A systemic perspective on crisis management and resilience in Germany

In the past decades, Germany was hit – in equal measure to other countries in Europe and beyond – by multiple transboundary and societal crises. We take stock of the ability of the German state to cope with the ensuing complexity in managing these exceptional situations. Conceptually, we apply a sys- temic perspective that asks about the resilience of the German state in the subsystems of policy-making in crises, implementation of administrative crisis management, as well as societal responses to crises. The paper draws on findings from a range of empirical studies assembled in this special issue, that focus either on the so-called refugee crisis of 2015/16 or the Covid-19 pandemic since 2020. Strikingly, the overall impression emerging from this research is generally favorable of the ability of the German politico-administrative system to mas-ter challenging crises – its resilience. But there are also areas for improvement. der administrativen Imple- mentation des Krisenmanagements sowie der gesell-schaftlichen Reaktionen auf die Krise untersuchen. Der Beitrag verarbeitet die Ergebnisse einer Reihe empirischer Studien, die in diesem Themenschwer-punkt zusammengestellt wurden. Diese beziehen sich entweder auf die so genannte Flüchtlingskrise von 2015/16 oder auf die Covid-19 Pandemie seit 2020. Die vorgestellte Forschung zeichnet einen po-sitiven Gesamteindruck hinsichtlich der Fähigkeit Deutschlands, die Krisen zu bewältigen – und somit auf seine Resilienz. Es gibt jedoch noch Raum für Verbesserungen.


Introduction 1
The early 21st century saw a striking clustering of extensive societal crises in Germany, Europe and beyond, caused by or related to widespread environmental disasters, a systemic failure of the financial market, exuberant migration movements -most recently with a new surge of refugees following Russia's attack against the Ukraine -, and a pandemic of global scope. Those events were characterized by an unprecedented interdependency and complexity, cutting across boundaries defined by political, administrative, or geographic jurisdictions, policy fields, and time. They also caused disruptions and major challenges for the functioning of our political, economic, and societal systems and institutions. Political science research has increasingly studied the processes, dynamics, and consequences of crises like these (Boin, 't Hart, Stern, & Sundelius, 2016;Boin & Lodge, 2016;Kuipers & Welsh, 2017;Rosenthal, Charles, & 't Hart, 1989). Yet, we only begin to understand how governance structures and processes, in particular in multilevel settings, can contribute to coping with this complexity and to enhancing systemic resilience in crisis situations.
Germany was hit by those crises in comparable measure to other OECD countries, whereby overall it tended to fare relatively well. 2 In the 2015/16 refugee crisis, Germany was among the countries receiving the highest number of asylum seekers (see e.g. Eurostat, 2022). But most cities and communities managed to provide accommodation and basic living supplies without major disruptions of their regular responsibilities. After the initial shock of the Covid-19 pandemic, the German economy recovered more quickly than in most of its neighboring countries. And the death toll in the pandemic was clearly below average. 3 Still, in the public discourse in Germany, governmental crisis management was recurrently harshly criticized.
One line of criticism was directed against inefficient political processes and administrative structures -decisions taking too long, hesitant implementation and lack of leadership. Lack of coordination between the 17 governments was seen as a major institutional obstacle to swift action in the refugee crisis (Bogumil, Hafner, & Kuhlmann, 2016;Riedel & Schneider, 2017;Thränhardt, 2020) and the pandemic alike (Lenz, Eckhard, Obermaier, & Hoffmann, 2022, S. 20). In consequence, a discourse has emerged on how state reform can contribute to making Germany more resilient against future crises (Dettling, 2022;Nationale Akademie der Wissenschaften, 2021).
The other line of criticism aimed at an alleged democracy deficit of German crisis management, with the executive exploiting the crisis situation for seizing power, and parliaments and the public willingly disempowering themselves (Merkel, 2020;Münch, 2021;Ooyen & Wassermann, 2021). Even though strong leadership may have been warranted, the government's political communication was widely seen as deficient (Blum & Kuhlmann, 2021;Siefken & Hünermund, 2021). In consequence, a discourse has emerged on how the democratic quality of crisis management could be improved in future crises (see e.g. Forst, 2021).
We position this special issue in between those two lines of criticism -the alleged lack of efficient decision-making and democratic legitimacy. It assembles and connects up to date empirical research on the nature of crisis management and the impact of recent crises on Germany. Contributions were selected to offer a systemic perspective on crisis management (see section 3) linking the subsystems and addressing interdependencies between (1) policy-making within and across levels of government, (2) policy implementation and adaptability of the (local) administration, and (3) society by way of citizen perceptions and repercussions on the legitimacy of the state. For each of these subsystems, contributions ask how well the various challenges were processed, how they were interconnected, and how well the overall crisis response contributed to solving underlying problems.
As their main finding, and opposed to the rather critical media discourse accompanying recent crises (Reus, 2021), contributions to this special issue add up to a general-ly favorable impression of the ability of politics and administration to cope with crisis. The findings hint at a problem-solving oriented rather than politicized style of national policy responses and effective local adjustment and crisis action. Citizens showed a remarkable willingness to contribute to crisis management with voluntary work and donations (around 25% in the 2015/16 refugee crisis, see Lenz in this issue). And the Covid-19 pandemic did not cause a major disruption of citizen trust in state institutions (Lenz, Eckhard, Obermaier, & Hoffmann, 2022). Still, there is scope for improvement. Towards the end of this introductory paper, we summarize the lessons learnt from the contributions to this issue on how to attain systemic resilience in the long run.
With its conceptual approach and findings, the special issue responds to a gap in the comparative literature of crisis management: Despite a growing bulk of studies, in particular driven by the Covid-19 pandemic, the German case has not been investigated broadly and deeply so far. Whereas previous efforts analyzed individual crisis events, such as the so-called refugee crisis of 2015/16 or the Covid-19 pandemic 4 , this special issue looks beyond the single crisis context and aims to identify more general trends and effects. We also broaden the perspective beyond (political) decision-making and (administrative) crisis management, by considering the societal response.
We begin by displaying our understanding of crisis, crisis management and resilience as core concepts based on widespread definitions in the social sciences (section 2). Next, we introduce our conceptual perspective and the three crisis subsystems (section 3). We then summarize extant research on each of these subfields consecutively and situate the contributions to this special issue in these debates, emphasizing their respective contributions (sections 4 to 6). In the conclusion, we summarize the main insights regarding weaknesses and strengths of those subsystems, their interrelation in crisis management and what can be learnt from this for improving resilience in Germany.

Core concepts: crisis and resilience
A crisis is a situation of an urgent threat to core values or life-sustaining functions under high uncertainty and time pressure for the decision-maker (Boin, 't Hart, Stern, & Sundelius, 2016, p. 42). Crises are "characterized by unclear problem definitions, complex causalities, conflicting goals and lack of standard solutions" (Ansell, Sørensen, & Torfing, 2021, p. 950), as is typical of so-called 'wicked problems' (Head & Alford, 2015). Yet, added to the 'usual' wicked problem structure is what Ansell, Sørensen and Torfing (2021, p. 950) call 'turbulence': an environment that is "surprising, inconsistent, unpredictable, and uncertain". Finally, a crisis contains the risk of turning into a disaster, yet also the chance of avoiding it. Here, the crucial role of decision-makers comes into play. Their tasks consist, first of all, in 'sense-making', i. e. in acknowledging that the situation is a crisis, and in taking the appropriate decisions to avoid a disaster (Boin, 't Hart, Stern & Sundelius, 2016, p. 15). Taking the right preventive action is, however, by no means an evident strategy, as Ferguson (2021, p. 10) illustrates with a quote of Henry Kissinger's 'problem of conjecture': "Each political leader has the choice between making the assessment which requires the least effort and making an assessment which requires more effort. If he makes the assessment that requires least effort, then as time goes by it may turn out that he was wrong and then he will have to pay a heavy price. If he acts on the basis of a guess, he will never be able to prove that his ef-fort was necessary, but he may save himself a great deal of grief later on. … If he acts early, he cannot know whether it was necessary. If he waits, he may be lucky or he may be unlucky. It is a terrible dilemma." The problem with crises is that each crisis individually is highly improbable, yet the risk that some crisis occurs at one point is rather high. The unspecific nature of the threat makes it difficult to prepare an adequate reaction. As Ansell, Sørensen and Torfing (2021, p. 949) put it: "it is not enough for the public sector to activate a predefined emergency management plan, call in the bureaucratic troops to deal with the crisis, and let them do their professional work …". As the probability for crisis situations has drastically increased due to the growing interdependence of the modern world (mobility, globalized economy, international regimes and interdependence of states), this begs the question how modern societies can best prepare for that which cannot be foreseen.
The most popular answer to this question is the need for systems to become resilient (Chandler, 2014), or to reduce their vulnerability (Miller et al., 2010) and increase robustness, flexibility, and adaptability (Ansell, Sørensen, & Torfing, 2021, p. 949). Resilience is a system's ability "to absorb change and disturbance and still maintain the same relationships between populations or state variables" (Holling, 1973, p. 14). Modern definitions are, however, more encompassing: -According to Elsner, Huck, and Marathe (2018, p. 31), a system is resilient when "demonstrating either one or a combination of the following capacities if confronted with a perturbation: absorbing (preventing a crisis); recovering (short-term coping and 'bouncing back' to the original state); adapting (long-term coping and 'bouncing forward' to a new state)". -Applied to the societal context, Norris, Stevens, Pfefferbaum, Wyche and Pfefferbaum (2008, p. 127) define community resilience as "a process linking a network of adaptive capacities (resources with dynamic attributes) to adaptation after a disturbance or adversity. Community adaptation is manifest in population wellness, defined as high and non-disparate levels of mental and behavioral health, functioning, and quality of life".
Resilience as a systemic quality thus presupposes learning from past crises to enhance the capacity for absorption; dynamic reaction, flexibility, and endurance to enable quick recovery; and quick and pragmatic adaptation to changing circumstances instead of safeguarding stability and conserving the status quo. In the next section, we elaborate on the challenges for decision-makers in a crisis situation, and the conditions supportive of an effective crisis response.

A systemic perspective on resilience enhancing crisis management: three subsystems
In the crisis management literature, a consensus has emerged that effective crisis management comes with a number of crisis management tasks. Boin, 't Hart, Stern, and Sundelius (2016, pp. 15 ff.), for instance, distinguish sense-making, decision making and coordinating, meaning making, accounting, and learning. Those tasks cut across the spheres of policy-making by elected politicians and the implementing administration. In addition to such 'internal' tasks of organizing a response and coping with uncertainty, Ansell, Boin, and Keller (2010, pp. 197 ff.) add an external perspective: To coordinate with other actors and -crucially -to communicate with the public. Indeed, crisis communication is fundamental for the above-mentioned notion of societal resilience, a concept that neatly connects the ability of the political-administrative system to respond to crises with the societal impact of that response. For this special issue, we deem such a comprehensive systemic perspective most appropriate and therefore distinguish conceptually between three subsystems of crisis management, as well as their interdependencies: The policy-making subsystem (including institutions, actors and processes), the administrative subsystem responsible for policy implementation, and the societal subsystem in terms of public trust and support as reactions to policymaking and implementation. Contributions to this special issue speak to each or several of these subsystems, as illustrated in Figure 1. As can be seen from Figure 1, we deliberately omit other systemic factors, such as the conditions, effects, and interactions of the fiscal and economic aspects of crisis management as well as the role of economic and civil society actors. We also limit our attention to national crisis responses, fully acknowledging that a further relevant aspect in improving resilience is to enhance governance and crisis management capacity in the European multi-level system (for a critical account on national crisis responses see Bouckaert, Galli, Kuhlmann, Reiter, & van Hecke, 2020;Ferrera, Miró, & Ronchi, 2021). Below, we discuss each subsystem in turn and emphasize the contribution that this special issue makes to the associated academic debate.

Challenges of policy-making
Political decision-making in times of crisis faces the challenge that swift and decisive action is necessary, but decisions inevitably must be taken under high uncertainty about the causes and dynamics of the crisis (Boin, 't Hart, Stern, & Sundelius, 2016, pp. 51 ff.). Calls for leadership recurrently make crises 'the hour of the executive' (Merkel, 2020;Münch, 2021), often supported by broad parliamentary and public acquiescence.
Facing an inherent uncertainty about potential consequences of action, policy-makers rely on knowledge provided by experts. Technocratic decision-making based on close cooperation between executives and experts may be the way to quick and potentially well-informed decisions. Yet, critical decisions, where competing values must be pondered, require a deliberative process involving a broad array of positions, interests, and opinions. This argument highlights the important role of parliaments in crisis policy-making. While parliamentary debate may slow down decisions, it provides an important forum for information, deliberation, and control of executive governance (Höhne, 2022, S. 3; with further references on the democratic and parliamentary deficit Hildebrand, 2020; and from a public law perspective Boehme-Neßler, 2021). On the other hand, public debates of crisis management in parliament risk to politicize measures. Crisis management can boost approval for government parties if they are generally regarded as successful, as the external threat is likely to produce a rally-'round-the-flag effect (Dietz, Roßteutscher, Scherer, & Stövsand, 2021) When public disapproval grows stronger, on the other hand, opposition parties have a chance of gaining profile and voter support by criticizing and opposing government strategies (Louwerse, Sieberer, Tuttnauer, & Andeweg, 2021). In an era of widening societal gaps and a high susceptibility to populism, as we currently seem to experience, public opinion and party politics are tightly linked.
As Sabine Kropp, Christoph Nguyen and Antonios Souris elaborate in their contribution in this special issue, party competition was de-emphasized during the first phase of the pandemic for the sake of an effective crisis response. In their content analysis of 202 debates in 'Länder' parliaments, they show that rather than engaging in detrimental blame-shifting and mutual recriminations that could polarize the political system (which we see in the United States), members of parliament claimed credit for their party's achievements, thereby avoiding disruptive dynamics.
The right balance between executive leadership and parliamentary control in the horizontal division of power is crucial for sustainable crisis management, yet needs to be recalibrated recurrently during a crisis. The same is true for the balance between centralized or decentralized decision-making, as it is not a priori obvious which level of government is functionally superior in managing the crisis. Regarding the 'centrallocal government continuum', recent studies generated valuable evidence (Hegele & Schnabel, 2021;Malandrino & Demichelis, 2020): Whereas centrally organized responses promise greater coherence for an entire polity, they seem more prone to failure in practice, such as in the case of the Hurricanes Katrina and Rita in 2005 in the US (Kapucu, Arslan, & Collins, 2010). By contrast, the advantage of decentralized systems is greater flexibility of action and adaptability to local conditions (Eckhard, Lenz, Seibel, Roth, & Fatke, 2021).
Most countries are located somewhere in between fully centralized or decentralized decision-making. A comparison of organizational structures and coordination mecha-nisms for crisis management in six European countries yields the predominance of a "composite system combining contradictory organizational principles that have evolved through institutional layering" (Christensen, Danielsen, Laegreid, & Rykkja, 2016, p. 316). Nowell, Steelman, Velez, and Yang (2018, p. 699) also identify "moderate core-periphery structures" in their analysis of multiple disaster management systems. Hegele and Schnabel (2021) show that even within and among federal states, decentralization of pandemic management varies greatly, a finding that is confirmed for Germany by Kuhlmann and Franzke (2021) who identified a shifting pattern between vertical coordination and decentralization in the course of a crisis.
Whenever policy-making is decentralized, this implies the problem how coordination can be achieved (Hegele & Schnabel, 2021, p. 18). Intergovernmental councils play a core role in this regard. Concomitantly, in Germany, the peak intergovernmental council, the 'Minister Presidents' Conference', gained particular importance as the prime body for taking joint decisions in crisis management (Behnke, 2021;Person, Behnke, & Jürgens, 2022). It provided not only an arena for negotiation and decisionmaking, but secured -equally important -information exchange to allow for voluntary adaptation (Benz, 2012). In their contribution, Johanna Schnabel, Rahel Freiburghaus and Yvonne Hegele tackle the coordination problem. They compare coordination activities during the pandemic between the peak intergovernmental councils in Germany and Switzerland. The activities of the German Minister Presidents' Conference intensified, while the Swiss Conference of Cantonal Governments did not expand its activities. Outputs of the German coordination body were, however, mainly aimed at internal coordination, while externally oriented outputs (such as concerted communiqués) were neglected. This encouraged diverging interpretations by heads of Länder governments generating the public impression of a lack of leadership.
The problem of regional diversification of crisis responses due to the decentralized power structure in Germany is the puzzle in the contribution by Nathalie Behnke and Christian Person. They investigate the extent of and drivers for the variation in policy responses (restrictive measures for health protection) at Länder level. As may be expected given the efforts to establish a coordinated strategy, responses generally followed a similar pattern in that restrictions increased when the pandemic became more severe. Variation between the Länder can in part be attributed to party ideology. On the whole, however, their findings confirm the findings by Kropp and colleagues of a prevalence of problem-solving oriented behavior in crisis management

The challenge of implementation
While decision-making is typically the task of the executive leadership in governments (and parliaments), a crisis is operationally fought at the local level. It is local administrations that carry the major weight of directly interacting with the people, e.g. by organizing and delivering immediate help, providing shelter, distributing money and living supplies, enforcing rules or processing requests. As was argued in the section above, crisis management in most countries works through a decentralized structure allowing for variation in responses adapted to local circumstances. Yet, the very nature of a crisis as a sudden, big, dynamic event defies routines and requests improvisation and 'bricolage' (Ansell, Sørensen, & Torfing, 2021, p. 953). Flexibility and improvisa-tion are, however, no typical strongholds of local bureaucratic structures and processes (Webb & Chevreau, 2006).
One core challenge in implementation is thus to boost local administrative capacity (Hahlen & Kühn, 2016;Klenk, Cacace, & Ettelt, 2021). Money, staff, and information processing and distribution capacities (Verenkotte, 2019) have repeatedly proven to be crucial for swift action and the capacity to process huge caseloads, such as registering and providing shelter for refugees or tracing the contacts on infected persons. In times of scarce public financial resources and after decades of reduction of public personnel, however, redundancy in staff necessary for a quick crisis response is obviously lacking. Most importantly, the protracted digitalization of German public administration in general and local administrations, in particular, as well as data protection requirements have seriously impeded large-scale coordination, processing, and exchange of information (Kersting & Graubner, 2020).
Local administrative staff is challenged in particular ways in its role as street-level bureaucrats who manage operational crisis management while interacting with society. In their contribution, Wolfgang Seibel, Christine Eckardt, Friedrich Huffert, Lisa Mende and Lorenz Wiese show that local administrations can activate additional 'resilience reserves'. Building on case study research and interview data (see Eckhard, Lenz, Seibel, Roth, & Fatke, 2021), they argue that such reserves can be mobilized by adjusting decision-styles within the administration (becoming more flexible) or by integrating external resources such as volunteers (becoming more participatory). Although the German system has sufficient resilience resources to handle one crisis, the authors conclude, it will likely be overwhelmed when multiple crises hit at a time.
In a decentralized crisis management system, coordination of crisis responses is another major challenge. German federalism includes three levels of government with 16 Länder, 401 counties and district free cities, and roughly 11.000 community level governments. Horizontal coordination across borders is important to avoid jurisdictional gaps or overlaps and to approximate a fair distribution of burdens. Information sharing could be eased by means of digital systems of information exchange and shared access to central information databases (Moynihan, 2009). However, not only are the means of digital information processing deficient, studies also show that crisis managers are typically more "concerned with receiving information from others than with providing information" (Bharosa, Lee, & Janssen, 2010).
Vertical coordination is just as relevant. Länder governments are in a focal position to bundle feedback from their local governments and channel it into coordinated decision-making, the result of which is then to be communicated back to the local level. Particularly federal legislation that is being updated in brief intervals in reaction to an evolving crisis situation needs to be adapted by local implementation. During the refugee crisis, local governments complained that they had no time to train their personnel and update them on the quickly changing legal requirements of case processing (Störr-Ritter, 2019). In the Covid-19 pandemic, local governments were primarily unsatisfied with the extent of information received from the Länder-level (Eckhard, Graf, & Lenz, 2021, S. 7).
The question how crisis responses can be coordinated horizontally and vertically, allowing for a flexible and adequate reaction to local circumstances is tackled by Jörg Bogumil and Sabine Kuhlmann in their contribution. They study the implementation of crisis measures in the so-called refugee crisis of 2015/16. The qualitative study offers a detailed analysis of the vertical integration (Verwaltungsverflechtung) between administrative levels in a crisis response. Similar to Benz (2020), they conclude that in principle, interlocking politics in a multi-level system is functionally beneficial, but that the practice in Germany is yet to be improved. Rather than calling for more or less decentralization, they argue that more differentiated integration is necessary.
6 The challenge of maintaining societal trust and support Crisis management by political leadership and local administrations is only as good as its acceptance and support by the public. The relationship between crisis response and public reaction is complex and barely understood Coman, Elsheikh, Gregor, Lilleker, & Novelli, 2021), not least because societal perceptions of crisis management are highly dynamic and volatile as a crisis unfolds over time. If people do not trust their government, if they are not convinced of the necessity and appropriateness of decisions taken, they will not act accordingly. To motivate the public to act in accordance with crisis measures is hence the absolutely basic condition for any crisis management to be effective.
Communication from political leadership to the public in order to explain the situation, the rationale behind decisions taken, the necessity of a joint effort, and also to signal responsiveness, is therefore of utmost importance (Boin, 't Hart, Stern & Sundelius, 2016, p. 78ff.), while "(i)nformation mismanagement can lead people to develop deep distrust in government" (Carlson, Jakli, & Linos, 2018, p. 671). In Germany, societal frictions, polarization, and radicalization were also side effects (or consequences?) of previous crises in Germany (Czymara & Schmidt-Catran, 2017;Hangartner, Dinas, Marbach, Matakos, & Xefteris, 2019). Rather, governments should deliver consistent messages to the public in order to incentivize citizens to some form of desirable behavior (Deslatte, 2020) and to foster societal resilience to face the situation (Fernandez-Prados et al., 2021).
In an era of social media, communication is hard to control. Messages need to be delivered in competition to alternative narratives, and (mis-)information abounds without any quality check. What is more, time pressure as well as constantly changing circumstances or new insights on causes and effects set a high bar for political communication to be effective and convincing. This 'infodemic' (Yang et al., 2021) undermines the acceptance of crisis management measures and stirs distrust, making it all the harder for governments to maintain public support. In Germany, the right-wing party AfD as well as various groups of anti-state sentiments were able to exploit the infodemic and to create communities of supporters against state crisis responses (Borucki & Klinger, 2021). Relative to their small numerical size, these protesters had a huge impact on the public discourse, creating the impression of a divided public when in fact the majority of the population supported the government policy (Lenz, Eckhard, Obermaier, & Hoffmann, 2022).
Yet, the contribution by Sebastian Jäckle and Uwe Wagschal in this special issue provides some reason to be cautiously optimistic regarding the German case. Their analysis of panel survey data shows that levels of specific political trust (state actors in office) decreased only moderately throughout the crisis whereas the average level of general political trust (state institutions more broadly) remained stable. This finding in-dicates that there is no contradiction between holding critical opinions about some aspects of German crisis management (such as federalism, see Lenz, Eckhard, Obermaier, & Hoffmann, 2022) and maintaining the pre-crisis level of trust in the state system.
Jäckle and Wagschal's findings dovetail with others who argue that effective crisis management can also be a chance to strengthen solidarity, identity, and resilience within a society (Boin, 't Hart, & Kuipers, 2018;Drabek & McEntire, 2003;Rosenthal, Charles, & 't Hart, 1989). As Christensen, Laegreid & Rykkja (2016, p. 889) summarize, this is the case "[w]hen governmental preparedness and crisis management match the expectations of citizens, the response process works well and governmental performance is perceived as good." Sustaining this insight, Borucki and Klinger (2021, p. 121) found for Germany that "government measures to contain the coronavirus were widely accepted and satisfaction with the government and communication was comparatively high". Questions remain, however, regarding the mechanisms between state action and citizen perception and the long-term effects on public trust and system resilience.
In her contribution to this special issue, Alexa Lenz offers one answer on how such mechanisms play out in practice. In her analysis of survey data from 900 volunteers who helped local governments in the refugee crisis, she shows that performance perception of the local administration, and the formalization of the different volunteer organizations, explain how volunteers' trust in the public administration changes over the course of the refugee crisis. This finding offers another perspective on how the state can influence public perceptions: Not only by means of communication and delivering effective results, but also the way how state representatives interact with volunteers likely has an impact on citizen perceptions, simply because volunteers are also citizens who may share their experiences in their personal networks.

Conclusion
The contributions assembled in this special issue address the question how the various subsystems in Germany reacted to the challenges posed to them by the so-called refugee or the pandemic crisis. Jointly, they give hints at how those reactions between subsystems were interdependent, e.g. how politics reacted to societal responses and vice versa, and how this interdependence contributed to or obstructed effective crisis management. Complexity and interdependence are inevitable characteristics of modern governance. Institutions and processes in democratic multi-level states must strive to mirror this complexity in order to deal with it successfully (Benz, 2020, S. 16). A comparative analysis across subsystems and across different crises, as provided here, reveals particular strengths and weaknesses of the political system in Germany to cope with crises. Based on those insights, it offers the chance to learn and improve systematic shortcomings.
Overall, the analyses in this special issue show that Germany performed reasonably well in crisis management. The locus of power in policy-making between the executive and parliament as well as in the federal division of authority gave rise to criticism, yet provided for quick and adequate responses. The changes in strategy between imposing restrictive infection protection measures and loosening restrictions mirrors the checks and balances between executive-driven crisis management on the one hand and the counterweights of parliaments, federalism, public opinion, and party competition on the other. Implementation of crisis responses at the local level put local governments under considerable stress, but they showed impressive action capacities as they responded to each exceptional situation with flexibility, great commitment, and ingenuity to obstacles. What is more, with joint support by volunteers and higher-level governments, personnel and financial resources could be amended to help process the enormous workload in due time. Finally, in spite of obvious protest movements and growing alienation among the population, overall public trust in the state and its institutions remained stable.
Still, the analyses also found room for improvement. As regards the policy-making subsystem, Kropp, Nguyen and Souris show that party competition contains a risk of jeopardizing political cohesion needed for sending a convincing message to the public. This warning is underpinned by the analysis of Behnke and Person who show that party politics account for some variation in policy responses across the German Länder. While party politics are an important link between voters and politicians, party competition should not promote populism and short-sighted responsiveness at the cost of responsibility. Communication is generally underestimated among policy-makers, as Schnabel, Freiburghaus and Hegele elaborated in their comparison of intergovernmental councils in crisis management in Germany and Switzerland. Governments need to find better communication channels to reach people not consuming traditional media.
As regards the subsystem of implementation, coordination within and across levels of government lacked not only successful communication to the public, but was found wanting -in particular between the federal and the local level in Bogumil and Kuhlmann's analysis. The complexity of the distribution of tasks and responsibilities between governmental levels fosters overlap, diffusion of accountability and regulatory gaps. However, street-level bureaucracies seem to be able to absorb formal deficits of the system to a large extent. As Seibel, Eckardt, Huffert, Mende and Wiese elaborate in their analysis on implementation and resilience, local governments can activate higher levels of flexibility and participation as 'resilience reserves'.
Studies on the subsystem of societal perceptions confirm generally high, yet vulnerable trust: As Lenz's analysis highlights, while local governments showed generally strong administrative capacities, street-level bureaucrats were not sufficiently aware of how their interaction with volunteers and the public shapes public perceptions and trust. This warning is also mirrored in Jäckle and Wagschal's panel analysis showing declining levels of trust in specific state actors over time (as opposed to the more general trust in state institutions).
While no individual contribution directly compared subsystem reactions across the refugee and pandemic crises, they jointly provide starting points for comparative reflections. Striking is an apparent shift as regards the most challenged subsystem in the two crises: While in the refugee crisis the administrative subsystem seems to have been under most stress, in the Corona pandemic the pressure was higher on the policy-making subsystem (and repercussions of political decisions on public trust). To some extent, this was a result of the nature of the crises. Reception and integration of refugees were mainly local or regional tasks, and horizontal coordination problems were limited to a fair distribution across the territory. In the pandemic, while local governments had to cope with implementing infection protection measures, the politics of coordination were the primary focus of public attention. As the infection protection law gave the Länder the right to issue their own regulations, coordination of crisis response was far more demanding.
Similarities can be observed in the challenges for local governments. In both cases, they suffered from a lack of resources to shoulder the immense case load, from insufficient means of data exchange and from constantly changing higher-level regulation. Financial means were provided, however, much faster by federal and Länder governments in the pandemic than in the refugee crisis. Also, party competition resulted similarly in distortions of crisis responses. As the crises lasted longer, tolerance towards party political bickering, recurrent restrictions and burdens and perceived inefficiency of crisis responses shrank. Under those circumstances, communication becomes ever more crucial to prevent a societal break-up.
All in all, lessons could be learnt for improving the resilience of German politics, administration, and society. However, chances for learning are limited, as this would require time to think about observed deficiencies and resources to change structures and processes. At the time of writing, the war in Ukraine began causing the next refugee crisis in Europe while the dust of the Covid-19 pandemic had not even begun to settle. Hence, it is likely that leaders and administrators "continue to repeat the same errors when a crisis occurs" (Lalonde, 2007, p. 17). Still, the magnitude of the past crises might make them game-changers in our thinking about public administration and leadership (Ansell, Sørensen, & Torfing, 2021). The evidence is there, in any case.
Notes 1 Previous versions of this contribution were critically discussed by Arthur Benz, Svenja Bauer-Blaschkowski and Alexa Lenz as well as by the journal editors Sylvia Veit and Thurid Hustedt. We gratefully acknowledge those remarks that helped to improve the manuscript. All remaining deficiencies are in the full responsibility of the authors. 2 As the contributions in this special issue provide empirical evidence on the refugee crisis of 2015/16 and the pandemic, we limit our argument to those two crises in particular, while we assume that the reflections related to resilience apply also to the other crises we experienced and which are yet to come. 3 Among OECD countries, as of February 2022, it ranked at place 14 of 38 in terms of the accumulated number of fatalities (1,447 deaths per one million inhabitants), see KFF (2022). 4 For the refugee crisis see e.g. Bogumil, Kuhlmann, & Proeller, 2019;Eckhard, Lenz, Seibel, Roth, & Fatke, 2021;Schomaker & Bauer, 2020. The covid pandemic has produced a pile of literature. Specifically regarding Germany, see e.g. Blum, Loer, Reiter, & Töller, 2021; Bundeszentrale für Politische Bildung, 2021; for a European perspective, see Goetz & Martinsen, 2021; international comparisons are provided e.g. by Chattopadhyay, Knüpling, Chebenova, Whittington, & Gonzalez, 2021;or Steytler, 2022.