Contextualizing supply chain risk governance in critical infrastructure sectors: insights from the Swedish food system

Abstract

Purpose

Critical infrastructure (CI) sectors and their resilience are vital for societies to function. In many countries, vital societal functions (VSFs) and CIs depend on inter-organizational and international supply chains (SCs) which combine public and private actors with often competing interests and unclear responsibilities that create discontents. While collaborative supply chain risk management (SCRM) can increase the robustness and resilience of VSF&CIs, their inherent characteristics complicate SCRM. To understand this, supply chain risk governance (SCRG) has conceptually been introduced, suggesting collaborative mechanisms that facilitate inter-organizational SCRM. The purpose of this study is to elaborate on and substantiate the theoretical and practical relevance of an existing SCRG framework, by empirically exploring governance of collaborative SCRM and suggesting future research.

Design/methodology/approach

An abductive case study was performed in a VSF&CI, the Swedish food system, to contextualize top-level governance of collaborative SCRM and elaborate on the conceptual SCRG framework. Archival data supplemented expert interviews with public and private actors representing direct and indirect SC actors.

Findings

Current discourse and interventions in the Swedish food system’s SCRM confirmed discontent in governance and the importance of SCRG mechanisms and supported further conceptualization, e.g. of legal mechanisms vs economic incentives, the importance of government inquiries and the understanding of the influence of indirect public SC actors.

Research limitations/implications

Considering the geographical scope of the study, transferability is limited but invites comparisons with both other countries and the SCRG of other VSF&CIs. Proposed research avenues guide future conceptualization and contextualization of SCRG.

Practical implications

The framework can support CI actors to jointly find and reduce discontents related to inter-organizational SCRM and support policymakers to increase public–private collaboration.

Originality/value

The novelty lies in empirically studying SCRG in critical infrastructures by combining SCRM and risk governance lenses.

Citation

Norrman, A. and Eriksson Ahre, E. (2024), "Contextualizing supply chain risk governance in critical infrastructure sectors: insights from the Swedish food system", The International Journal of Logistics Management, Vol. 35 No. 7, pp. 33-59. https://doi.org/10.1108/IJLM-10-2023-0444

1. Introduction

Critical events in recent years, such as the COVID-19 pandemic, increased geopolitical tensions and transportation disturbances have accentuated the importance of global supply chains (SCs) for society and its vital functions. COVID-19 influenced supply shortages and demand spikes for medicine and medical supplies (Suda et al., 2022; Ahlqvist et al., 2023) and for food SCs (Kumar et al., 2021) especially, demand shocks and supplies of fresh and perishable products (Coluccia et al., 2021) and logistics services (Hohenstein, 2022). Medicine shortages were noted as an increasing global problem even before the COVID-19 pandemic (de Vries et al., 2021) and continued afterwards (Pharmaceutical Group of the European Union, 2023).

The Russia–Ukraine war, and related sanctions, have had an impact on trade in general but on food supply and food prices in particular as the countries involved are important producers of cereals, vegetable oils and fertilizers (Glauber and Laborde Debucquet, 2023). Jagtap et al. (2022) argue that the impact on the effectiveness and responsiveness of global food SCs will be significant. The Russia–Ukraine war has also impacted energy supply, especially natural gas. To exemplify transportation disturbances, the trade route through the Suez Canal regularly faces problems with piracy incidents and can become blocked (e.g. by the Ever Given ship in 2021). Disturbances create delays and freight price increases, illustrating the vulnerability of global SCs.

Foodstuffs, medicine, transportation and energy are defined by the USA and the EU as vital societal functions (VSFs) or critical infrastructures (CIs). The European Council (2008) defines critical infrastructure as an asset or system which is essential for the maintenance of vital/critical societal functions. Their damage, destruction or disruption will have a significant negative impact on a society's security and the well-being of its citizens. Definitions of VSF and CI have developed over time and vary across countries (Pursiainen, 2009; Moteff and Parfomak, 2004) but tend towards a broad, holistic understanding: networks of systems of vital functions of society as a whole or the infrastructures of those systems (Pursiainen, 2009). From 16 January 2023 (EU, 2022), the EU's member states are to adopt a national strategy and carry out regular risk assessments to identify entities considered critical or vital for society and the economy. The two terms are often combined into VSF&CIs (Swedish Civil Contingencies Agency (MSB), 2014). The difference between the two terms is that VSF is more general, depending on but also encompassing infrastructures (Pursiainen, 2018).

VSF&CIs often depend on inter-organizational and international SCs, combining public and private actors with often competing interests, limited overviews and unclear responsibilities. A disruption in one part of the SC can cause ripple effects (Norrman and Jansson, 2004; Ivanov et al., 2014) and affect other parts. Due to complex relations and dependencies in VSF&CIs, many risks cannot be managed by a single actor but require collaboration (Renn, 2008).

Despite often having collaborative intentions and being defined as such (e.g. Norrman and Jansson, 2004; Fan and Stevenson, 2018; Friday et al., 2018; Friday et al., 2021), both in research and practice, supply chain risk management (SCRM) has mostly concentrated on a single company’s risk management work as the unit of analysis (Ho et al., 2015; Fan and Stevenson, 2018). While improved SCRM can increase a VSF&CIs robustness and resilience, the inter-organizational characteristics of VSF&CIs make collaborative SCRM both difficult and very important.

SCRM was originally developed for private companies, focusing on potential strategies that companies (can) adopt in preparedness for potential disruptions (Ho et al., 2015; Ahlqvist et al., 2023). This is consistent with extant SCM literature largely being limited to relationships between for-profit firms while neglecting the government perspective, which needs more attention (Quarshie and Leuschner, 2020). Research gaps are increasingly identified in the public–private interface in SCs and its governance. Recent reviews highlight the need for understanding inter-organizational governance and collaboration between public and private, direct and indirect, actors in SCRM (Ahlqvist et al., 2020; Friday et al., 2021; Tsolakis et al., 2023). Scala and Lindsay (2021) argue that CI SCs can strengthen their resilience through public–private partnerships. Sodhi and Tang (2021) call for research on the growing role of government interventions needed for critical products. Darby et al. (2021) and Pournader et al. (2020) also call for more research to explore the role of policymakers in SCRM. However, Pournader et al. (2020) also find that government interventions further complicate SC responsiveness and coordination, adding another layer of risk. They suggest research adopt a multi-level approach, incorporating individual, organization, SC and actor interconnections and their impact on SC resilience.

Ahlqvist et al. (2020) synthesized different concepts and perspectives from different sources and introduced a conceptual multi-level framework for supply chain risk governance (SCRG). The framework was grounded in interdisciplinarity and combined elements from SCRM and the domain of risk management and societal safety, a domain concerned with societal risks and how public organizations deal with emergencies but without a focus on private actors and a SC perspective. It highlights the need to increase inter-organizational SCRM work in general, and specifically for public–private global SCs of VSF&CIs but is purely conceptual and has not yet been explored empirically. In their SCRG framework, Ahlqvist et al. (2020) introduce an SCRG layer that emphasizes inter-organizational collaborative governance mechanisms as operationalized elements of governance. Thus, governance is used as a concept to broadly describe an institutionalized decision-making process spanning many independent actors, and governance mechanisms are seen as more operationalized concepts that direct the behaviours of various actors. It chimed with recent general interest in SC governance mechanisms (e.g. Tachizawa and Wong, 2015; Dolci et al., 2017; Bonatto et al., 2022; Gemente et al., 2024), but which not focus either on SCRM nor the public–private interface.

This study aims to fill the research gaps discussed above, including the need for more empirical studies, and to increase understanding of the phenomenon of governing collaborative SCRM (especially in public-private VSF&CIs). The purpose of this study is to elaborate and substantiate the theoretical and practical relevance of an existing SCRG framework, by empirically exploring the governance of top-level collaborative SCRM, and to suggest future research.

To contextualize the phenomenon of SCRG in the VSF&CI context and elaborate on Ahlqvist et al.’s SCRG framework, a case study approach (Ketokivi and Choi, 2014; Schwarz and Stensaker, 2016) was used to gain critical insights into context and real-life challenges. We studied the Swedish food system’s high-level governance of SCRM, with direct and indirect as well as public and private actors as informants.

Our study offers an original perspective on the governance of collaborative SCRM in the public–private setting in VSFs/CIs and contributes to prior SCRM research in many ways. Firstly, it responds to calls for more empirical research on collaborative SCRM and SCRG and illustrates multi-level complexity in practice. Secondly, novel theoretical propositions are generated that help researchers and managers understand the interactions between SC governance mechanisms, collaborative SCRM and public–private SC structures. Thirdly, by researching the SCs of public–private actors, including the government and policymaker perspective, it invites future SCRM research in this context. Fourthly, Ahlqvist et al.'s (2020) multidimensional SCRG framework is refined with the addition of new SCRG governance mechanisms and additional roles of public actors. Although it focuses on the food system, our study is relevant for other sectors as well, showing how different actors can understand and improve collaborative SCRM.

Below, we present the theoretical underpinnings, including the guiding conceptual SCRG framework in Section 2. This is followed by a description of our research design in Section 3. Section 4 discusses empirical observations, and we elaborate on the SCRG framework and submit propositions in Section 5. Finally, we present conclusions and limitations and suggest issues for future research in Section 6.

2. Guiding conceptual framework

2.1 Supply chain risk governance

Ahlqvist et al.'s (2020) multi-level conceptual SCRG framework, which guides this research, was synthesized from extant literature from the domains of SCM and risk management and societal safety, combining concepts related to SCRM and risk governance. The term risk governance (Arvidsson and Cedergren, 2016; Dubreuil, 2001; Gheorghe et al., 2007; Renn, 2008; Sajeva and Masera, 2006) describes how multiple actors are affected by risks and involved in risk management, and how this should enable inter-organizational risk evaluation and collective decisions to handle complex, uncertain and ambiguous risk (Gheorghe et al., 2007; Klinke and Renn, 2012; van Asselt and Renn, 2011; van der Vegt, 2018). However, Ahlqvist et al. (2020) find that this domain focuses less on private actors and especially does not take into account the SC perspective of following flows upstream and downstream. Applying a systems approach and combining public-private perspectives, Ahlqvist et al.'s (2020) conceptual SCRG framework, as shown in Figure 1, illustrates three layers: the SC including actors and flows through nodes and links; the SCRM work done individually or jointly by different actors; and governance mechanisms influencing the degree of collaborative SCRM.

The bottom SC layer illustrates the SC structure and flows, including direct and indirect actors (Mentzer et al., 2001; Carter et al., 2015), but also differentiates between private and public actors. While direct actors are physically related to the critical flows, indirect actors can influence the VSF&CI flows in other ways, such as by creating platforms for interactions (e.g. industry federations) or intervening through policies and decisions (e.g. various authorities at sectoral, regional and local levels). This mapping of the SC structure, in line with guidelines for SC mapping (Gardner and Cooper, 2003; Henkow and Norrman, 2011; Fabbe-Costes et al., 2020), is useful for understanding complex, multi-actor SCs with different types of actors whose responsibilities and involvement might vary over time and depending on the risk situation. While Carter et al.'s (2015) SC theory included supportive actors such as financial institutions and forwarders, Ahlqvist et al.'s (2020) SCRG framework extends this by suggesting that it is important also to understand influencing and intervening actors.

The SCRM processes layer illustrates risk management processes (including, e.g. proactive risk identification, assessment, mitigation and monitoring as well as reactive actions – as described in numerous SCRM articles (e.g. Norrman and Jansson, 2004; Fan and Stevenson, 2018). These may be put in place collaboratively or separately by diverse public and private actors. The SCRG framework highlights the fundamental idea that increased collaborative inter-organizational SCRM can improve the SC's capability to manage risk (e.g. Norrman and Jansson, 2004; Fan and Stevenson, 2018; Friday et al., 2018, 2021). Examples might be, e.g. whether and how risk mitigation strategies of using extra stock and its positioning are used and coordinated or the practice of risk identification and mapping and whether actors do this in isolation or jointly.

Ahlqvist et al.'s (2020) conceptual SCRG framework focuses on the top-level layer. This operationalizes the supply chain risk governance structure into distinct mechanisms (e.g. joint understood context and concern, aligned strategic goals, collaborative performance measurement system, collaborative information sharing and communication, inter-organizational institutional mechanisms, collaborative and transparent decision-making). These more operationalized SCRG mechanisms resemble what other authors term SC collaborative risk management capabilities (Friday et al., 2021) or interface variables (e.g. Simatupang and Sridharan, 2005; Friday et al., 2018). Ahlqvist et al. (2020) found unique mechanisms suggested both in the SCM literature and in the risk governance literature, but most concepts overlapped across domains although different terms were used. In general, the risk governance literature focused on societal elements at a higher level, while the SCRM literature focused more on operational mechanisms for process integration in the SCRM layer.

2.2 Governance and governance mechanisms

Governance is a higher-level concept describing an organizational construction or institutional framework, while governance mechanisms are the concrete management and control activities (Ahlqvist et al., 2020). Governance mechanisms describe how the organization's required behaviour will be motivated, influenced and established. They are thus more “administrative tool[s]” (Martinez and Jarillo, 1989) and represent an actual operative practice between the parties. Governance mechanisms are sometimes divided into formal and relational, which complement each other (Poppo and Zenger, 2002). Hoetker and Mellewigt (2009) describe formal governance mechanisms that specify each party’s role, performance expectations and dispute resolution mechanisms, such as business plans, service level agreements, prices and performance indices, as well as ways of monitoring the partners, such as economic efficiency calculations and reporting requirements. Relational governance mechanisms mitigate potential opportunism by building trust and social identification, and Hoetker and Mellewigt (2009) provide examples of establishing teams, task forces and committees; direct managerial contact through trips, meetings and even the transfer of managers; mechanisms for shared decision-making; and formal systems for conflict resolution that rely on two-way communication and joint problem-solving. Recent literature on supply chain governance uses this terminology but is dominated by literature reviews and focused on conceptual discussions of high-level formal and relational mechanisms (e.g. Ghosh and Fedorowics, 2008; Tachizawa and Wong, 2015; Bonatto et al., 2022; Gemente et al., 2024), while some focused on trust (Ghosh and Fedorowics, 2008; Bonatto et al., 2022) and Dolci et al. (2017) added transactional mechanisms and tested their influence on performance.

The guiding SCRG framework (Ahlqvist et al., 2020) suggests the two governance mechanisms strategic alignment and collaborative performance measurement systems aim to ensure a shared commitment between the supply chain actors. The strategic alignment of actors is crucial, e.g. the strategic goals of public authorities and private firms in relation to other actors’ values, society, political ambitions and ethical and legal issues (Cao et al., 2010; Sajeva and Masera, 2006; van Asselt and Renn, 2011). Collaborative performance measurement systems should follow the strategic intentions (Friday et al., 2018; Simatupang and Sridharan, 2005). Whether and how key performance indicators are developed, implemented and used between actors are important. They guide other governance mechanisms that aim to increase collaboration and coordination (Simatupang and Sridharan, 2005). The SCRM framework clusters a set of formal governance mechanisms, including legal frameworks, policies and contracts (Dubreuil, 2001; Fan and Stevenson, 2018; Renn, 2008; Urciuoli et al., 2014) under the term inter-organizational institutional mechanisms. These assist in defining accountability and responsibility as well as aligning incentives and creating goal congruence among the actors.

In parallel with the inter-organizational institutional mechanisms, other more relational governance mechanisms such as collaborative information sharing and communication and collaborative and transparent decision-making should increase collaboration and coordination (Simatupang and Sridharan, 2005). Collaborative information sharing and communication are characterised by openness, visibility and the exchange of relevant data (related to both critical flows and risk management processes) (Bak, 2018; Cao et al., 2010; Govindan and Chaudhuri, 2016; Li et al., 2015; van Asselt and Renn, 2011). Collaborative and transparent decision-making (Dubreuil, 2001; Friday et al., 2018; Sajeva and Masera, 2006) describes how joint and coordinated decision-making can be facilitated and take place both pro-actively and reactively. These SCRG mechanisms aim to enable more integration of the often-segregated SCRM processes (in the SCRM layer).

The risk governance literature adds the importance of having mechanisms for establishing a joint pre-estimation of the context (Klinke and Renn, 2012; Renn, 2008), which can influence an aligned strategic focus. The involvement of many actors with different perspectives makes it important to jointly understand the political and legal contexts, cultures, values, ethics, etc. of various actors. The public also constitutes an important actor as its “votes” can be important for actors and decision-makers who are politicians. While this type of governance mechanism was barely discussed in the SC governance or SCRM literature, the risk governance literature provided an understanding of the societal context and contributed by pointing out important antecedents to facilitate interaction to obtain more coordinated and joint SCRM, characterized by a level of trust, legitimacy, social competence, respect and formal and informal networks (Boholm et al., 2012; Dubreuil, 2001). Finally, Ahlqvist et al.'s (2020) SCRG framework included mechanisms for collaborative learning to stress the importance of developing and improving SCRM processes both internally and between actors (Cao et al., 2010; Dubreuil, 2001; Kache and Seuring, 2014; van Asselt and Renn, 2011).

3. Research design and method

3.1 Research design

This research on the phenomenon of top-level governance of collaborative SCRM, and its currently evolving risk context for VSF&CIs (pandemics, geopolitical situation, climate change and natural hazards) followed the phenomenon-driven research tradition and used qualitative methods (Schwarz and Stensaker, 2016). Also, the novelty of the term SCRG, the aim to extend existing insight into a complex and multifaceted phenomenon, and our ambition to study an emerging and contemporary phenomenon in its real-life context, motivates a case study (Meredith, 1998; Yin, 2014). To substantiate and “reshape” Ahlqvist et al.'s (2020) conceptual framework with empirical data and serve as a springboard for future research (Imenda, 2014), this case study research exhibits features of both theory-generating and theory-elaborating approaches (Ketokivi and Choi, 2014). The study thus aims to theoretically contextualize an emergent conceptual SCRG framework (Ahlqvist et al., 2020), adapt it to contemporary practices and challenges (Ketokivi and Choi, 2014) and contrast, specify and further structure its concepts. In short, we sought to discover how Ahlqvist et al.'s (2020) SCRG framework holds for studying top-level supply chain risk governance within the chosen context (the VSF&CI being the Swedish food system) and extend it where necessary.

To obtain an empirical foundation, we designed a single embedded case with the unit of analysis being top-level supply chain risk governance in VSF&CIs and with different supply chain risk governance mechanisms as subunits of analysis.

We examined different relevant actors' perspectives on current SCRG and their understanding of the reasons behind and implications of recent changes and interventions. To develop context-specific conceptualization, the single case study allowed us to explore geographical and temporal elements that gave important insights into organizations’ contexts and real-life challenges (Flyvbjerg, 2006).

3.2 Selection of case and informants

Firstly, the case has to be selected. Sweden has defined 11 societal sectors with VSF: Energy supply; Financial services; Trade and industry; Health, medical and care services; Information and communication; Municipal technical services; Foodstuffs; Public administration; Protection, safety and security; Social security; Transport (Swedish Civil Contingencies Agency (MSB), 2014, p. 13). The researchers’ starting point in logistics/SCM directed a purposeful case selection of a VSF more characterized by physical flows than by immaterial flows or services. Such a theory-driven confirming case (Miles and Huberman, 1994) taking the physical product flows as its starting point made us elaborate on the initial framework. We selected Foodstuffs as we judged it to be less complex and diversified than trade and industry or health, medical and care services. To explore the interface between public policymakers and private actors involved in top-level governance, we purposefully selected the top-level governance of SCRM in the Swedish food system as a case. The food system also mirrors general characteristics in VSF&CIs, such as the mix of public–private actors and globalization but also an on-going discourse on governance issues with potential interventions being investigated, suggested or implemented.

Secondly, informants must be selected. To understand inter-organizational dimensions, we aimed to obtain a theory-driven and confirming representativeness of perspectives from relevant public and actors along the SC. According to the Swedish Civil Contingencies Agency (2014, p. 14), the VSF for foodstuffs includes distribution, primary production, inspections and manufacture of foodstuffs, activities that we refer to as the food system. The target entities that own or operate VSF&CI are municipalities, county councils, county administrative boards, national authorities and private sector operators (Swedish Civil Contingencies Agency (MSB), 2014, p. 9). They represent perspectives in the guiding framework as public–private and direct/indirect actors. Insight into the top-level SCRG came from relevant types of actors such as (Table 2): two municipalities, two county councils and county administrative boards, the two main national authorities involved (the Swedish Food Federation and the Swedish Civil Contingencies Agency [MSB]), a representative of an ongoing governmental inquiry (Committee of the Swedish government inquiry N 2022:02 “A new food preparedness”) and six representatives of the private sector. The food system is large and diversified with many private manufacturers and distributors of different sizes and with different characteristics, but Swedish firms are organized in different industry federations which represent them in discussions with other entities. We included the two industry federations (the Swedish Food Federation and the Swedish Food Retailers Federation) as informants. From the very concentrated Swedish grocery retail sector, we included two of the biggest retailers (together representing 70% of turnover) to understand direct actors. From food manufacturing, two manufacturers were chosen to represent insights on how global firms with Swedish production view SCRG in Sweden. We selected one of the largest and most diversified food manufacturers which produces all kinds of food (seafood, meat, vegetarian, bread, beverages, desserts, snacks, candy, etc.), has many different Swedish factories and sells through different channels both within and outside Sweden. The other produces a very critical foodstuff, selling finished products both to many other food manufacturers and to consumers via retailers. In this case, the Swedish factory network has gradually become highly consolidated, making it critical. Swedish farmers, representing primary production of food, are organized in the Federation of Swedish Farmers (LRF), which we interviewed. Board members and officers working at LRF are normally farmers themselves, as was the person we interviewed. Insights from LRF thus both represent primary producers and align with a top-level perspective. Similarities and differences between the actors interviewed support the theory-elaborating nature of this study (Miles and Huberman, 1994; Fisher and Aguinis, 2017), mitigating possible problems with our single case selection.

Thirdly, informants were sampled to maximize conceptual insights and information richness (Flyvbjerg, 2006); the approach was information-oriented rather than looking for generalization properties (Patton, 1990). Informants were selected to adhere to the boundaries of our research purpose (Miles and Huberman, 1994), to represent diverse types (Voss et al., 2002), and based on their potential to provide details about the different elements of the phenomenon investigated. Respondents from different perspectives (public-private, direct/indirect) increased confirmability and conceptual insights and understanding. We contacted the different organizations with a request to interview the person responsible for SCRM and/or involved in inter-organizational forums within the food system related to the topic. The participation list of the most important national forum set up (known as SKAL) also helped us identify key informants. To obtain rich insights, informants selected from municipalities and counties were pointed out by other informants as being very knowledgeable and engaged in pioneering risk management work related to the food system from a local public perspective.

3.3 Data collection and analysis

In planning and performing the case study, we followed general guidelines proposed by Yin (2014), Gioia et al. (2013) and Voss et al. (2002). The conceptual SCRG framework (Figure 1) guided our data collection and analysis, and the research protocol included a semi-structured interview guide (Appendix 1). To understand the recent discourse, it focused on development as well as challenges related to SCRG, along with motivations for interventions and their implications.

Multiple data sources (Meredith, 1998) and perspectives (public–private, direct/indirect) were triangulated. The informants from different organizations (Table 1) gave primary data capturing aspects of SC and SCRM work but focused on issues related to discontent and different SCRG mechanisms. Interviews lasted between 90 and 120 min and were recorded and transcribed. To help informants prepare (Voss et al., 2002), we shared the themes of the questions before the interviews.

Archival data (websites, inquiries, news articles, white papers, investigations and annual reports) from public authorities, industry associations and private actors supplemented the interviews. This triangulation was important for content analysis and to validate interview data. Of special importance were legislation, government bills and government inquires that illustrate how policymakers' perspectives and ambitions have changed over time, as well as showing formal directives. Using a case study database strengthened a formal chain of evidence (Voss et al., 2002).

To elaborate on the theory, data analysis proceeded through abductive iterations between the framework and the data (Ketokivi and Choi, 2014), which typically include interpreting or recontextualizing individual phenomena to improve previous and current understanding (Kovács and Spens, 2005). Capturing informants' perceptions helped us explore the differences that existed between actors, and later to reconcile findings within the broader perspective of the overall case study (Voss et al., 2002).

The analysis included concurrent data reduction, data display, interpretation, coding and drawing and verification of conclusions (Miles and Huberman, 1994) (first individually then jointly by the two researchers). The initial first-order codes were refined to second-order themes that emerged from the data (Gioia et al., 2013) and moved data to a more theoretical and abstract level of interpretation. To identify relevant elements of SCRG mechanisms, we first openly coded the transcripts to display critical events, interventions and phases of changed SCRG. In cross-informant pattern matching, some categories emerged as being most interesting in terms of discontents and development of SCRG. Emerging patterns were systematically compared, e.g. results from the analysis with Ahlqvist et al.'s (2020) conceptual SCRG framework and the information our informants provided (Gioia et al., 2013). Similarities and differences between the different actors were coded and contrasted (Miles and Huberman, 1994) to elaborate the theory. Using the conceptual framework as a starting point increased the study’s theoretical calibration, i.e. the degree to which theory is captured in the research design. By systematically combining empirical evidence with matching literature, we improved understanding and extended the prior framework (Kovács and Spens, 2005).

3.4 Research quality and trustworthiness

We applied different actions (Table 2) to increase research rigour and trustworthiness (transferability, credibility, dependability and confirmability (Lincoln and Guba, 1986). Despite the measures taken to increase transferability, generalizability limitations naturally exist when using a single case study, as will be elaborated on when discussing limitations and further research avenues.

4. Observations of SCRG discontents and mechanisms

We observed discontents related to SCRG and SCRM and examined how different SCRG mechanisms can support or act as barriers towards collaborations and more resilient SCRM in a national food system. To understand which SCRG mechanisms actors perceive to be important, we observed mechanisms that are already established or under implementation, but also mechanisms actors judge to be needed to further address discontents. All informants confirmed that the discourse on collaborative SCRM had increased recently and described interventions carried out or planned.

Representatives from public and private actors agree that ongoing global trends (e.g. the COVID-19 pandemic and Russia–Ukraine war) first highlighted the need for, and then created, a joint understanding of the discontents at hand which was previously lacking (such as lack of an SC overview, clear goals, clear decision-making and especially a lack of food supply preparedness such as stockpiling (4C Strategies, 2020; Dir. 2021:65; Dir. 2022:33; Dir. 2022:72). They further stressed the importance of public–private collaborations.

This awareness increased the need for more forums and ways to meet, both within and across public and private sectors and share experiences and information. Since 2016, the Swedish Food Agency has established a forum, called SKAL, which facilitates meetings between private and public actors. The forum first met twice a year, but the COVID-19 pandemic created a need to meet more often and frequency then further increased with the Russia–Ukraine war. Now the SKAL forum meets every or every second week and has more members (up to 80 participants). Informants highlight the SKAL forum as their primary meeting point and all are pleased with how it works: “It works very well in my opinion. They have done a great job; they are very far ahead” (C_publ).

The SKAL forum has two main objectives. One is more operational, intending to help actors manage incidents they are facing or experiencing. One example cited by several informants was how the Swedish Food Agency permitted an exception to be made to food safety legislation when the Russia–Ukraine war caused a sudden shortage of sunflower oil: “They had printed packaging for millions of bags. It says sunflower oil, but they are having to fry crisps in rapeseed oil instead. Can we still release this? Then we made an exception [from the legislation] and said that in this case, we can accept a correction label. Because there is no health risk” (G_publ). For private actors, this increased their trust in the public authorities to be flexible when needed: “The fact that the Swedish Food Agency could decide quickly on an exemption to the legislation increased the industry’s trust that the authority can find solutions if needed” (D_priv).

The second objective relates to information sharing. The food system is in a phase where much effort and work revolve around high-level information sharing and jointly describing and understanding the current situation, facilitated by the SKAL forum: “… what is working, what is not working? What do we fear will not work? What are we hearing from our different sides and corners? Everyone has their networks, of course. Not least those in business who monitor what is happening with their suppliers very well. Both within and outside the country’s borders” (G_publ).

However, several informants (e.g. A_priv, F_Priv) point out that information sharing takes place at a too high and too broad a level and does not get down to the details that retailers, for example, need to discuss to get something done operationally. One perceived barrier to more concrete joint SCRM actions which was highlighted by several informants (e.g. E_priv, G_publ) is uncertainties related to what information they are allowed to share under competition law: “if it's going to be really crisp and concrete, then we need to talk about things that we might not be allowed to talk about. Absolutely, it’s an obstacle when we come down and want to talk actual damn facts. But it’s all more of these overarching general processes when you would like to go much deeper on certain matters” (F_priv).

One insight from interviews and archival data is that government-initiated inquiries and industry investigations are important mechanisms to create joint contextual understanding and start aligning overriding goals. Several government inquiries are currently ongoing, indicating high attention from the political leadership. These inquiries will guide the government's continued work and establish prerequisites for joint strategies and goals. Informants especially highlight the recently launched food preparedness inquiry (Dir. 2022:33) (led by B_publ) as being very important for future food system resilience work: “You have to see that one (Dir. 2022:33) as a further increase in ambition because it will put forward a proposal for steering mechanisms and some funding principles and suchlike” (E_priv). At the same time, industry organizations (e.g. E_priv and D_priv) are increasingly publishing reports. These reports allow private actors to establish (and lobby for) their goals and expectations. Besides collecting and compiling information, inquiries and investigations create platforms for meetings (expert interviews) and information exchanges (consultation). However, the current large number of public inquiries is also stressful for the private sector: “We receive an incredible number of requests to participate in investigations. It’s quite … The requests come from regions, it can be municipalities, counties, authorities in the form of the Swedish Food Agency on behalf of Swedish Civil Contingencies Agency and so on. That whole thing isn’t so damn good for us right now” (F_priv).

All informants expressed a lack of common goals and joint strategies regarding what a future food system’s resilience should look like as being a barrier to moving beyond the high-level information sharing of today, to, e.g. sharing detailed information on stock levels or capacities to better coordinate SCRM strategies along the chain. The expert authorities lacked clear goals and directions from the policymakers. The private actors lack knowledge of the public sector’s expectations of them in joint SCRM/preparedness work. This is perceived as a barrier for collaborative SCRM – which both private and public informants sensed: “How should we dimension the participation of the private actors? What do we expect from them? There is a lack of overall goals that give dimensions and can be boiled down to more concrete orders” (G_publ); “That said, we have to know who pays, when we pay, who decides/steers/controls, who doesn’t decide/steer/control, what is the expectation, what we should be able to deliver” (A_priv). While one joint performance measure mentioned is being able to hold 3 months of food supply in a crisis, many informants point to the lack of definitions as to what this is and how it can be measured. Thus, the importance of joint performance measurements in the sector can be seen, e.g. in Swedish Food Agency (2021), a report proposing performance indicators for the sector.

While joint goals and strategies are needed to proceed towards a more resilient food system, other key discontents are repeatedly mentioned by the private sector in interviews (quote above) and in their investigations (e.g. 4C Strategies, 2020). There is a need for a clear decision-making structure, as well as for clear and explicit economic incentives and funding principles for work related to resilient food systems (e.g. for increasing capacities or safety stocks) – who should pay and how? One logistics-oriented issue relates to stockpiling for preparedness. While Sweden used stockpiling both for military and public purposes during the Cold War, this system was subsequently dismantled. Part of ongoing inquiries (for different critical infrastructures) concerns issues regarding which tier of an SC should hold stockpiles (e.g. as raw material and other inputs, as intermediary products or close to consumers) and who should own and bear the financial investment in stockpiles and how should they be managed operationally and replenished to as not to become obsolete. Concerns are being raised that economic funding and incentives can become a potential area of conflict regarding the participation of the private sector: “I think the State will come back with a set of demands on the companies, and then there will probably be a bit of a squabble about who will pay for this. That’s the question that will probably arise” (E_priv). Similar reasoning can be found in reports written by private actors (4C Strategies, 2020). In line with the previous discussion, the overall takeaway is that clear funding principles are not in place, and private and public actors see clear financial incentives as being necessary for the future.

Some felt that the constitutional mechanism in Sweden acts as a barrier to efficient crisis decision-making. The Swedish constitutional model means that public authorities are not directly controlled by the politically elected ministers but are more independent expert authorities. Furthermore, Sweden has the principle that no authority should interfere with another authority. According to the website of the Swedish Civil Contingencies Agency (MSB, C_publ), the agency “is responsible for issues concerning civil protection, public safety, emergency management and civil defense as long as no other authority has responsibility”, indicating a “gap-filling” role. These constitutional mechanisms cause the informants to claim that confusion regarding roles and responsibilities leads to decision-making difficulties: “I think this is a matter of clarifying the responsibilities between authorities, regions, and municipalities. This is probably confusing for everyone – those interfaces are the messiest” (F_priv). Informants from the expert authorities, and the secretary of the government inquiry (B_publ), also pointed to discontent between politically elected decision-makers and expert authorities.

To plug gaps between expert authorities, in 2022, the government assigned ten “sector responsible” authorities that will take a more significant role in coordinating resilience efforts. The Swedish Food Agency was assigned responsibility for the food system (including primary production that falls under the Swedish Board of Agriculture), which informants found positive for SCRM collaboration. Despite this, the decision-making issue remains, as the “sector responsible” role does not grant a mandate over other authorities, regions or municipalities. The complexity created is expressed as follows: “What we lack is that we have ten authorities that have sector responsibilities [for different critical infrastructures], but there is no-one coordinating them under the government and that is a flaw. That’s why I keep returning to the government’s goals and strategy” (A_priv).

To summarize, different governance mechanisms for collaborative SCRM are connected and interdependent – for example, as stated by one representative: “Regulations and steering mechanisms are essential. But also, as I said, in combination with some form of dimensioning and goal setting so that you steer towards the right things” (G_publ). These connections and interdependencies must be understood if we are to understand how to govern collaborative SCRM, i.e. a system perspective is needed.

5. Findings: empirically elaborated framework for supply chain risk governance

Building on the empirical observations and our analysis, we will now elaborate on, substantiate and reshape Ahlqvist et al.'s (2020) conceptual SCRG framework, thereby extending its theoretical and practical relevance. Data from interviews and archival sources (e.g. citations, with text examples summarised in Appendix 2) was structured into first-order concepts of discontents and barriers as well as antidotes and interventions (Figure 2), that led to second-order themes with a focus on observed discontents related to SCRG and SCRM. The analysis also showed that different informants (Appendix 3) shared the view of the different themes and status of the SCRG mechanisms. Findings were finally aggregated to dimensions describing how different SCRG mechanisms can support or act as barriers towards collaborations and more resilient SCRM. Below this is discussed and propositions are developed.

The purpose of the elaborated framework (Figure 3) is to support a holistic, systematic and cross-disciplinary understanding of how collaborative SCRM can be improved and thus support preparedness and risk handling of disruptions in VSF&CI SCs to make them more resilient. Although complex in itself, the framework intends to systematically illustrate the complexities and interrelations that characterize this SC’s complex web of actors and stress the importance of inter-organizational governance. Describing how multiple actors (policymakers, public direct and private actors) are involved in risk management can enable inter-organizational SC risk evaluation and collective decisions to handle complex, uncertain and ambiguous risks.

Informants were explicitly asked about the content and structure of the guiding SCRG framework (Ahlqvist et al., 2020). All found it interesting and relevant, isolating key questions (as SCRM actions from its governance) and addressing important discontents and mechanisms, e.g. “Many barriers relate, as you stress, to governance issues. Who is resourcefulness for what? What are the legal mechanisms like? … What is the public commitment and what is in their toolbox?” (B_publ).

The two bottom layers in Figure 1 have not been changed, as the focus of the study was on the top-level governance mechanisms. Furthermore, we did not observe any new actor types or obtain any new insights regarding the SCRM processes. However, what must be highlighted is the importance of mapping and understanding the SC structure and flow (the lowest layer), separating private and public actors but also including indirect actors that may influence and intervene in the critical flows. For the second layer, one interesting example of an SC risk mitigation strategy currently discussed among both policymakers and private actors is where and how stockpiles for preparedness should be positioned, financed and managed. To be effective from a nation’s point of view, each direct actor cannot (or will not) take these decisions without any coordination.

We suggest additional elements to include, based on interviews combined with the analysis of what was both emphasized as being important in the private actors’ investigation reports and focused on in recent government inquiries. One such element missing in Ahlqvist et al. (2020) was “inquiries and investigations” which, in this Swedish context, was demonstrated to be an important mechanism for high-level information sharing used increasingly by both private and public sectors to improve joint contextual understanding and create information exchange platforms. As this is quite a tangible formal governance mechanism which is increasingly used, we distinguish it from “joint understood context and concern”, which more represents its outcome (also influenced by more relational mechanisms). Additionally, we separate legal mechanisms and economic incentives. Both were heavily discussed and were often seen as opposing formal mechanisms to create action. Private and public actors had different views on how they should be balanced.

In comparison to Ahlqvist et al.'s (2020) framework, we further separate indirect public actors involved in SCRG into political decision-makers (such as the government) and expert authorities, to illustrate the different policymakers. Even more complex interrelatedness was observed by certain elements. Using a systems approach, we illustrate this with interdependent arrows. All in all, an extended SCRG framework should be designed with the following proposition in mind:

P1.

The more complex governance interrelations in public–private SCRG, the higher the need there is for governance mechanisms designed to increase interaction and coordination and accommodate potentially conflicting SCRM actions.

The SCRG framework’s focus is on the highest layer, which has been elaborated by this study. The observations showed the importance of having a collaborative breeding ground, including antecedents and interrelated mechanisms, such as creating a joint and understood context and having high-level information sharing realized by public–private meeting points and inquiries and investigations. Thus, as in Ahlqvist et al.'s (2020) SCRG framework, the elaborated framework points out the importance of antecedents to facilitate interaction, such as the level of trust between actors, legitimacy and respect and whether formal and informal networks are in place between actors (Boholm et al., 2012; Dubreuil, 2001). As Ghosh and Fedorowics (2008) and Bonatto et al. (2022) point out, trust is an important relational SC governance mechanism which is also needed between public and private actors. Observations show that many of these antecedents have been lacking – but also that they have recently increased, driven by outside triggers (threats) and that actions taken have increased joint trust. Another important relational mechanism is a joint understanding of context and concerns (Klinke and Renn, 2012; Renn, 2008), especially as misunderstandings seem to grow more easily between private and public sectors than within them. Examples of this mechanism in this context are understanding what type of decisions different actors can take and what drives them. We submit the following proposition:

P2.

Antecedents (e.g. trust, respect) facilitating interaction and a joint understood context between private and public SC actors, established before crisis events, will increase proactive SCRM.

It is critical for private actors to understand how the public sector functions and is governed, which differs between countries. Informants observed and discussed constitutional mechanisms, e.g. the power balance between policymakers such as ministers and expert authorities. In the context studied, the constitutional mechanisms caused confusion regarding responsibilities and mandates in the decision-making process, as pointed out by both private actors and public municipalities and regions. Compared to Ahlqvist et al. (2020), the extended framework more explicitly points out different types of governance actors and distinguishes between elected politicians (such as the high-level decision-making government and Parliament) and expert authorities. However, both types of actors indirectly influence the “physical” SC. The framework highlights the importance of understanding country-specific governance factors, especially for multinational private actors. Information sharing is a critical mechanism to increase joint understanding and antecedents such as trust. In the elaborated framework, we propose dividing this into high-level vs detailed information sharing. Based on our analysis, high-level information sharing has recently increased considerably (indicating its importance). One formal mechanism observed here was inquiries (commissioned by the government and performed by authorities), but also investigations carried out, e.g. by industry associations “for lobbying”. A relational side-effect of inquiries and investigations was that different actors were interviewed and later involved in consultations, which created meeting points between private and public actors and further information sharing. However, other meeting points were also increasingly established or intensified recently, emphasizing their relational importance. This confirms the risk governance literature that suggests the importance of having mechanisms for joint pre-estimation of the context (Klinke and Renn, 2012; Renn, 2008; Renn et al., 2011). The following proposition is submitted:

P3.

Platforms for high-level information sharing (including forums, inquiries and investigations) can strengthen antecedents, create a jointly understood context and align strategic focus.

Both private actors and public authorities found a lack of strategic congruence and requested that the political decision-makers formally decide on clear strategic national focuses and goals that the other actors can “align with.” A related discontent discussed was the lack of clear performance measures, which actors are asking for. The Swedish Government uses inquiries, as formal mechanisms, to define strategic focus, which explains why this relation is illustrated as being reciprocal in the framework (Figure 3). As in Ahlqvist et al. (2020), an aligned strategic focus with clear goals is a key element. In the expanded framework, it is thus positioned in the centre, emphasizing its significant influence on other coordinating mechanisms, but also clearly indicating how antecedents (e.g. trust), high-level information sharing and joint understanding improve the opportunity to obtain an aligned strategy and aligned goals.

Aligned strategic focuses increase in importance given the large number of independent actors (private companies, authorities, municipalities, regions, politicians) with their own objectives and strategies involved in VSF&CIs. Clear directives are sought, especially when actors find the context becoming more turbulent. However, according to private actors, directives must be balanced, as too detailed national directives might decrease companies’ competitiveness in a global export market. Furthermore, informants did not perceive that any explicit metrics or performance measures were used, a formal mechanism that according to, e.g. Friday et al. (2018) and Simatupang and Sridharan (2005) should follow from strategic intentions to monitor and drive aligned behaviour. The following proposition is submitted:

P4.

The more turbulent the context, the more important it is to have explicit and aligned strategic focuses and goals for direct actors distant from central authorities.

To increase coordination and integration of SCRM work (Level 2 in the framework), distinct coordination mechanisms are needed but these are hard to develop without first having strategic congruence. Such SCRG mechanisms related to information sharing, collaborative decision-making and incentives are crucial – but need clear and congruent strategic goals to be aligned with.

More detailed information sharing is required for coordinated decision-making, especially related to the SCRM processes. Private actors discussed how more detailed information sharing of operational data, e.g. sharing detailed information on inventory levels or capacities, is a necessity for concrete and coordinated joint SCRM actions. However, this often interferes with legal acts, e.g. related to competition or security; thus, this interrelation must be understood. The importance of collaborative operational information sharing for joint SCRM processes means that public actors, such as competition authorities and politicians, should support private actors in detailed information sharing through guidance and legal mechanisms. The following proposition is submitted:

P5.

Collaborative information sharing and communication must be allowed and in place, to inform operational SCRM process work.

In general, legal mechanisms are important formal mechanisms in outlining what is allowed, who is responsible (influencing decision-making) and when changes occur due to, e.g. definitions of crises or war. Both private and public actors emphasize that clearer legal requirements can create a better understanding of what is expected of private and public actors in preparedness work. However, as mentioned, legal mechanisms sometimes create barriers to efficient risk mitigation actions if no exceptions can be made when necessary. In addition, global private actors highlight that too strict legal requirements can act as a barrier to their participation in national preparedness work. While clear economic incentives are other formal mechanisms sought by the private actors to drive their operative decisions (e.g. on risk mitigation strategies related to stockpiling) rather than too detailed regulations in laws, informants also express the need for a balance between legal and economic incentives. The following propositions are submitted:

P6a.

Legal mechanisms and economic incentives must be explicit, balanced and aligned when it is assumed that private actors will manage societal risks.

P6b.

Institutional mechanisms coordinating SCRM actions (e.g. preparedness stockpiling) must respect both public and private logic and thus balance explicit (legal) regulations with economic incentives.

Private and public actors describe how SC structures in VSF&CIs are characterized by a diverse range of involved actors with a lack of synchronized or collaborative decision-making. Private actors represent global and local supply chains as well as various ownership structures, while public actors include, for example, municipalities, regions and authorities. Both private and public actors emphasize how a lack of clear decision-making mandates combined with the diversity of actors involved act as a barrier towards increased collaboration. The explicit lack of (and perceived need for) clear decision-making (and incentive structures) related to overall strategic goals, gaps between public actors and private SC structure, especially when insecurity in the context increases, motivates a focus on decision-making mechanisms and the following proposition:

P7.

The more insecure the context and the more diverse SC actors are, the more important it is that decision-making processes and incentive structures for collaborative SCRM actions are explicit and transparent.

Finally, this study of the Swedish food system showed that systematic learning between private and public sectors remains limited and has not been focused, although individual positive examples exist. Certain activities such as joint courses and some public-private scenario training have been performed, although in few cases as yet and not in a structured way. As collaborative learning (from mistakes and successes) is a critical mechanism for developing SCRM (Cao et al., 2010; Kache and Seuring, 2014; Norrman and Wieland, 2020) and risk governance (Dubreuil, 2001; van Asselt and Renn, 2011) and informants felt it was lacking, this mechanism has been retained in the SCRG framework.

6. Concluding discussion, limitations and future research agenda

6.1 Contributions

Taking the importance of global SC for society and its vital functions and the challenges of collaborative SCRM as its starting points, this study explored top-level governance mechanisms between public and private actors. The aim was to contextualize the phenomenon of SCRG in the VSF&CIs context and to extend the theoretical and practical relevance of Ahlqvist et al.'s (2020) conceptual SCRG framework by empirically exploring, theoretically elaborating on and further conceptualizing SCRG.

This study makes several contributions. Firstly, the study is one of the first to study the topic of governing and managing risks in VSF&CI supply chains empirically and to conceptualize the importance of public–private interactions in more detail. It confirms the importance of understanding how to manage SC risks in VSF&CIs and explores issues of mixing private and public actors. It thus answers calls for a better understanding of joint inter-organizational governance and collaboration between public and private direct and indirect actors in SCRM (Ahlqvistl et al., 2020; Friday et al., 2021; Tsolakis et al., 2023) and for research related to the roles and interventions of policymakers in SCs (Pournader et al., 2020; Darby et al., 2021; Scala and Lindsay, 2021; Sodhi and Tang, 2021).

Secondly, it empirically extends the recent literature on SCRG (Ahlqvist et al., 2020) by describing issues found in practice related to the current increased turbulence in food systems and related preparedness. It contributes more detailed empirical observations and insights regarding discontents and how practice is developing related to SCRG. In terms of general theoretical relevance, it further develops and elaborates on Ahlqvist et al.'s (2020) conceptual SCRG framework by proposing new elements. The initial SCRG framework is substantiated by adding more public actors and drawing a distinction between political decision-makers in the form of government and expert authorities. The element of information sharing is divided into two (on different levels) and the importance of inquiries and investigations in the public-private interaction of VSF&CIs SCs is stressed. Inter-organizational institutional mechanisms are separated into legal mechanisms and economic incentives, to consider their differences and how they interact positively as well as negatively.

Thirdly, empirical-driven insights into SCRG further develop the general conversation on governance mechanisms in the SCM domain (e.g. Ghosh and Fedorowics, 2008; Tachizawa and Wong, 2015; Bonatto et al., 2022; Gemente et al., 2024) and particularly related to SCRM.

The framework is practically useful as it can guide policymakers', public actors' and private actors' managers in how to map, understand and address important issues related to improving resilience and joint handling of risks in CI supply chains. The study contributes to the knowledge of policymakers and SC practitioners by describing real-word examples of discontents related to SCRG and SCRM and how different SCRG mechanisms can support or act as barriers towards collaborations and more resilient food systems. It can support a holistic, systematic and cross-disciplinary understanding of how collaborative SCRM can be improved and thus support preparedness and risk handling of future disruptions, making VSF&CI supply chains more resilient. Specifically, it can support policymakers’ understanding of the importance and complexity of SC flows and nodes, e.g. related to preparedness stocks and capacities. It provides support in understanding which governance mechanisms private actors find important. If policymakers do not understand risk sharing and the economic incentives necessary for private actors to hold the inventory stocks required for VSF&CIs, the risk is high that no stock will be kept when critically needed.

6.2 Limitations

Despite the measures taken to increase transferability (Table 2), limitations exist. As the study only addresses the food system in one country, Sweden, future research could increase transferability by investigating the usefulness of the SCRG framework in systematically structuring and describing SCRM discontents and SCRG mechanisms in other contexts. Additional studies should investigate whether, e.g. inquires and investigations are typical Swedish mechanisms, or whether other countries also use these to increase the jointly understood context and aligned strategic focus. If major differences are identified between countries, this invites research on contextual factors, e.g. related to history (such as recent experiences of war and crisis), constitutional mechanisms (e.g. the power balance between ministers and expert authorities) and the influence of national cultures (such as uncertainty avoidance and individualism-collectivism (Hofstede, 1980)).

Another limitation is the focus on top-level governance, with selected informants being more high-level rather than involved in operational details. As food systems are very multi-faceted and wide-ranging, especially in food production, one limitation is the low number of respondents. Despite including industry federations to represent a balanced view from industry and retail and purposefully selecting global firms involved in different food types with Swedish production, future studies could focus on potential differences between different parts of the food industry, e.g. reflected by specific laws, regulations and behaviours. The specific industry setting may differ, and restaurants are one part of the food system which was not studied.

6.3 Future research

This explorative study opens up many new research avenues in addition to those motivated by research limitations. If additional empirical evidence is also collected for other VSF&CIs in other countries, the framework’s transferability and operationalization can be further developed.

One research issue relates to the potentially necessary adoption of governance mechanisms in different situations on the spectrum from normal business via crisis, and threat of war, to war. Related to proposition 1 and complex governance interrelation, informants clearly pointed out the interdependence between different VSF&CIs, both in the SC layer (structure and flows) and in the top-level governance layer. For the food system, supplies of energy, diesel, water and transport were specifically pointed out as critical supply – but cyber-attacks on IT or financial flows can also interrupt the SC. Future research should thus develop the SCRG framework to also include critical interdependences between different VSF&CIs.

The balance between formal and relational governance mechanisms is interesting. We observe top-level policymakers and inquirers focusing more on improving the formal mechanisms, while private actors explicitly ask for these to be balanced with relational ones. Building on proposition 2 and linking to extant SC governance research (Ghosh and Fedorowics, 2008; Bonatto et al., 2022), future research could study the importance of trust as an antecedent or governance mechanism between policymakers and companies in a public-private SC. Related to propositions 3 and 5, more studies should investigate what kind of information-sharing mechanisms work best in public-private VSF&CIs, given issues with information security and competition law.

Decision-making, priority-setting and their dependence on strategy and goals direct more research (related to proposition 4) into what explicit goals can be used to define the needed preparedness level to enable it to be understood among different sectors and SC tiers.

The SCRG framework's conceptualization and systems approaches can both be further developed by future research. The interrelatedness between different top-level governance mechanisms and specific risk mitigation strategies (Level 2) could be explored, e.g. collaborative decision-making related to preparedness stockpiles and surplus production capacities. (This could answer what buffers should be invested in, where in the SC buffers should be positioned, who should own and invest, who should handle inventory management and replenishment, whether the control should be via legal mechanisms or economic incentives, etc.). Building on propositions 6 and 7, the interrelatedness between legal mechanisms and economic incentives should be further investigated, e.g. the balance sought between legal and economic mechanisms in different situations – and what contextual; factors would influence this.

Finally, although we claim that SCRG is more intriguing in VSF&CIs that depend on both private and public actors, the issue of governing collaborative SCRM also exists in an SC that only consists of private actors (e.g. Norrman and Jansson, 2004; Fan and Stevenson, 2018; Friday et al., 2018; Friday et al., 2021). The usefulness of the suggested SCRG framework to analyse collaboration SCRM issues in private SCs must be examined further.

Figures

Figure 1

Conceptual multi-level framework for SC risk governance

Figure 2

Data structure for coding and aggregation process

Figure 3

Extended conceptual multi-level framework for SC risk governance

Table 1

Informants

Informant	Title/position	Organization	Type of actor	Private or public	Date and time (h)
A_priv	Manager Risk and Security (grocery retailer)/Project manager, preparedness (industry organization, grocery retail)	Grocery retailer 1/Svensk Dagligvaruhandel – the Swedish Food Retailers Federation	Industry organization grocery retail/indirect Private company/direct	Private	Jan 2023 (2)
B_publ	Chief Secretary	Commission/committee of Swedish government inquiry (N 2022: 02 A new food preparedness)	Appointed by the government/indirect	Public	Jan 2023 (1,5)
C_publ	Researcher	MSB – Swedish Civil Contingencies Agency	Government agency/indirect	Public	Jan 2023 (2)
D_priv	Economic policy expert, focus areas: crisis preparedness, supply capacity and total defense	Livsmedelsföretagen – the Swedish Food Federation	Industry organization food production/indirect	Private	Jan 2023 (1,5)
E_priv	Senior Strategy Advisor	LRF – the Federation of Swedish Farmers	Industry organization farmers and other agricultural production/indirect	Private	Jan 2023 (1,5)
F_priv	Chief Strategic Supply Chain	Grocery retailer 2	Private company/direct	Private	Feb 2023 (2)
G_publ	Part of preparedness unit	Livsmedelsverket – the Swedish Food Agency	Government agency/indirect	Public	Feb 2023 (1,5)
H_publ	Security coordinator	Municipality 1	Municipality/direct	Public	Mar 2023 (2)
I_publ	Preparedness coordinator	Municipality 2	Municipality/direct	Public	Mar 2023 (2)
J_priv	Head of factory	Global manufacturer 1 with Swedish factories	Private company/direct	Private	Mar 2023 (2)
K_priv	Research and Strategic Partnership Manager	Global manufacturer 2 with Swedish factories	Private company/direct	Private	Mar 2023 (2)
L_publ	Risk and security manager	Region 1	Region/direct	Public	Mar 2023 (2)
M_publ	Process manager, food strategy	Region 1	Region/direct	Public	Mar 2023 (2)

Source(s): Authors’ own work

Table 2

Research quality actions

Transferability (compare External validity)	• Stated theoretical aim (theory elaboration), unit of analysis and justification of case study; described logic and criteria applied for case selection to allow analytical generalization • To be able to transfer results to different contexts, the food system shortly described. However, some aspects of the informants are not included to allow for anonymity • Data on different actors' view and context to enable comparison with other contexts • Results presented to knowledgeable researchers to check for transferability • Pilot interview with one informant to test interview questions based on guiding framework • Conducted cross-actor analysis considering different discontents, interventions and mechanisms
Credibility (compare Internal validity)	• Explicated the interrelations between key constructs; used different bodies of literature (SCRM, risk governance) to interpret findings • Case and informants selected following a structured and purposeful sampling approach • Involved multiple informants (representing different actors and perspectives on governance) to compare, contrast and triangulate responses • Used a literature-based guiding framework • Informants respond similarly to the same problems and mechanisms • Investor triangulation: two researchers discussed and judged data and interpretations • The raw data (transcripts) put into software (NVIVO) and then coded on different levels (1st, 2nd, 3rd) to connect it with different theoretical categories • Presented results to peer researchers (from some and other disciplines) familiar with the studied phenomenon to check and obtain feedback about case analysis
Confirmability (compare Construct validity)	• Described the method choices and the research design; part of the paper has been through scientific review processes for conference • Conducted systematic data collection • Details on how the data is collected in the method chapter to assure integrity of results • Multiple perspectives and sources of data collected (interviews, presentations, inquires, reports and websites). Primary and secondary data are triangulated and used to augment data • Involved multiple respondents (representing different internal perspectives and actors) • Interview guide provided • Engaged peer researchers to review research protocol and interview guide • Gave examples of detailed quotes and first level-coding • Engaged researchers from other disciplines familiar with the studied phenomenon to obtain feedback on case analysis and framework development • Guiding framework presented to respondents to compare and contrast views • Terminology and guiding framework explained to informants to avoid misunderstanding • Triangulated data using multiple sources of evidence (interviews, internal presentations, industry reports, websites) to triangulate respondent responses • Illustrated the chain of evidence using matrices, tables and figures. • Conducted first-, second- and third-level coding and aggregation illustrated in coding tree
Dependability (compare Reliability)	• The method choices and the research design described • Explained data analysis procedures • Preliminary results presented at conferences to get input from other researchers • Developed a standardized interview guide for data collection across the informants • Developed and continuously updated a case study database (including, secondary and archival sources, interview transcriptions, themes, codes and memos) to ensure complete documentation of the data analysis procedure • Displayed data in different structures, like coding-tree and tables, to facilitate cross-informants analysis and find pattern • Explained interrelations between key concepts (the framework) and triangulated different literature (SCRM, risk governance)

Source(s): Authors’ own work

Appendix

The supplementary material for this article can be found online.