Emergency management through enduring collaborative networks: Lessons on phases and levels
Abstract
This article explores the conditions that underpin networks as enduring collaborative relationships. To identify key features that sustain long-term collaboration, we use the Norwegian system of emergency management as our empirical case. Norway is characterized by the important role played by volunteers and voluntary organizations as partners to the police in search and rescue operations. In particular, the article focuses on two possible explanations for the endurability of these networks: the broad involvement of volunteers in the different phases and the role of intermediaries. The article concludes that intermediaries can compensate for a lack of broad involvement, which may be particularly relevant for emergency management where many different resources have to be mobilized in a short time span and in a fashion that enables fluid interaction. This finding is a reminder that both phases and levels need to be explored in empirical studies of collaborations.
1 INTRODUCTION
With the blooming literature on collaborative governance, our understanding of collaboration is developing at a considerable pace (Bianchi et al., 2021). A substantial amount of research has been carried out to understand, for example, how collaboration is influenced by institutional design and leadership (Ansell & Gash, 2007; Crosby & Bryson, 2005). While many collaborations between the public sector and societal actors are established to handle single events, the collaborators often deal with permanent challenges, and maintaining the collaboration over time therefore represents an important issue (Peng & Lu, 2019, p. 19). The conditions under which a collaboration will last over time are, however, underexplored in public administration research (Bryson et al., 2015; Gazley & Guo, 2020). This article will explore this issue by emphasizing two elements we suggest are significant in this regard: (a) the extent to which collaborators are involved in the different phases of a collaboration process and (b) the existence of intermediaries between the public sector and individual collaborators.
The empirical context for the following discussion and analysis is emergency management. In the scholarly literature on emergency management, cross-sectoral collaboration has become an important subfield (Nohrstedt et al., 2018), and the abovementioned concerns about endurable collaboration also comprise a topic in this subfield. Emergency management is characterized less by routine and more by unexpected events calling for improvised actions, making the coordination of collaborative networks a major task (Eller et al., 2015). Lasting relationships and stable, enduring collaborations are thus believed to be of major importance. This follows partly because such incidents often require contributions from non-public partners, such as volunteers and civil society. Consequently, dealing with new and unknown risks and situations requires trust among the involved, and trust is built through repeated actions and learning over time (Kapucu & Hu, 2016; Peng & Lu, 2019). In part, lasting collaborations are also believed to strengthen motivation among volunteers, since they allow these individuals to become involved in the different phases of a collaborative process, which, in turn, is believed to increase their motivation (Andrew et al., 2015; Schie et al., 2014).
While it is common to invite volunteers ad hoc when larger crises hit (Johansson et al., 2018; Villagran et al., 2006; Waugh & Streib, 2006), the system we analyze in the following is a stable and enduring collaboration with long historical roots that is less dominated by a specific profession (McGuire, 2009). In Norwegian emergency management, volunteers are not only a welcome supplement to public resources in everyday emergency management, but volunteers and voluntary organizations also make up an integrated, important part of the national emergency management system. This emergency management system would collapse if such volunteers suddenly disappeared. Particularly in search and rescue, most personnel conducting the search are typically volunteers organized in one of a handful organizations that have a formalized collaboration with the public sector. Furthermore, some of these organizations have specialized skills, like mountain climbing, that you will not find in public sector organizations. These aspects make Norway a typical case of enduring collaboration and, accordingly, an ideal context for studying characteristics that facilitate long-lasting, durable collaboration (Seawright & Gerring, 2008).
Most emergency management systems depend, in some way, on volunteers, although there are significant differences regarding how volunteers are integrated. For example, individual volunteers can provide spontaneous assistance when an incident occurs but still have no role in the formal networks for emergency management (Rivera & Wood, 2016). Volunteers can also be involved as members of formal civic organizations and extend their activities to cover emergency response when there is a need (Whittaker et al., 2015). By studying the Norwegian context, we examine a case where volunteers are involved through formal voluntary organizations and where their organizations are acknowledged as partners in a formal network for emergency management.
We examined emergency management through a qualitative study of collaborative networks in six areas in Norway. Our analysis is based on 37 interviews with representatives from the public sector and leading persons from voluntary organizations. These collaborative networks join forces to routinely handle search and rescue operations but can also be involved in other incidents, such as wildfires, avalanches, and flooding. Much of our understanding of emergency management is based on big “focusing events” that frame policymakers' and researchers' views on good emergency management (Birkland, 1998; Kapucu & Garayev, 2016). Despite every emergency event being unique and requiring customized actions, our study examines networks characterized by more routinized interactions related to, for example, searches for missing persons. Such research can broaden the scope of our understanding of emergency response networks and provide insights relevant to the broader literature on collaboration.
In conclusion, the article points towards a possible trade-off between broad involvement in all phases of emergency management and intermediaries in terms of voluntary organizations. We observe that volunteers are less involved in preparations and learning the more they are involved in larger organizations. This apparent paradox can be explained by the intermediary's level. Even if volunteers are not deeply involved in the emergency management networks' preparations and learning activities, they are all the more involved in activities organized within their single voluntary organizations. This indicates that to understand the collaboration process in depth, we need to understand the different levels and orders at play, including the individual, organizational and network levels. In the concluding section, we discuss the practical and theoretical implications of this finding, as well as pointing out limitations in our study readers should be aware of.
2 THEORY: PUBLIC–CIVIL INTERACTIONS IN EMERGENCY MANAGEMENT: PHASES AND LEVELS
Collaboration is a somewhat contested concept that has a number of different definitions. We base our study on Bryson, Crosby and Stone (2006, p. 44), who define cross-sector collaboration as “the linking or sharing of information, resources, ties, and capabilities by organizations in two or more sectors to achieve jointly an outcome that could not be achieved by organizations in one sector separately.” While numerous studies are based on this definition, there is a shortage of studies that have a specific long-term focus on collaboration (Getha-Taylor, 2019). Since the integration of specialized capacities into network efforts is laborious, securing enduring collaboration is considered valuable when efficient service production models are obtained (Weber, 2012).
Collaborative emergency networks are often mixes of centralized commands and interdependent relationships between autonomous organizations (Chen et al., 2013; Head, 2008), and the relations between participants can change throughout the different emergency phases. In the preparedness and learning phases, the relationship is often described as horizontal (Ansell et al., 2010), whereas in the response phase, public agencies will typically have formal responsibility and lead and coordinate response efforts, even if the same actors are involved. The simultaneous involvement in vertical and horizontal interactions represents a specific challenge, as autonomous actors must subject themselves to the control of otherwise equal actors during operations. Such semi-hierarchical structures have been shown to work well in non-catastrophic crises but to be less successful in more complex ones (Moynihan, 2009; Waugh Jr., 2009).
The complexity increases, we believe, as more volunteers and voluntary organizations become involved in emergency management. Some studies indicate that to collaborate efficiently in the response phase, collaborative skills, and trustful relationships must be developed in “horizontal” arenas in other, less acute phases of the emergency process (Chen et al., 2013). Nevertheless, the actual impact of collaboration on different arenas before, during, and after operations, as well as how such arenas should be devised, has received only limited attention (Nohrstedt et al., 2018).
Collaboration between state and non-state actors is a dynamic process, suggesting that we need to probe the changing relations among the partners and how they develop throughout the cycle (Bryson et al., 2015). While it has become mainstream in the research literature on collaborative governance to understand collaboration in terms of process models with different phases (Ansell & Gash, 2007; Susskind & Cruikshank, 1987), scholars tend to assign different names and content to the specific phases. It is common to employ terms that refer to the policy process stages, such as “problem setting,” “direction setting” and “implementation” (Gray, 1989). Understanding collaboration in terms of stages may be important for calling attention to different conditions and strategies in different contexts. It is also argued that collaboration is cyclical rather than linear (Ansell & Gash, 2007) and depends on virtuous cycles between different elements (Huxham, 2003).
Turning to the subfield of collaboration in emergency management, it is common to differentiate between phases such as mitigation, preparedness, response and recovery (McGuire & Silvia, 2010; NGA, 1979), acknowledging that the number of organizations involved as well as the type of interaction will vary in the different phases (McGuire et al., 2010). The interconnectedness of the various phases stems from not only the perceived benefits of the activities in each individual phase (Boin & McConnell, 2007) but also how they are related. To assess this interconnectedness between the phases, we thus need to develop an understanding of the roles of the activities throughout an efficient emergency management cycle, which we will do in the following sections. In the following, we will elaborate on three different aspects of the emergency management cycle—preparation, implementation and learning—which are the phases we analyze in this article.
2.1 Phase 1: Preparation—Planning, designing, and training
Most studies concerned with the phases of collaborative emergency management take planning as the point of departure. A key element here is to make interaction fluid and reduce transaction costs once an incident occurs. Indeed, in their study of the Boston Marathon bombing, Hu et al. (2014) identified pre-event planning in a formalized and institutionalized fashion as crucial for developing shared capacities and promoting shared training and, consequently, more efficient emergency responses. Additionally, the development of informal friendship bonds can aid fluid interaction during the various stages of the emergency management cycle (Kapucu & Hu, 2016). Based on a literature review, Nohrstedt et al. (2018) see antecedents–process–outcome models as promising contributions to explaining successful collaboration in emergency management. They are perceived as crucial for determining the extent to which stakeholders have planned for different scenarios and, in this way, have become familiar with the different resources of nonprofits and government organizations (Simo & Bies, 2007). An underlying premise of these contributions is that preparing for incidents is necessary to achieve a constructive outcome. Furthermore, good planning will reduce uncertainty and make complex interactions in later stages more manageable (Persson & Granberg, 2021).
In addition to planning and designing, training is widely seen as “the heart of crisis preparedness with many potential benefits, but also hurdles to overcome” (Persson & Granberg, 2021, p. 1337). Some studies challenge both the need for training and the viability of prioritizing training for incidents that rarely occur (Boin & McConnell, 2007; McConnell & Drennan, 2006). Yet, these conclusions are based on the crisis management of rare and unforeseen incidents and may not be transferable to a context in which search and rescue operations take place more frequently. Furthermore, an important distinction is that between training within an organization and training among organizations in a collaborative network. In the latter case, a review by Khorram-Manesh et al. (2016) found successful implementation to be based on pre-disaster collaboration developed through training, making it paramount to identify the right form of training. Failure to do so may well be the principal cause of crisis response failures (Boin & Hart, 2010).
2.2 Phase 2: Implementation—Responding to incidents
Implementing an emergency management operation is a matter of different actors coming together to create a public good. Ideally, such an operation should eliminate the waste of resources and efforts via communication, coordination, establishing partnerships and interoperability (Kapucu et al., 2010). The realization of these goals is often dependent on what happens in the other phases.
Implementation is thus at the core of emergency management and is the phase that all the other ones lead up to. Efforts made in the preparation and learning phases are thus means of optimizing the implementation phase. The latter is, accordingly, a reference category, while the assessment of the other phases is based on the extent to which they are important for successful implementation (Nohrstedt et al., 2018).
2.3 Phase 3: Learning
Learning is an important aspect of the endurance and success of collaborative arrangements (Ansell & Gash, 2007; Gerlak & Heikkila, 2011). Still, learning is the phase of the collaborative process that has received the least attention from researchers (Nohrstedt et al., 2018; Sørensen & Torfing, 2021). The role of learning depends on how it is used. One example is evaluation, which can be used primarily as a form of auditing to assess the extent to which predefined goals are met and to identify the contributions of the different participants (Boin & Hart, 2010). Both the selection of relevant measures and which actors should be subject to such evaluations are discussed in the scholarly literature (Kapucu & Hu, 2020, p. 172).
A possibly more demanding form of learning is evaluations that have collective learning as their primary goal. This form of evaluation is also the more cyclical type in that it contributes to the development of planning and training for new incidents (Getha-Taylor, 2019). The ability to learn from experiences in a process of continuous learning is important for building resilience in an emergency management network as it not only facilitates incremental improvements but also helps overcome failures (Sommer et al., 2017).
The following analysis will be organized in line with the three phases explained above: preparation, implementation and learning. Three aspects of these phases are especially relevant to our study. First, while many studies investigate collaborations between public and non-public actors, little attention seems to be paid to the potential differences between the involved actors. Volunteers may need a different form of involvement compared to public professionals. Second, while some research assesses the importance of one particular phase, such as preparation, for a successful outcome, fewer studies have examined the importance of all the phases in the cycle. Third, while some studies question particular activities in the respective phases, most embrace the importance that all the phases can potentially hold in terms of successful emergency management. Accordingly, an understanding exists that the more volunteers and voluntary organizations are involved in the different phases of emergency management, the likelier a successful and enduring collaboration is (Andrew et al., 2015; Peng & Lu, 2019; Rivera & Wood, 2016).
Based on these arguments, we expect that the durability of collaboration is ensured by broad engagement in the sense that volunteers are involved in the emergency management work of public agencies throughout the different phases of the collaboration process.
2.4 Civic involvement by intermediaries
Handling emergency situations often requires resources beyond those available to public authorities (Eller et al., 2015). Combining public, private and civic resources is itself no easy task. Consequently, the role of the public sector in engaging with non-public actors is a central theme in the academic literature on collaborative governance (Ansell & Gash, 2007). Some scholars argue that even when there is no formal hierarchy in place, collaboration will unfold in “the shadow of hierarchy” (Sørensen & Torfing, 2009, p. 236). Other scholars have pointed to the different institutional logics that come into play when public and civil society interact (Thornton et al., 2012). During a collaboration, these institutional logics must be negotiated for the successful implementation of shared policies.
Civic resources can be integrated into emergency management in two significantly different ways: (1) by engaging voluntary organizations that then recruit individuals or (2) by recruiting individual residents directly. In the latter case, individual residents can be recruited ad hoc and only when public resources become too scarce (Johansson et al., 2018), or they can become involved through more permanent arrangements (Krogh & Lo, 2022).
Resident involvement through voluntary organizations is believed to be important for several reasons. It is vital for emergency planning because social relations developed through relevant activities leave local communities better prepared during emergency responses, and such involvement grants access to a broader set of community resources (Andrew et al., 2015; Peng & Lu, 2019). While spontaneous volunteers can be valuable resources in an emergency, when the emergency management capacity is overwhelmed (Daddoust et al., 2021), they cannot be counted on to the extent that the emergency management system can base its organization on such contributions (Waldman et al., 2018). Put simply, it is argued that voluntary organizations are needed to mobilize volunteers in a predictable fashion.
When mobilized, volunteers need to be coordinated to play fully constructive roles. Indeed, uncoordinated volunteers can create health, safety and security issues at the scene and, thus, distract responders from their duties. Furthermore, coordinating volunteers may be something that professional responders are not trained for or do not have time to do during an operation (Fernandez et al., 2006). By mediating the relationship between emergency managers and individual volunteers, the organizations can thus alleviate the responders' task burdens, ensure volunteers do not increase the risk exposures and direct volunteers in how to make constructive contributions during operations.
In addition to coordinating the actions and interactions between different institutional members of the emergency response network, such organizations also coordinate internally. By coordinating the volunteers between emergency incidents, the organizations can establish activities that produce habit strength, which increases the likelihood of retaining these individuals over time (Mullan et al., 2021). Stabilizing who contributes as volunteers in an emergency management network may be a key aspect of ensuring the durability of such networks.
Retaining volunteers over time also facilitates the development of relevant skills. While it is not likely that individual volunteers will have the necessary knowledge to develop or an interest in learning skills useful for potential future emergency scenarios, a voluntary organization can provide such training for its members, increasing the skill levels of its volunteers (Aasland & Braut, 2019a). Furthermore, through the internal process of verifying the skills of individuals, the organizations can increase the trust public managers have in the skill levels of volunteers and, thus, reduce the transaction costs of such interactions (Krogh & Lo, 2022).
Based on these arguments and insights, we therefore expect that voluntary organizations play important roles as the durability of a collaboration is ensured by intermediaries between individual volunteers and the public sector in the sense that volunteers are members of voluntary organizations that provide more stability than single individuals do.
3 CONTEXT OF THE STUDY, DATA AND METHODS
Norwegian emergency management functions are characterized by a combination of fragmentation and integration. On the one hand, many different public organizations, including local governments, the county governor, national authorities, and the military, are included. On the other hand, voluntary organizations and volunteers play a more significant role than in many other systems. It is not only a system based on contributions from individual volunteers, but also one in which, to a large degree, volunteers become involved through their organizations and where voluntary organizations are recognized as partners in formal emergency management networks (Krogh & Lo, 2022; Krogh & Røiseland, 2023). The different actors, including the volunteers and their organizations, are integrated through two main national rescue centers, which lead and coordinate all sea-based and some large land-based emergencies. Most land-based incidents are delegated to ad hoc local rescue centers under the leadership of the local police service (Hovedredningssentralen, 2018).
Rescue helicopters are the only tool the two main national rescue centers fully control when larger emergencies occur. All the remaining resources necessary for a rescue operation must be mobilized and coordinated by the local rescue center. In the public sector, these resources include the military, hospital system, fire services and local and regional governments, which are mandated to provide resources commanded by the rescue center. The non-public resources are to a large degree—and in significant contrast to most other systems—rooted in civil society. The voluntary organizations have thousands of members ready to contribute, and many have specialized skills that the public sector lacks (Krogh & Røiseland, 2023; Prop. 1 S (2018–2019), pp. 298–299).
The Norwegian emergency management system deals with many different types of incidents. According to statistics collected by the two main rescue centers, about 8000 incidents were registered in 2021. The different events included fires and wildfires, missing vessels, strandings, missing persons, snow- and landslides, severe floods, serious traffic and railway accidents and plane crashes (Hovedredningssentralen, 2022). Most of these operations involved volunteers and their organizations in some way, for example, to search for missing vessels or people, search through snow and landslides and fight against wildfires, sometimes by aiding existing public resources, such as the fire brigade, and sometimes by acting as the only available resource. In particular, a set of seven voluntary organizations are deeply involved in search operations on land and in the mountains, including the Red Cross and organizations offering medical services, rescue dogs, flight services, radio communication and cave climbing (Aasland & Braut, 2019b). These seven voluntary organizations have formed the Voluntary Professional Rescue Organizations' Forum (Frivillige Organisasjoners Redningsfaglige Forum – FORF), an organizational umbrella with a central website that has developed manuals for different types of incidents. In some areas, the FORF has also developed into a coordination unit, organizing the different voluntary resources involved.
This study was based on semi-structured interviews with 37 informants from 6 Norwegian collaborative emergency management networks. The 6 networks were spread out geographically and included both smaller localities with populations of less than 3000 and larger cities with populations near 300,000.
When selecting the networks, we first identified three police districts, as the police are core actors in the networks. The districts were selected from different parts of Norway, and we ensured that each district contained both larger urban areas and rural districts with a certain frequency of engagement in search and rescue operations. We had no information on which networks in the country are the most successful, but the overall emergency management model is the same across all the police districts. We thus sought to achieve diversity in terms of the contexts, implying that features that were consistent across the contexts had a general nature (Ragin, 2004). In each district, we selected the largest city as one case network. Additionally, we selected a rural district that had a popular hiking or skiing destination, prompting repeated search and rescue operations. We thus managed to cover important variations in the sorts of challenges typically faced by Norwegian collaborative emergency management districts.
Within each network, we identified the individuals from the various parties most involved in the collaboration. We interviewed police officers who were at the scene when emergencies arose and, thus, carried out the collaboration with volunteers on the spot as well as police officers responsible for nurturing the relationships with volunteers during the other stages. From the voluntary organizations, we interviewed the local leaders most involved with the public sector in all the phases. Additionally, we interviewed those responsible for emergency management in the municipalities involved, and the county governors, who are the regional state representatives that play the role of supervising local governments. In total, we interviewed 10 police officers, 19 representatives from voluntary organizations and 6 employees from municipalities and 2 from the county governor's office. Table 1 sums up the number of interviews per role and emergency management district.
| Case | Interviews | In total | |||
|---|---|---|---|---|---|
| Police | Voluntary organization | Municipal representative | Country governor | ||
| Police district 1 | 4 | 7 | 2 | 1 | 14 |
| Police district 2 | 5 | 5 | 2 | 1 | 13 |
| Police district 3 | 1 | 7 | 2 | 10 | |
| Number of interviews | 37 | ||||
The interviewees all had extensive experience in emergency management work, and all the volunteers had some form of leadership experience in their organizations. Of the interviewees, 5 were women and 32 were men, reflecting the uneven gender balance in Norwegian emergency management. Their ages ranged from individuals in their early 30s to those over 60 years old. Younger people participate in search and rescue operations but seldom reach leadership positions before they turn 30.
- How is it decided who will participate in which situations?
- Are there any disagreements between the actors regarding in which situations or how they should cooperate?
- Are there formal regulations or rules that shape the nature of the collaboration?
- What skills do different actors bring to the collaboration in the various phases?
The interviews were conducted in 2021 and 2022. Due to the pandemic, most of the interviews were performed online, recorded and later transcribed. When analyzing the transcribed interviews, the authors conducted a “structured, focused comparison” (George & Bennett, 2005), where the focus was on the various phases and how they were connected and on the role of voluntary organizations as intermediaries; we also tried to identify alternative explanations for the durability of the emergency management system. We conducted an inductive, thematic analysis where we identified themes related to our research interest of what sustains enduring collaboration. The identified themes were organized according to phases, which also reflected the nature of the interview guide. To ensure inter-code reliability, the authors convened in a workshop setting and compared, discussed and agreed on codes and themes. In Table 2, we present sub-codes and examples for each of the three phases.
| Parent code | Sub-code | Example |
|---|---|---|
| Preparedness | Training | I have been in the rescue service for 32 years, and I can probably count on one hand the times we have been invited [by the police] to exercises. (Interview 7) |
| Formal interaction in planning | Our contact with the police is sporadic, but with one exception. That is when the police train new operational managers. Then we are invited to talk about the voluntary rescue service and what kind of capacities we have. (Interview 30) | |
| Implementation | Mobilization | … the ambulance comes with five people, three of them are operatives and two are managers. We come with 20 people. From then on, it is really us who run the operation for the next 2 days. (Interview 12) |
| Authority | All happens in consultation with the police leader on the spot. It is the police who is responsible, and primarily to ensure the safety of our crews. (Interview 34) | |
| Dependency | Most operations we have are search and rescue where we are very dependent on voluntary organizations. (Interview 35) | |
| Learning | Debrief | After an operation there is a debrief where everyone is called back again and then the police go through how the scenario went. (Interview 30) |
| Institutionalized form of learning | The Red Cross, Norwegian People's Aid, the rescue dogs, and a couple of the others meet on the first Tuesday of the month. We have an evening meeting with the police where we go and look at rescue operations that have taken place in the last month. (Interview 29) |
4 FINDINGS
In this section, we will explore how volunteers and their organizations participate in the Norwegian emergency management system in relation to the three phases explained above. As we analyzed the interconnectedness of the different phases, we assessed the relationship between the public sector organizations and volunteers and the role of the intermediating voluntary organizations.
4.1 Preparation
At the local level, there was some variation among the different regions. The variation ranged from practically no planning beyond the meetings taking place during incidents to networks with regular encounters between the police and volunteers. Where there are meetings to plan actions, this is often done through the voluntary FORF umbrella. A police officer with responsibility for emergency planning explained as follows:We have developed this handbook for the search and rescue of missing persons on land. Recently, it was revised, and the police have been an important contributor. This handbook means that the education of policemen and their training is similar to our training, meaning that the handbook has made us both more effective. Therefore, common planning and common education are necessary (Informant from the Red Cross and the FORF).
In this region, we see that representatives of the volunteers have institutionalized participation in planning activities in the police district. Local volunteers corroborate their involvement in planning activities. At the same time, the other police districts have quite different approaches to planning. In one district, both volunteers and the police reported next to no contact between incidents with only annual meetings. One typical example is the following:The FORF leader in the South-West, he is simply part of the police chief's rescue management in the local rescue center. At an operational level, we have a committee that we call the training committee, which deals with all training activities in the police district and coordinates between all the cooperative actors. Here, the FORF is represented by its leader. And based on that committee, all the training activities that we carry out together with the volunteers are organized. This means that the volunteers help to design how we practice routines and perform planning in the rescue service (Emergency planner in a police district).
In this case, the police report annual meetings for planning activities initiated by the voluntary organizations. This is somewhat paradoxical since both parties point out that during incidents, it is the police that is in command.At least once or twice a year, we have a meeting with the Red Cross about preparations for avalanches and rescues in the mountains. The Red Cross is good at inviting [us] and ensuring we have annual meetings about the handling of such situations (Informant from the police).
Despite the police being the public sector representatives who most frequently interact with volunteers, local governments also have a formal role as planning authorities in emergency management. Among the volunteers, an understanding exists that the public actors, including the police and local government, are responsible for initiating training activities. Both have fundamental responsibilities when incidents occur, while they must rely on volunteers to handle many situations. Still, in our material, the volunteer organizations were often the parties that initiated training, and volunteers do not seem deeply involved in local government planning processes.
There may be several reasons for the lack of training within the network as a whole. As mentioned above, local governments do not seem to recognize this as their role. In addition, practical challenges also exist when public and civic actors operate together. A leader of the Red Cross explained how both work hours and budgets become an obstacle to common training:We have a yearly helicopter exercise in collaboration with the air force that all newcomers must take part in. We train in our methods when using helicopters. In addition, we train in rescue procedures on the ground four to five times a year. However, most of the competencies among the members of the alpine rescue group are acquired by the mountain climbing our members do in their leisure time. Some even have climbing and mountain guiding as a profession (Informant from the alpine rescue group).
Training within the voluntary organizations goes beyond developing precise skills to use in operations. It is also a means of helping volunteers develop habits of engaging in emergency preparedness work. For this informant, this is something that needs to be seen in the context of training:When we train with volunteers, it is normally over a weekend or in the evenings. This often means the police need to pay policemen overtime to take part in the training. Some of the policemen are very engaged, and we have seen policemen meeting to train without payment, simply because they see training as so important for the police. […] However, the police often cannot take part (Informant from the Red Cross).
Summing up the preparation phase and regarding the extent that volunteers are involved in such preparation, it is mostly related to operational planning, such as nationally developed handbooks and manuals developed under the FORF umbrella for different incidents. There is less training within the emergency management networks as a whole, while the different voluntary organizations organize many training sessions, both within single organizations and across organizations and sometimes also together with the police. Notably, we see some variation between the different police districts in terms of including voluntary leaders in their preparations, but, as we shall see, this variation does not seem to have consequences in the implementation phase of the cycle. The most important form of coordination in the preparation stage takes place through the nationally developed handbooks and locally within and between the voluntary organizations and not in their relationships with the public authorities. These findings echo observations that interaction in the preparedness stage is beneficial for emergency management (Brudney & Gazley, 2009), but at the same time, in SAR operations that take place with a certain frequency, it is not vital to continuously engage in network wide training as long as individual skills and collaborative cultured is nurtured in the individual organizations (Boin & McConnell, 2007; McConnell & Drennan, 2006), and established scripts for interaction among organizations is well-developed.It is not optimal if months go by between each search and rescue operation. Therefore, the meetings we organize in between are extremely important. Not only are they necessary for maintaining and developing our skills but they are also important for newcomers in the organization. We need to include them, make them part of our community. This is very hard to achieve if you do not meet regularly (Informant from the Red Cross).
4.2 Implementation
All the efforts in all the phases are carried out to achieve a successful outcome in the implementation phase. This is where literal life and death issues must be resolved. Search and rescue operations are dominant in our material, as they are in Norwegian emergency management in general (Hansen & Winsvold, 2021). Regarding search and rescue, a shared finding in our material across all collaborative networks and from both public sector officials and volunteers was that collaboration leads to success in terms of being able to aid individuals in need of help.
Concerning the mobilization of voluntary resources, we found some variation across the collaborative networks, but a shared feature was that the coordination of the resource mobilization is, to a certain extent, performed by the volunteers themselves through their inter-organizational network FORF:No matter what, we are dependent on the voluntary resources. If you look at the calls for searches for missing persons, then the police will act as leaders […] and the remaining [individuals] involved on the ground will be volunteers. The volunteers can easily amount to 70%–80% of the total resources on the ground (Informant from the police).
The mobilized resources often have skills that surpass those found in the public sector organizations. The voluntary organizations have volunteers who have completed years of training, giving them unique skills. This leader of a voluntary organization described his road to becoming a leader in avalanche incident response:In our region, basically all alarms from health and police will be sent to the so-called FORF-watch. We have volunteers as watchmen: 2 people, 24 hours a day, year round. They receive the alarm message and then communicate the message to the FORF umbrella. When doing this, they also weigh the different resources available in the FORF (Informant from Norwegian People's Aid).
Interestingly, this respondent contrasted the time and devotion put in by him and others in voluntary organizations with what is possible within the public sector, concluding that it is simply not feasible for public employees to reach the level of skill needed in these situations.It requires some years of experience and frequent training. I do not think you can achieve this in, let's say, a local fire brigade. You must have a special interest (Informant from the Red Cross).
Even if the formal responsibility is duly placed with the police, the reality on the ground may vary. There is some variation regarding skill levels among both police officers and volunteers. Experienced police officers will naturally assume leadership in most situations, while less experienced police will grant more leeway to experienced volunteers. Furthermore, certain volunteers have such specialized skills that the police are not capable of determining what constitutes a fruitful course of action.What is extremely important, and something we need to be open about, is who has the full responsibility. It is the police that are responsible through their staff members who are involved in various situations. And the police must consider several needs in search and rescue operations, such as being prepared for this to end up as a crime scene (Informant from the police).
Summing up the implementation phase, it seems there was strong mutual recognition among the participants involved of the interdependency of public and civil resources in search and rescue operations. This interdependency also influences the leadership function in specific actions. While it is clear on article that the police are in charge, in practice, leadership is both shared with and sometimes even outsourced to volunteers in situations where the latter, more than the police, have the necessary experience and competencies. In this way, practical roles often differ from formal ones. This means that while the police are formally in charge of all the volunteers contributing to the operation, they will, in reality, mostly work with the leaders of the voluntary organizations, who, in turn, coordinate their own members in terms of both mobilizing for an operation and carrying it out.
The role of the FORF is also worth noting. This organizational umbrella, formed by seven voluntary organizations, has a special agreement with the emergency services, and the seven organizations are the main pillars of the voluntary sector of the emergency system. It thus represents an additional layer of coordination among the voluntary organizations that further reduces the complexity of working with volunteers for the police, who only need to alert the FORF to access the resources of its seven member organizations. These findings echoes results from a UK study which demonstrated how coordination in the implementation phase of emergency management improves when the communication is decentralized (Brown et al., 2021).
4.3 Learning
As an example of learning, a leader explained that communication is a topic often raised during evaluations:After a search and rescue operation, there is always a debriefing led by the police, where we evaluate the main actions. Individuals who have been exposed to extreme sights are offered follow-up. In addition, in our organization, we have a system intended to help members who have had unpleasant experiences (Informant from Norwegian Rescue Dogs).
Each operation provides an opportunity to learn. However, in one of the networks we studied, evaluation and learning also took place on a more permanent basis and included the leaders of the voluntary organizations. One of the leaders explained as follows:One typical observation is that we are not good enough at communication at headquarters [gives an example]. This is something we work on every time we evaluate: improving communication (Informant from Norwegian Rescue Dogs).
Summing up the learning phase, we observe it is common to debrief after operations, when the equipment, leadership and collaboration are briefly evaluated. The learning activities at the network level seem rather ad hoc, and long-term and systematic processes of learning involving most of the volunteers seem to be lacking. This is in line with other studies of policy learning in general (May, 1992) and in emergency management in particular (Albright & Crow, 2015; Haque et al., 2019), that scaling experiences from routine operation to system level is difficult to achieve. However, within the different organizations, there are many learning activities, and the evaluations of actions seem to blend with ordinary organizational life.Here, in our region, we meet: the Red Cross, Norwegian People's Aid, the association for search and rescue dogs and a couple more. We meet with the police the first Tuesday of every month in the evening. During these meetings, we evaluate any operations from the last month—Did we succeed? Are there any lessons to be learnt? Things to improve? Things that went very well and that we intend to do more often? (Informant from Norwegian Rescue Dogs).
5 CONCLUDING DISCUSSION
Our case is an example of an emergency management system that, according to all the parties, functions effectively in achieving its main goal: searching for and rescuing people. This subjective view of the network participants is backed up by objective data that demonstrate how vast resources are mobilized, far beyond what the public sector musters for this public task (Prop. 1 S, 2018–2019, p. 299). Furthermore, the collaboration is described as smooth, with mutual recognition and trust in task solving. Additionally, the collaborative network has a stable and long-standing capacity that is frequently mobilized to carry out operations.
Based on the extant literature, we formulated two expectations for such an enduring, network-based emergency management system. First, we expected broad participation from volunteers in all phases of the collaboration. Second, we expected that the durability of collaboration is ensured by intermediaries between individual volunteers and the public sector. Surprisingly, we did not see broad engagement between public sector agencies and volunteers outside the implementation phase. Following on from similar studies in other national contexts, this should mean that the actual response becomes less effective (see, e.g., Kapucu & Hu, 2016; Peng & Lu, 2019). However, what we saw in all our cases was fluid and integrated interaction in the implementation phase and unsystematic variation between the cases in terms of both preparation and learning. While involvement in the preparation and learning phases varied among the cases, this variation is not, apparently, reflected in the implementation phase.
One clue for understanding this apparent paradox may be the role of intermediaries that facilitate different levels of coordination. The Norwegian emergency management system is inherently polycentric, with different levels and sources of authority. The individual volunteers are coordinated at different levels, gradually decreasing the complexity of the interactions in the implementation phase.
At the ground level, volunteers are recruited, trained and approved by individual organizations. These organizations have highly institutionalized structures that ensure the quality of and compliance with these processes. The different organizations thus ensure that individuals are prepared with the basic skills for both search and rescue and interacting with other organizations. Much of the preparation and learning is thus done separately in each organization.
At the second level, the voluntary inter-organizational network FORF coordinates the voluntary organizations, further reducing the complexity of the collaboration. At this level, joint planning and training are carried out. In addition, through the FORF, the voluntary organizations can approach the public sector with one voice, facilitating collaboration with the entire network as the public sector gets involved in the implementation phase.
The third level of coordination is demonstrated in the implementation phase, where the volunteers, through their voluntary organizations, collaborate with the police to solve tasks. At this point, prior collaboration among volunteers is a prerequisite for successful collaboration in the entire network. This entails various levels of authority: organizations, the FORF umbrella and the public sector. The durability of the network is based more on the volunteers fulfilling the ascribed duties in each phase than being strongly integrated into the various phases.
The broad involvement of individual volunteers in all phases of emergency management would thus not square with a civil society model in which such interaction is mediated through voluntary organizations. The organizations play the pivotal role of training, retaining and mobilizing volunteers; thus, more direct interaction between the public sector and volunteers could undermine this role of the organizations. However, the perspectives of each party are broadly present in all the phases. This is achieved through the national development of a shared framework for emergency management—for example, the abovementioned handbook for search and rescue—and through formal and informal meetings between leaders from the police and the voluntary sector.
Understanding the dynamics of the different phases is thus key not only to understanding how coordination takes place in each phase but also to grasping the roles of the different levels of analysis. Indeed, the phases are important for the outcome of the implementation, but to understand this, we need to look at the levels of authority and coordination and not solely at the functioning of the entire network. This can unmask important features of emergency management in the different phases. For example, as mentioned in the theory section, an unresolved issue in the literature concerns the importance of common training (Boin & McConnell, 2007; Persson & Granberg, 2021). Our case suggests that preparation is indeed paramount but does not necessarily need to include the entire collaborative network. If only parts of the network prepare and train, this might be more useful for developing specialized skills, while preparation and training for the whole network can often be focused on the collaboration and not necessarily on the core skills themselves.
For the study of emergency management, in theoretical terms, our study indicates that the cyclical nature of collaboration, in which the different phases together form the outcome, needs to be expanded to encompass how intermediating actors can virtually create a new institutional design in each phase. While each phase plays an independent—yet interrelated—role in achieving the outcome, the institutional dynamics in one phase do not determine those in a different one. To understand these dynamics in collaborative emergency management, one must understand and take into account the characteristics of civil society in a given system. This is a reminder that comparing emergency management across different national systems is extremely challenging, although, we argue, still important.
Referring to the larger literature on collaborative governance—for example, as summed up by the now classic review of Ansell and Gash (2007)—this study is a reminder of the complexity of collaborative governance. Not only are there different phases with their distinctive dynamics, there are also different levels of coordination that must be accounted for to understand the dynamics behind an enduring collaboration. This implies that establishing universal theories about collaboration across different cultural and national contexts is difficult and more complicated when compared to theories related to the public sector only. Digging deeper into these conditions and formulating theories and models that can be applied in empirical research across different national contexts are important tasks for future research.
Finally, our study has some limitations that should be mentioned. Our qualitative material put limits on our ability to generalize beyond the units we have studied, and the Norwegian context may deviate in significant ways from other national contexts. Therefore, generalizability of our results to other contexts should be avoided.
ACKNOWLEDGMENTS
This research work was funded by Research Council, Norway (Grant # 296064). The authors are grateful for valuable comments on earlier versions from Marte Winsvold, Bernard Enjolras, Dag Wollebæk, Vibeke W. Hansen, Christian Lo, Andreas H. Krogh, and Arild Wæraas, as well as from the three anonymous reviewers.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.
Open Research
DATA AVAILABILITY STATEMENT
The data applied in this study are available on request from the corresponding author. The method for collection of data has been approved by SIKT Data Protection Services, Norway. Data are not publicly available due to privacy or ethical restriction.
