It has become a crisis manager involving in-house or agency-based competencies and developing cooperation among member states (Blondin and Boin 2020; Boin et al. 2013). It has also become involved in regulating national crisis management policies through multilevel arrangements spanning across domains and levels. The involvement of the EU in a traditional core state power, such as crisis management, raises important questions about governance and the emerging nature of the EU polity in general, and the relationship between policy design and implementation in particular (Genschel and Jachtenfuchs 2013). Indeed, responding to transboundary crisis has been identified as the modern-day equivalent of military threats for the development of the EU’s (regulatory) state-type capacities (Freudlsperger and Schimmelfennig 2022). Yet, most of the literature on the EU and crises has focused thus far on EU level politics and policies (Brack and Gürkan 2020, Ferrara and Kriesi 2022), leaving aside questions of implementation, relations between member states and the EU, and more broadly, how dynamics of multilevel governance affect the EU’s ability to solve crises.

Addressing these dynamics of the rise of crisis management in the EU, Jones et al. (2021) suggest that crisis politics in the EU follow a ‘failing forward’ pattern in that initial (liberal intergovernmental) lowest common denominator responses generate further political and policy instability because of their incompleteness, and are followed up by enhanced EU-institutional (neo-functional) strengthening, which, turn, generate further integration shortcomings. This article offers some nuance by pointing to the importance of expected differential policy implementation in the formulation of differential policy integration, and how ongoing concern over differential implementation encourages continued contestation over policy approach.

This article focuses on the role of the anticipation of differential implementation in EU transboundary crisis management regimes. It extends Jones’ ‘failing forward’ argument by taking into account the role of concerns over differential implementation in shaping ex-ante the design of EU policy regimes. On the one hand, the (inevitable) ‘customisation’ at the member state level that occurs during the implementation of EU-level policies is regarded as an efficient way to secure legitimacy (Thomann 2018; Zhelyazkova and Thomann 2022). On the other hand, however, member states’ customisation is a potential and significant source for transboundary crisis itself. As a consequence, transboundary crisis management regimes are formulated in view of anticipated customisation, thereby seeking to mitigate potential sources of crisis emerging from differential implementation (Thomann and Sager 2017a, 2017b). In responding to the anticipation of customisation, transboundary crisis management regimes vary in the degree to which they (1) allow for a shift of authority to the EU level in terms of standard-setting, and (2) seek to enforce a high degree of prescriptiveness. Second, anticipation and perception of differential implementation at the member state level represents a ‘failing forward’ factor that influences ongoing contestation over transboundary crisis management regimes. The initial choice of transboundary crisis management regime also points to tensions and paradoxes that generate demand for further reform: inbuilt contestations and tensions shape the path of further institutional reform. The reform path, however, does not necessarily follow the kind of institutional strengthening expected by ‘failing forward’ accounts, but rather leads to further differentiation. The inevitable and continuous contestation over appropriate transboundary crisis management regimes renders both crises and crisis management ultimately ‘un-solvable’.

In view of this special issue’s interest in differential implementation, this paper therefore contributes to two key discussions: first, it highlights the link between anticipated differentiated policy implementation and resultant differential policy integration, and second, it highlights how the scope for differential implementation is regarded as a key dimension for generating institutional capacity for dealing with transboundary crises (Zhelyazkova et al. 2023).

Our analysis relies on a comparative qualitative case study across domains and levels of decision making. We define crisis management regimes as the set of (EU) policies, including instruments that seek to substantively address (sources of) crisis and that allocate authority across different actors across levels of governance (see also Hood et al. 2001). This paper examines four different crisis management regimes in the financial, infrastructure, social and environmental domains: banking, electricity transmission networks, youth unemployment and invasive alien species. These four domains are associated with crises involving urgency, threat, and uncertainty. Across the four domains, regimes have emerged to address transboundary crises through the adoption of EU legislation seeking to regulate national and cross-border crisis management. The four domains also present considerable differences in terms of institutional arrangements (in terms of authority and prescriptiveness), salience or distribution of actors’ resources (problem-solving capacity).

In the next section, we identify four ways of organising multilevel transboundary crisis management regimes: governance by delegation, centralisation, benchmarking, and decentring. These regimes seek to manage the interaction between European institutions and member states in the governing of transboundary crises whilst minimising the risks of member state customisation for the overall regime functioning. As noted, these initial choices are inherently unstable, given tensions over degrees of centralisation and prescriptiveness. Ongoing tensions point to instabilities in multilevel governance generated by actual crisis and ongoing concerns regarding differential implementation. We subsequently apply this framework to transboundary crisis management in the EU to our four cases, highlighting how conflicts over differential implementation concerns define structural tensions in each of these domains. Transboundary crisis management therefore needs to be understood as a continuous process of contestation which is characterised by a continuous ‘hunting around’ for institutional arrangements across the two dimensions of the centralisation of institutional authority and prescriptiveness.

Multilevel governance, prescriptiveness and authority in implementation

The literature on EU multilevel governance and implementation offers a useful starting point to explore how member states and European institutions seek to coordinate transboundary crisis management both horizontally and vertically (Börzel et al. 2012; Scholten 2017). This literature has highlighted the importance of studying actual rather than mere legal transposition, as well as the need to develop a better understanding of administrative capacities (Hartlapp and Heidbreder 2018; Heidbreder 2014). The focus on transposition and compliance has usually been limited to national authorities and aimed to show the adaptation of EU policies at national level as well as problem-solving capacity issues stemming from dispersed implementation (Benz et al. 2016; Gollata and Newig 2017). Research has pointed to diverse implementation patterns, whether it is due to the growing complexification of governance arrangements at the EU level, national political choices or varied national and sub-national administrative capacities. For example, Scholten (2017) noted that the practice and study of enforcement in a multilevel context is complexified by the direct ties between EU and national-level regulators, the proliferation of cross-national enforcement networks (Martinsen et al. 2022), and the direct oversight by EU-level authorities in some domains (see also Yesilkagit 2011, Yesilkagit and Jordana 2022).

So-called customisation of EU law provides for potential additional legitimacy and effectiveness as it resolves domestic conflicts and enables member states to meet their supranational commitments (Thomann 2018). Such customisation is particularly critical in the area of crisis management, a traditional domain of core state power (Genschel and Jachtenfuchs 2013, 2016) in which the authority to define when and how to declare, manage and end ‘states of emergency is still mainly located at the national level. A reliance on customisation to translate EU arrangements to coordinate crisis management may be particularly important to secure member state support and obtain an overall degree of legitimacy.

Transboundary crises, and the way jurisdictions prepare for and respond to these crises, represent a fundamental challenge for national crisis management systems as authority to resolve crises is distributed across borders, and/or to the next level ‘up’ – the EU. It is not just a matter of sharing core state powers that is of central importance for determining authority, but also the actual management of crises. For example, it matters who decides whose electricity should be switched off or whose investments should not be ‘bailed out’ in times of crisis. National elected leaders rather than EU institutions are likely to be held accountable in the eyes of voters. Therefore, member states’ concerns reflect not just how crisis management regimes evolve at the EU level so as to reduce their potential exposure to transboundary crises (see Freudlsperger and Schimmelfennig 2022), but also the implementation of these regimes and their consequences for their authority. At the same time, their own crisis management capacity (and legitimacy as sovereign power) critically depends on the national capacities of other jurisdictions to respond to transboundary crises in ways envisaged by shared policy regimes.

These central tensions in transboundary crisis management point to the centrality of customisation. Member states’ customisation represents a distinct risk to the EU and member states. Indeed, differential implementation may reduce the overall effectiveness of a transboundary crisis management regime as financial, energy or environmental crises can spill over or cascade – often rapidly and unpredictably – into other jurisdictions with often destabilising effects on other national regulatory systems. It also creates a risk to the EU as a whole, whether through the accumulation of crises threatening the legitimacy of the EU or because of threats to the integrity of the EU policy system.

Thus, the level of concern over the likelihood and impact of problems arising from national implementation is likely to shape ex ante institutional choices in EU transboundary crisis management regimes to limit the scope for differential implementation. The initial formulation of transboundary crisis management regimes will pay considerable attention to the expectation of differential implementation and adopt institutional devices that reduce the scope for ‘customisation’ that is anticipated to become a source for a transboundary crisis itself. Moreover, the contested nature of the emerging varieties of transboundary crisis management regimes generates continued tensions, leading to further differentiation. In other words, concerns about discretionary customisation play a central role in the variety of transboundary crisis management regimes and generates, over time, further differentiation and variety.

Two dimensions are central to the ways in which the scope for customisation (and, thus, differential implementation) can be curtailed: the level of authority in decision making and the degree to which administrative activities across levels of governance are prescribed (Hooghe and Marks 2003). Taken together, these two dimensions point to four distinct multilevel transboundary crisis management regimes. First, centralisation denotes a move towards full EU-level competence with specific administrative capacities and resources to address particular policy issues. Second, delegation relies on centrally legally binding provisions, but where there is a reliance on transposition by member states. Third, benchmarking relies on non-binding indicators and recommendations with a low degree of prescriptive demands for information or required policy compliance. Fourth, decentred governance reflects limited prescriptiveness, with only centralised binding codes and guidelines to guide behaviours and enable information flows, typically developed in the context of ‘co-regulation’ with significant industry involvement. The notion of ‘decentred’ is taken from the regulation literature to denote the exercise of authority beyond the confines of the state, such as different types of self- and co-regulation as well as transnational governance arrangements.

The four modes of governance incorporate their own built-in tensions that shape subsequent regime dynamics. Tensions emerge due to ongoing concerns with the location of authority and the degree of prescribed discretion. Acute crisis will lead to calls for change. Similarly ongoing conflicts between member states as to what constitutes ‘discretionary compliance’ and what is ‘non-compliance’ are likely to shape ongoing contestation. These tensions are illustrated in Table 1. Centralisation can be expected to suffer from information asymmetry problems, concerns over gaming and cheating by subordinate actors, limited processing capacity at the centre and concerns over ‘one size fits all’ responses to specific and diverse policy contexts. Delegation is likely to attract concerns about diverse transposition and compliance across member states. Benchmarking is likely to suffer from limited policy responsiveness at the national level if incentives for compliance with benchmarks are not aligned. Finally, decentring is shaped by industry interests and might be said to lack ‘hierarchical’ qualities in that no ultimate decision-making authority exists.

Table 1. Modes of multilevel governance in the EU.

		Authority
		Decentralised	Centralised
Prescriptiveness	High	Governance by delegation	Governance by centralisation
		Centralised standards, decentralised information-gathering and decentralised behaviour-modification	EU-level agencies shape standards, oversee national agencies and engage in information-gathering and behaviour modification
		Advantages: adjusts to diversity across Member states while encouraging a consistency in goals	Advantages: establishes leadership responsibilities through formal authority
		Pathologies: limited interest in Member states to comply	Pathologies: information asymmetries and dealing with diversity
	Low	Governance by benchmarking	Governance by decentring
		Centralised indicators and recommendations, reliance in peer-review and decentralised information gathering and behaviour modification	Centralised, but co-produced binding codes/guidelines, decentralised behaviour-modification and information-gathering
		Advantages: limited adjustment costs reflect contexts of high diversity	Advantages: joint working across different stakeholders
		Pathologies: little coordinative capacity and potential problem of actual compliance with the intent of regimes	Pathologies: lack of central leadership, high differentiation

We expect that the more member states are concerned with the uniform interpretation of joint provisions, the further they will seek to prescribe implementation efforts. Such concerns relate to the likelihood of spill-overs of any given crisis, either on member states or EU institutions. Low prescriptiveness is likely to reflect, in contrast, a low level of concern with how other member states conduct their affairs, either because of the limited likelihood and impact of differential implementation activities becoming a source of crisis, or because integration in a given domain does not cause the kind of pressure for prescribing similar approaches. Conflicts over the level of authority are also likely to be influenced by considerations of member states’ diverse implementation structures.

Regarding the European Commission, the EU body tasked with oversight the implementation of EU policies, we should not expect a uniform demand to move to a ‘centralised authority’ model of transboundary crisis management across all domains. Rather, considerations regarding the reputational implications of having to police member states’ implementation efforts and the need to maintain political coalitions are likely to lead the Commission to advocate diverse transboundary crisis regimes depending on particular situational constellations and ‘crisis profile’.

At the same time, perceptions regarding the risks of differential implementation as a source of transboundary crisis are unlikely to be shared by all actors. Consequently, we expect that ongoing disagreement over degrees of centralisation and prescriptiveness will lead to continued conflict over how to organise transboundary crisis management and, given reactive policy adjustments across levels of government, to ever further differentiation. The ‘failing forward’ account by Jones et al. (2021) notes how institutional gaps and frictions caused by initial intergovernmental bargains among reluctant member states trigger calls for further ‘deepening’ of EU integration that is perpetually incomplete. According to our account, the initial institutional choices seeking to curtail the risks of customisation are a source of policy instability. However, the outcome is a multilevel governance system structurally rife with tensions in view of experiences of acute crises as well as evolving expectations as to assumed effects of customisation on overall regime capacity to manage transboundary risks. Transboundary crisis management regimes therefore continuously evolve in response to irresolvable conflicts over levels of authority and prescriptiveness.

Case selection and research

This study is based on a qualitative comparative analysis of multilevel transboundary crisis management in the EU to identify variations in the mode of governance identified above. Comparison is both on a country and domain basis to capture variations in the distribution of crisis management powers and capacities and national responses.

We focus on four domains selected on the basis of similarity and difference, namely banking, electricity transmission networks, youth unemployment and invasive alien species. All four cases include traditional core state powers that member states have shared reluctantly. In each case, the call for EU-level governance has arisen from the recognition that transboundary effects and potential or existing crises (defined by the shared properties of urgency, threat and uncertainty) exist that are likely to overpower any national crisis management regime and where policy externalities emerge from national administrative action: banking crisis and electricity blackouts might be seen as ‘traditional’ crisis areas in which cascading effects are likely to be widespread. Youth unemployment presents a social crisis to member states and the EU as they might incur considerable population mobility across national boundaries and threats to political legitimacy. Invasive alien species represent often urgent threats to local ecologies and have become an increasingly transboundary issue, not just in terms of international trade and climate change-induced species migration, but also in view of the need for coordinated administrative actions to tackle cross-border invasions.

Over the last two decades, the four domains all witnessed the development of crisis management arrangements. However, each of these domains feature diverse governance arrangements with varying levels of authority and prescriptiveness. In terms of uncertainty, therefore, crisis management not only sought to address the inherent issues of substantive uncertainty during a crisis, but also the administrative uncertainty arising fron differential implementation at the national level.

The research involved the analysis of formal provisions, documents and media coverage regarding the emergence and operation of the four domains’ regimes dealing with transboundary crises. In addition, the paper draws on 135 elite interviews across EU, national and local levels, involving both state and non-state actors involved in the policy process (both in terms of development and implementation). Semi-structured interviews covering the evolution of EU and national regimes and their interaction were conducted between early 2016 and March 2018 under conditions of consent and non-attributability, and covered different member states (France, Germany, Italy, the UK, Spain, Portugal and Hungary). Member states were selected based on their centrality to the process and operation of a particular regime. While the following cannot, due to space limitations, give justice to the richness of the empirical material, the combination of documentary and interview material was particularly suited to uncover and analyse the logics shaping the mechanisms that shaped the debates regarding the different regimes.

Multilevel transboundary crisis management in the EU

The following section discusses the evolving and varied transboundary crisis management regimes in each of selected domains. The aim is to highlight how these regimes have evolved across the two dimensions of centralised authority and degree of prescriptiveness. Choices across these two dimensions reflect member states and European Commission views regarding the potential impact of differential implementation across member states on the overall capacity of transboundary crisis management. We subsequently note how inevitable concern over differential implementation of these regimes encouraged continued contestation.

Banking crisis management: towards centralised governance

In the aftermath of the financial and, more particularly, the Eurozone crisis, the EU set in place a banking union for the Eurozone, transferring regulatory powers to the EU. The banking union, including a single supervisory mechanism and a single resolution mechanism, consisted of one of the most significant shifts in recent years in transfers of authority to the EU-level (Howarth and Quaglia 2016). It also coincided with a significant strengthening of procedural prescriptions to guide member states’ authorities conduct. This strengthening reflected the concern among some member states that the limited compliance and capacity of certain national regulators and financial authorities, as well as acute coordination challenges, had aggravated the financial crisis and the subsequent sovereign debt crisis.

Prior to the financial crisis, only limited coordination existed through regulatory networks to deal with cross-border groups and the effects of financial integration in a single currency area given an overall consensus that national regulators would ‘do the right thing’. At the same time, a select group of critics noted that national regulators were primarily interested in defending ‘national champions’ rather than the overall Eurozone stability. These concerns over ‘home bias’ and the potential ‘doom loop’ between failing banking sectors and the sovereign were limited. However, the banking crises of 2008–2009 and 2011–2013 highlighted the limits to such a low centralised authority and low prescriptiveness approach. The crises also revealed that failing banks could not be managed through the EU’s state aid regime alone. The peak of the sovereign debt and banking crises in the Eurozone in 2011–2012 represented a turning point as countries such as Spain came on the brink of default, leading to a shift in member states’ views in favour of centralised banking authority at the EU level to ensure consistent supervision and banking crises resolution (Epstein and Rhodes 2016; Rynck 2016).

A set of measures negotiated by banking regulators and first adopted by the Basel Committee and the Financial Stability Board were transposed in EU law at that time (Quaglia 2014). These measures included a strengthening of supervision with the adoption in 2012 of a ‘Single Rulebook’ (focusing on capital requirements) and, in 2014, of a ‘Banking Resolution and Recovery Directive’ that required all member states to develop national resolution authorities– in other words, to organise and plan for crisis management. Following a decision from the European Council, Eurozone members established a ‘banking union’ consisting of a Single Supervisory Mechanism (in 2013), attached to the ECB and tasked with directly supervising the most significant banks of the Eurozone, and a Single Resolution Board to deal with banking crises. This near-independent agency oversaw the planning and execution of the resolution of those most significant banks under the Single Supervisory Mechanism and those of a cross-border nature. The Council and the Commission were required to sign off before any resolution action was to take place. This centralisation of authority at the EU level went hand-in-hand with a growing emphasis on prescriptiveness to ensure information exchange among EU-level actors and that member states authorities cooperate and consistently undertake their activities.

Despite the centralisation of banking crisis resolution and stringent prescriptiveness, the implementation of the resolution agenda and, more generally, of the banking union revealed considerable tensions. As suggested in Table 1, a centralised regime was likely to face issues dealing with information asymmetries and diversity in implementation. Although resolution legal authority rested with EU institutions, implementation still relied on national administrative capacities. One crucial division in this regime was between the large, significant banks (about 123) directly supervised by the ECB and the Single Resolution Board, and the thousands of small, regional banks (as found in Germany, Italy or Spain) still under the supervision of national authorities (even though EU institutions still supervised the supervisors).

Second, because resolution constituted an entirely new policy area and a new profession, the development of the Single Resolution Board remained a work in progress. National resolution authorities that had gained experience during the crisis had some leeway to promote their favoured resolution strategy (as in the case of Banco Popular in Spain in 2017), while the Single Resolution Board could influence those that lacked expertise or personnel. In addition, national bankruptcy laws and bank models still varied significantly, which provided for an additional level of ‘customisation’ to the European resolution regime. For example, Italy used its specific liquidation regime in 2016 to proceed with a precautionary recapitalisation of Montei dei Paschei, and put into resolution two of its small banks, even though the Single Resolution Board had deemed that resolution was not in the public interest (Donnelly and Asimakopoulos 2020). These cases, as well as concerns over highly prescriptive common rules, exposed how even the most centralised regime faced challenges in dealing with incomplete integration, problem-solving capacity challenges, and national diversity (Asimakopoulos and Howarth 2022).

These concerns re-emerged in view of US banking failures (starting with Silicon Valley Bank) and the buy-out of Credit Suisse by UBS in early 2023. These episodes highlighted not just whether any form of resolution and ‘bail in’ was still credible (as the US immediately declared that it guaranteed all depositors) but also whether certain banks were, indeed, ‘too big to fail’ and that assumptions about who would be affected, and in what order, would not stand the test of an actual crisis (as in the case of Credit Suisse). Subsequently, debates re-emerged over the degree of centralised prescriptiveness over the handling of smaller bank institutions (over national discretion to use taxpayers’ money), following on from a previous return of a conflict over the creation of a EU-wide deposit insurance scheme.

Invasive alien species: towards delegated governance

Invasive alien species represented a ‘new’ policy domain that responded to growing concerns regarding the impact of alien species on local, if not national biodiversity in the early 2000s. It also responded to the realisation that legal authority over invasive alien species fell between the cracks of existing habitat- and biodiversity-related provisions – both at member state and EU-levels. International conventions existed, but individual nation-states showed varied enthusiasm in taking action. Problems with invasive alien species increased with the deepening of the Single Market which facilitated the free circulation of plants and animals that, in different habitats, could prove destructive to local ecosystems and biodiversity.

In 2014, the EU passed a Regulation to prevent and respond to risks and crises posed by invasive alien species (Justo-Hanani and Dayan 2020). While standard-setting was delegated to the EU, tasked with defining a ‘black list’ of species of concern, continued reliance was placed on member states to engage in prevention and crisis management (through eradication notably). There was only limited centralisation of authority for transboundary management; most of the regime relied on implementation by member states with considerable scope for ‘customisation’ even in the context of an EU Regulation. Given limited capacities in most countries to deal with an issue often not considered a priority and cutting across administrative boundaries, developing regulatory tools and administrative capacities to engage in prevention, early warning, rapid responses, and comprehensive control strategies proved challenging. Indeed, apart from the UK, which had a dedicated secretariat, most member states had limited and dispersed capacities to deal with invasive alien species.

At the same time, the development of risk assessments for inclusion onto the ‘blacklist’ relied on a high degree of prescriptiveness. The demand to prescribe risk assessments was advocated by experts and the European Commission to put forward ‘objective’ and ‘neutral’ science-informed perspectives to reduce likely accusations of political biases. A scientific committee was created to assess which species were to be ‘blacklisted’ and to define common risk assessment measures and standardised crisis prevention and response strategies. Subsequently, a management committee consisting of national representatives considered these assessments in view of wider economic and social impacts of decisions. Such a highly prescriptive approach was viewed as essential for encouraging decentralised administrative capacity building and ensuring consistency. The Regulation required member states to upgrade their laws and administrative capacities to enhance attention paid to the topic and facilitate a more standardised approach across member states and cooperation.

The Regulation reflected conflicts across member states and EU-level institutions about the appropriate level of authority and degree of prescriptiveness. For some member states, a list at the EU level was of questionable value as ecosystems and ecological vulnerabilities across member states varied considerably – and so did the administrative feasibility of addressing certain invasive species. Similarly, certain actors (member states, MEPs and industry interests) were concerned about the ‘neutrality’ of risk assessments, fearing that these would be used to either include or exclude particular species given lobbying pressure.

The eventual choice of a Regulation was justified by actors within DG Environment by the need to signal ‘legal force’ and therefore prescribe national administrative attention in implementation. The use of the Regulation-tool also proved problematic for member states in their decentralised adjustment to the legal provisions’ requirements which were framed in anticipation of these provisions being passed as a Directive. The high degree of legal prescriptiveness created difficulties in implementation at the national level as administrative, legal, environmental and economic contexts varied greatly.

Electricity transmission networks: move towards decentred governance

European electricity networks are largely interconnected across Europe following long-standing cooperation among transmission network operators and as a result of EU-led initiatives to encourage network interdependence via the creation of interconnectors (to increase incentives for liberalising markets and reducing vulnerabilities). Concerns about transboundary cascading effects and blackouts have also been long-standing, and operators have addressed these by creating networks codes and operational handbooks to prevent and handle emergencies in a common manner. Yet, the European Commission repeatedly called for measures to enhance interconnection and coordination among member states to ensure ‘security of supply’ as enthusiasm for both was often lacking, especially in times of crisis when operators (and their government) are always tempted to serve their national interests first (a concern that became particularly prominent in gas given the dwindling of Russian gas supplies prior to and following the invasion of the Ukraine).

Renewed interest in how to manage crises across borders emerged in the mid-2000s, following blackouts originating from Italy (2003) and Germany (2006), as well as the growing integration of energy markets. In this context, member states were willing to accept more institutionalised forms of coordination for crisis management. However, due to the subsidiarity principle, questions of security of supply remained a national competence (as formalised under the Lisbon Treaty). Attempts to move the issue to the EU level could only be initiated via the Commission’s competence on environmental or competition bases (Buchan and Keay 2016). Initial directives and regulations sought to establish conditions for an ‘adequate’ level of interconnection and generation, and encourage ‘solidarity’ among member states. A European regulator, ACER, was established in 2009, albeit with limited powers, especially vis-à-vis member states regulators.

Given member states reluctance to grant the Commission centralising authority and thereby lose elements of central core state power – namely the power over deciding on continuous electricity supply – centralisation took a different route through an increasing reliance on coordination between transmission operators. In 2009, the European Network of Transmission Operators – Electricity (ENTSO-E) was given the role of drafting formal rules known as network codes, with ACER in an advisory role (Jevnaker 2015). In particular, a code on Emergency and Restoration was adopted in 2017 to harmonise crisis prevention and management across operators. Operators also set up regional joint crisis management centres, as trust between regional partners was higher.

Subsequent debates, encouraged by capacity shortfalls during the 2016 and 2017 cold winters, led the European Commission to call for enhancing the prescriptiveness of provisions governing risk and crisis management and strengthening supranational actors in crisis management – thus leading to a growth in levels of authority and prescriptiveness. Member states resisted such a shift, as did national regulators and transmission providers, as they tended to favour bi- and multilateral agreements outside the EU (such as the Pentalateral Forum). Eventually, this led to the adoption in 2019 of a Regulation on Risk Preparedness in the Electricity sector (2019/941), setting up a ‘common framework of rules on how to prevent, prepare for and manage electricity crises’ in the hope that common standards would facilitate coordination in times of crises.

In sum, this domain witnessed growing attention to the management of transboundary crises. However, this attention was met by a particular form of centralisation in the context of the EU, through the use of co-regulatory strategies that centrally involved transmission operators, in response to differential and sometimes weak implementation across member states. While the development of network codes did involve prescription, these codes granted providers some discretion, not just allowing for their participation in these processes but also permitting regional differences in network development.

Youth unemployment: move towards governance by benchmarking

Following the 2007–2009 financial and economic crisis, youth unemployment (and unemployment more generally) increased considerably in several EU Member states, especially in southern Europe. The issue of youth unemployment had traditionally been regarded as a matter for national welfare states. However, the rise of significant youth unemployment, above 50% in countries such as Greece, Italy or Spain, turned this issue into one of European interest, in part because of the burden on welfare states and migration in the short-term, reflecting unequal administrative and budgetary capacities to deal with such high unemployment numbers, and in part, because of the long-term demographic and economic implications. Leaders perceived the problem as a threat to the legitimacy of the EU in response to growing concerns that only financial and economic aspects of the 2008–2011 crisis had been addressed to the detriment of social issues, potentially fuelling discontent amongst youth. An informal policy-making network spanning the European Commission (led by DG Employment), the European Parliament and civil society groups advocated at that time a more systemic response to the social crisis that offered a counter-balance to the dominant economic discourse.

Two summits in 2012 in Berlin and Paris led by Angela Merkel and François Hollande placed youth unemployment on the EU agenda; identifying youth unemployment levels in certain member states as a crisis in itself, but also part of the wider EU response to the financial crisis. Subsequently, the 2012 Youth Unemployment package focused on job creation and labour market reform. The European Council adopted in 2013 a Recommendation to implement a Youth Guarantee to support ‘school to work transitions’. Its objective was to ensure that young people received, within four months of being unemployed or inactive, an offer of employment, training or education. Whilst framed as an ‘emergency response’ addressed to a particular sub-section of the labour market (the so-called NEETs, ‘not in education, employment or training’), the Youth Guarantee was seen by some as the first-ever EU social right. In addition, qualifying Member states were also granted access to additional financial means in response to the perceived emergency and unequal budgetary resources (further depleted by the financial crisis). A dedicated budget of €3.2bn for a Youth Employment Initiative was matched with another €3.2bn of national allocations from the European Social Fund (ESF), and another €6.2bn would be drawn from the ESF, in response to perceived differences in implementation capacities. The crisis, therefore, constituted a turning point for EU youth unemployment policies that, until then, had consisted of the loose benchmarking of national state policies (as part of the ‘open method of coordination’).

Nevertheless, the overall regime remained distinctly limited in terms of centralisation of authority or prescriptiveness. In light of varied enthusiasm to participate in EU social policies (Savage and Howarth 2018), youth employment policies were included in the European Semester (Copeland 2019; Copeland and Daly 2018). A scoreboard of indicators was created to establish ‘early warning tools’ to incentivise the implementation of the Recommendation and to ‘socialise’ the EU response to the financial crisis. There were debates about what a ‘good’ (or ‘quality’) position was, and how to define it so as to count as part of the Youth Guarantee. There were also differences among and within Member states about whether job seekers would be placed in further education or directly in employment. However, a much higher degree of prescriptiveness existed when it came to the distributive aspects of youth unemployment policies under the umbrella of the ESF.

In terms of authority, the diversity of welfare states and the Treaty base stood in the way of enhanced centralised authority. At a different level, the lack of central authority was also noticeable in the bifurcation that emerged in this domain between those particularly interested in indicator development and peer-review, and those, more senior politicians, who were largely interested in financial transfers. In addition, varying levels of national administrative capacities also hampered the implementation of EU policies, even when substantiated by significant EU funding from the ESF (Spain received nearly €1 billion). Overall, the reliance on benchmarks and ‘experimentation’ reflected the limited ‘spill-overs’ from differential implementation onto other member states.

Structural tensions in multilevel crisis governance

The previous section explored the extent to which perceptions about the implications of differential implementation for effective transboundary crisis management shaped the formulation of transboundary crisis management regimes (as summarised in Table 2). The following section highlights how debates continued to focus on the two dimensions of level of prescriptiveness and locus of authority.

Table 2. Multilevel crisis governance arrangements and accountability biases.

	Crisis regime characteristics	Differential implementation concern
Banking resolution	Reliance on EU-level actors to develop standards enforcement and govern decision making	National ‘doom loop’ given incentives to provide for loose oversight over national banks leading to EU authority and high prescriptiveness
Invasive alien species	Reliance on risk-assessment driven ‘black list’ and national implementation	Concerns with national prioritisation over EU black list leading to use of ‘Regulation’ device
Electricity transmission networks	Reliance on industry-developed codes and industry-regulator implementation	Concerns with unequal capacity of different national transmission networks, with lack of coordination during crises, and with transferring authority to the supranational level
Youth unemployment	Reliance on indicators and criteria for funding	Concerns over differential administrative and budgetary capacities

Each domain was characterised by specific governance arrangements that displayed particular implementation tensions. In terms of centralised governance, there was criticism regarding the distribution of authority across EU and national banking authorities: the ‘one-size-fits-all’ approach to resolution was criticised for not sufficiently taking into account national banking contexts and consequences on small creditors and the economy. In terms of decentring, there were concerns about the entrenched power of industry actors (organised through ENTSO-E), but also regarding the differentiated ways in which member states encouraged the maintenance and expansion of their own transmission networks. In terms of delegation, the invasive alien species regulation attracted criticism for being both over-prescriptive and granting too much leeway to member states in responding to the Regulation. Finally, in terms of benchmarking, there was considerable concerns about how different member states utilised the Youth Guarantee without much of a meaningful central overview over the implementation of these programmes, allowing member states to report and use financial resources creatively.

Furthermore, across the four domains, conflicts existed over the two governance dimensions, namely the degree of EU/member states authority and the degree to which member state actors were to be granted more or less discretion in their activities. One such conflict was about the lack of a central authority: in the banking sector due to the complex architecture governing supervision and resolution; in the electricity sector because of the dominance of industry actors in defining network codes; in the case of youth unemployment with the ‘lack’ of centralised measures tackling the social consequences of market integration; or in the case of invasive alien species with the lack of central backing for the regime.

At the same time, there were concerns about a lack of subsidiarity, referring to judgments about over-reliance on EU-level competencies and capacities, and calls for placing more authority at the level of the member states or other non-EU regional arrangements instead. Such criticism reflected, for example, concerns about the need to take into account different ecological systems in the case of invasive alien species (that conflicted with a uniform ‘black list’), the lack of decision-making autonomy granted to financial regulators and resolution authorities despite different national-legal sensitivities, the need to reflect diverse labour markets and welfare states in the context of youth unemployment or the concern with granting the European Commission or ‘regional coordinators’ too much authority in risk and crisis managing electricity transmission networks.

Such conflicts also affected the ‘prescriptiveness’ dimension. The concern with a lack of prescriptiveness focuses particularly on discretionary member state activities, in other words, the consequences of permissive customisation by member states. For example, there was concern about discretionary banking supervision and resolution activities by member states that were accused of representing industrial policy in all but name and were seen has having the potential to lead to European-wide crises. Similarly, the discretion granted to member states in developing their capacities for invasive alien species management was seen as undermining the objectives of the Regulation and encouraged calls for more prescription. The same criticism was used by those concerned with the varied application of the Youth Guarantee within and across affected member states. Similarly, concerns with different levels of member states compliance when it came to ‘risk managing’ their electricity transmission networks were seen by some as a reason to call for greater prescription, as was the concern by the European Commission that a lack of prescription could be exploited for purposes of protecting national markets and providers.

However, there was also a concern across all four domains with a lack of flexibility. This related, for example, to the concern in banking resolution with the need to adjust for national insolvency law regimes, in the case of invasive alien species over questions about what ‘managing’ and ‘eradication’ might involve given the administrative feasibility of such efforts in different jurisdictions, the criticism that prescriptive requirements for electricity transmission networks would be inappropriate in view of radically different networks, and the need to deal with questions of labour market flexibly.

In sum, across all four domains, demands existed for either more or less centralisation and prescriptiveness across both dimensions. These debates highlighted the irresolvable tensions inherent in the emerging EU crisis management regimes, which in turn reflect not just differential implementation across member states, but also how contestation over differential implementation might trigger dynamics that inevitably further differentiation rather than lead to ‘deepened centralisation’ as Jones et al. (2021) would expect.

Conclusion

Transboundary crisis management offers a particularly challenging area to explore differential implementation. Whereas customisation at the member state level is usually regarded as an ‘efficient secret,’ enabling member states to address domestic concerns whilst meeting EU standards, in the case of transboundary crisis management, concerns with the costs and risks of customisation are particularly prominent given the possible consequences of crises and authority of national and EU leaders. These concerns over customisation or differential implementation remain central to deliberations over transboundary crisis management regimes, or ‘differential policy integration’ as Zhelyazkova et al. (2023) would argue. Indeed, differential policy integration is a both an anticipatory and reactive response to differential policy implementation, seeking to constrain the potential of differential implementation leading to transboundary crises itself.

We observed, on the basis of the two dimensions regarding the level of authority and prescriptiveness, four different ways in which the EU and its member states have organised to address different degrees of concern with member states customisation. The paper has also highlighted that regardless of regime choice, inevitable contestation emerged. Transboundary crisis management was far from inevitably moving towards deeper integration.

Such a varied pattern should not come as a surprise. As EU citizens experience crisis management at the front line of their national and or local administration, the observed patterns point to a dispersion of decision-making authority within EU regimes and executive responsibilities at the national level. Such complexity leads to confused lines of accountability. For example, the likelihood that a particular bank has to be resolved may have decreased thanks to the presence of an EU-level transboundary crisis management regime, but when such crises do occur, the presence of a European regime increases the cost on national politics to account for EU banking resolution-related decisions that impose costs on national citizens. In other words, transboundary crisis management in the EU may support increased capacity to prevent and manage crisis. However, it increases the cost on national politicians to account for crisis management decisions that are no longer inside their jurisdiction.

The four empirical cases highlighted how the degree of anticipated customisation during implementation at the member state level and associated risks to transboundary crisis management capacity encouraged more prescriptiveness and attempts by the Commission to centralise authority – as was the case in banking and electricity. Although there was a perception that youth unemployment could become a threat to the EU, the consequences were mostly felt and managed at the national level, thereby limiting attempts to transfer authority and increase prescriptiveness (in addition to legal limits due to the subsidiarity principle). Finally, for invasive alien species, while a high degree of prescriptiveness was advocated by scientists, such an approach was rejected by member states. Anticipation of limited enthusiasm by member states to implement the EU measures then led the Commission to push for a Regulation rather than a Directive, which invited further tensions given the considerable administrative and issue diversity facing member states.

Our article offers some broader insights into debates about the problem-solving capacity of the EU. It has become fashionable to bemoan the limited problem-solving capacity of the EU, to point to the highly asymmetric nature of governance arrangements, and to complain about the centralising tendencies of the European Commission (an ‘Einheitsstaat’, as some might put it), as exemplified in discussions about the politics of emergency in the EU (Kreuder-Sonnen and White 2021). Empirically, the emergence of transboundary crisis management regimes, a central core state power, points to considerable flexibility in regimes. The research did not encounter a highly centralised variant involving central EU-level authority that was combined with a highly prescriptive approach constraining member states’ conduct. Even in the case of banking, centralisation was contested, and the regime incorporated considerable scope for (contested) flexibility, as debates about regulatory standards illustrated. We have found calls for centralising responses across our four domains, but equally, our discussion has highlighted opposition to such arrangements and the existence of alternative governance arrangements.

Similarly, contemporary EU crisis management debates focus on contrasting national and EU-level responsibilities and consequent challenges, given the dominance of national identities in the context of crisis. Our discussion does not deny the presence of such tensions but points to a far more differentiated picture that places emphasis on interdependence and a better understanding of how different jurisdictions can work together. These tensions are generated less by concerns about legitimacy as such (although they undoubtedly play a role), than concerns regarding the potential impact and probability of a transboundary crisis, and by concerns about the perceived impacts of differential implementation of transboundary crisis management regimes.

Despite the remarkable growth of crisis management capacities over the past two decades or so across a range of domains, transboundary crisis management remains, therefore, ‘unsolvable’. What remains ‘unsolvable’, and reflects a state of evolving differential implementation, are inevitable conflicts over appropriate levels of authority and degrees of prescriptiveness. In other words, differential implementation feeds back into the evolution of crisis management regimes, creating further tensions. It is in this sense that the European crisis is ‘unsolvable’. To legitimise governing under such conditions requires a continued focus on managing tensions, namely by focusing on the interaction and interdependence across levels of government, rather than focusing on a particular normative worldview about the power of EU institutions or member states respectively.

Ethical approval

The relevant ethics documentation governing the research is available here (https://www.lse.ac.uk/accounting/carr/research/TransCrisis/TransCrisis-Publications) and received institutional approval.

Acknowledgments

We thank the organisers of the special issue for their continuous support and insights.

Disclosure statement

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

Additional information

Funding

The research for this article benefitted from funding of the ‘TransCrisis’ project under the European Commission’s Horizon2020 programme call H2020-Euro-Society-2014.

Notes on contributors

Lydie Cabane

Lydie Cabane is an Assistant Professor of Crisis Governance at the University of Leiden. She studies the development of crisis policies in various governance settings. She recently published The Government of Disasters: State Formation and Disaster Management in South Africa, Springer/Palgrave Macmillan (2023). [[email protected]]

Martin Lodge

Martin Lodge is a Professor of Political Science and Public Policy, Department of Government and Co-Director of the Centre for Analysis of Risk and Regulation at the London School of Economics. His research interests are in the areas of regulation and executive politics. [[email protected]]