Undercurrents: resource dependence strategies across the adaptive cycle in a critical raw material-dependent supply network

Geopolitical tensions among nation-states are reshaping the design and strategy of global supply networks (SNs) (Golgeci et al., 2025). Recent examples, such as Brexit, national COVID-19 mandates and the US–China trade war, have unsettled SNs, leading to the redesign of manufacturing locations, reconfiguration and increased complexity of SNs (Charpin and Cousineau, 2025; Moradlou et al., 2025). As a result, research on geopolitical disruptions is expanding rapidly. Yet, geopolitical tensions persist and escalate, necessitating further empirical research (Golgeci et al., 2025). Critical raw material (CRM) SNs are at the centre of geopolitical tensions (OECD, 2022), but this context has received limited empirical investigation (Bhamra et al., 2025).

CRMs are metallic and non-metallic elements and alloys with unique physical and chemical properties, limited substitutability, and are endemic to specific geographical regions (Shiquan and Deyi, 2023). They are important in strategic industries, including those advancing the green transition and modern technologies (Bhamra et al., 2025). Consequently, producing countries have adopted protectionist measures, while dependent countries increasingly frame reliance on CRM as a national security concern and intervene to strengthen autonomy and safeguard critical industries (Shiquan and Deyi, 2023). For example, China's export controls on rare earth elements in response to the US trade war (Lv and Munroe, 2024) and Russia's curtailment of its supply of CRMs, such as helium and nickel, to the EU bloc following sanctions (OECD, 2022) illustrate these dynamics. Concurrently, the EU has developed the CRM Act, which seeks to reduce the geopolitical risks associated with dependency on specific locations and strategically scale localised production within the EU (European Commission, 2024). Similarly, the USA has enacted Section 232, which imposes tariffs on the imports of CRMs to promote local value addition (Bureau of Industry and Security, n.d.).

The uncertainties surrounding geopolitical tensions in CRM supply require firms to continuously adapt their dependency strategies to respond to the evolving global environment (Bednarski et al., 2023). According to resource dependence theory (RDT), firms respond to dependency uncertainty by restructuring dependencies (Pfeffer and Salancik, 2003). There is, however, scant evidence on how geopolitical disruptions and tensions influence responses to manage dependencies. Prior literature applying the RDT mainly assumes a homogeneous environment (Jiang et al., 2023). Yet geopolitical disruptions and tensions require firms to develop adaptive strategies to navigate a continuously changing rather than a static environment (Zheng et al., 2025).

Unlike prior research, we elaborate on the RDT by examining the adaptive responses of CRM-dependent firms within SN to ongoing geopolitical tensions in a continuously changing environment. By examining the dynamic evolution of adaptive strategies, we align with recent calls for process-based studies that assert SNs as adaptive systems that constantly respond to external influences (Grimm et al., 2024). We integrate the adaptive cycles of panarchy theory (Holling, 2001), which conceptualise adaptation as a cyclical process of growth, conservation, release and reorganisation in response to environmental change, with the RDT to explore the research question: How do firms in SN, dependent on CRMs, respond adaptively to geopolitical disruptions and tensions?

We conducted a case study in the context of Nordic manufacturing SN dependent on CRMs and exposed to contemporary geopolitical disruptions and tensions. Using purposive sampling, we conducted 19 semi-structured interviews with 3 CRM experts and 16 senior executives from the Nordic manufacturing firms embedded in CRM-dependent SN. We extended our analysis by collecting longitudinal SN data for the firms represented by our interviewees to capture system-level dynamics associated with adaptive cycles. Network data were obtained from Bloomberg SPLC for the period 2013–2025. Analysing these data allowed us to complement the interview insights and strengthen the temporal and system-level understanding of SN adaptation across the adaptive cycle phases.

We found that during the conservation phase, firms prioritise stability and continuity in dependency management, as firms tend to normalise geopolitical risk and defer costly reconfiguration. At the release phase, firms primarily engage in negotiating with the environment and avoiding dependencies to restore short-term flexibility amid high uncertainty. During reorganisation, firms emphasise absorbing the environment and avoiding dependencies as they redesign dependency structures to regain control over critical CRMs. Finally, in the exploitation phase, firms shift towards strategic renewal and growth, increasingly engaging in creating the environment and negotiating with the environment to shape favourable institutional, geographic and relational conditions.

This study makes the following theoretical contributions. Firstly, we extend the RDT view of managing uncertainties in the external environment. Whereas prior studies have focused predominantly on bridging and buffering as strategic responses (Bode et al., 2011; Kalaitzi et al., 2018), we identify that firms develop adaptive responses of absorbing the environment, avoiding dependencies, creating the environment and negotiating with the environment amid geopolitical disruptions and tensions. Also, we respond to calls to integrate RDT with complementary theoretical lenses (Hillman et al., 2009; Jiang et al., 2023) by combining RDT with the adaptive cycles of panarchy theory to explain how firms and their SN structures evolve in response to geopolitical shocks. We show that dependency restructuring unfolds differently across adaptive cycle phases: at the release phase, firms engage in partial or full release by either avoiding dependencies or negotiating with the environment, depending on the severity of geopolitical disruptions. Finally, contrary to the assumption that firms adapt to geopolitical shocks, we identified a limitation to this viewpoint. We show that adaptation is selective, temporally uneven, and may be strategically avoided under conditions of extreme dependency and structural constraints.

The adaptive cycle of the panarchy theory is useful for studying persistence, adaptation and transformation over time (Wieland, 2021). The adaptive cycle is conceptualised as a four-phase process that moves through exploitation, conservation, release and reorganisation (Holling, 2001), as illustrated in Figure 1, and these phases are further grouped into the front and back loops. In this study, we consider firms within an SN as adaptive systems and demonstrate how they progress through the four phases of the adaptive cycle.

The front loop of the adaptive cycle commences with the exploitation phase, characterised by firms forming new relational and structural capital to pursue growth and value creation (Mirzabeiki and Aitken, 2023; Vega et al., 2023). Relational and structural capital among firms and SN actors begins to accumulate and become increasingly interconnected, thereby transitioning into the conservation phase. In the conservation phase, network structures and relational rents such as practices, relational-specific assets, and routines are reinforced and stabilised (Wieland, 2021). However, the increasing connectedness among firms may create rigidities whereby repeated investments in established relationships and network structures develop lock-in effects that constrain alternative strategic actions (Sydow et al., 2009) and limit adaptive capacity (Holling, 2001). These rigidities inhibit alternative actions due to the constraints of higher switching and sunk costs (Sydow et al., 2009). Organisational inertia emerges in conditions of high rigidities, whereby firms are disincentivised to adapt to disruptions (Barker et al., 2026). Yet, in the conservation phase, the system becomes overconnected, and the accumulated capital is predisposed to increased vulnerabilities, loss of resilience and becoming “an accident waiting to happen” (Holling, 2001, p. 394).

When disturbances originating from cross-level linkages exceed the adaptive capacity of accumulated rigidities in relational and structural capital, they pave the way for creative destruction that dissolves rigidities (Holling, 2001; Vega et al., 2023). The back loop of the adaptive cycle begins with the release phase, where flexibility enables the release of accumulated capital, creating a transient space for innovation and adaptation during the reorganisation phase (Wieland, 2021). The transition from release to reorganisation is marked by high uncertainty and requires firms to respond adaptively to unexpected disruptions (Holling, 2001). The reorganisation phase indicates high flexibility, experimentation and innovation, enabling the accumulation of new relational and structural capital and laying the foundation for the next adaptive cycle (Holling, 2001). The adaptive response observed during the reorganisation and exploitation phase can be understood as sensing environmental changes, seizing new opportunities, and reconfiguring internal and external capabilities and network structures in the face of uncertainty (Teece, 2007). Overall, the adaptive cycles emphasise two contrasting perspectives: the front loop highlights growth and accumulation, whereas the back loop depicts flexibility, innovation and reassortment (Holling, 2001).

Recent studies in operations and supply chain management (OSCM) have increasingly drawn on panarchy theory and the adaptive cycles to examine how firms and their SN actors interact with broader socio-economic, political-economic and planetary levels. Studies have shown that SNs' cross-level linkages with other systems reinforce one another (e.g. Madonna et al., 2024; Mirzabeiki and Aitken, 2023; Vega et al., 2023). More recently, the higher-level political system has brought seismic shifts to global SNs (Free et al., 2025). Our study examines how firms within an SN respond to geopolitical tensions that disrupt their CRM dependencies through the lens of panarchy theory's adaptive cycles. Rather than examining the cross-level linkage between firms in an SN and their mutually reinforcing linkages with geopolitical forces, we investigate how geopolitical tensions trigger adaptive responses among firms dependent on CRMs. This delimitation is appropriate, as the objective is not to investigate how firm-level responses shape the higher-level geopolitical system but rather to provide a focused understanding of how firms progress through the phases of the adaptive cycle.

The contemporary era of deglobalisation has seen a rise in protectionism, nationalism, international wars and economic sanctions, potentially leading to failures and structural changes in global SNs (Golgeci et al., 2025). Several OSCM studies have focused on protectionist measures such as the US–China tariff imposition and resulting trade wars. These studies show that the US–China trade war reduces spatial complexity (Chakkol et al., 2024), increases inventories and decreases profitability among firms impacted by the trade wars (Fan et al., 2022). Moreover, the research describes firms' adaptive responses (Zheng et al., 2025) and the strategies they use to lock in their buyers (Fan et al., 2024) during heightened trade wars.

Wars such as the Russia–Ukraine war have disrupted the flow of raw materials, contributing to global price hikes in commodities such as grains and metals (Bednarski et al., 2023). Studies on the Russia–Ukraine war examined how downstream firms adopt digital capabilities to improve agri-food supply chain resilience strategies (Belhadi et al., 2024), and unhook their relationships and reconfigure their SNs from conflict zones (Srai et al., 2023). Despite growing research, several global dynamics, such as geopolitical tensions and economic policy interventions, persist incessantly and continue to adversely affect SNs, calling for further investigation (Golgeci et al., 2025).

Extant studies examine geopolitical disruptions across sectors such as healthcare, agrifood and general manufacturing; CRM-intensive sectors remain underexplored (Bhamra et al., 2025). Bednarski et al. (2023) stated that research on geopolitical disruptions does not address CRM SNs. CRMs are the backbone for crucial sectors, including those supporting the green transition and the production of advanced technologies such as semiconductors, electronics, magnets and defence systems (Bhamra et al., 2025). Overall, CRM SNs occupy a central position in geopolitical tensions, as they are significant for nations' global economic and political leverage (Bednarski et al., 2023). Unsurprisingly, the race for these materials has resulted in geopolitical turbulence that exists in two forms. Firstly, nations endowed with extraction, refining and production capabilities have advanced protectionist measures to exert control over these resources (Bhamra et al., 2025). Secondly, nations that lack reserves or production capacity view their dependencies as a national security concern and have enacted industrial policies to gain autonomy, increase domestic value-addition and reduce supply risks (Shiquan and Deyi, 2023). These macro-level shifts create a turbulent operating environment for manufacturing firms in CRM dependent SNs. These higher-level shifts in the political system require firms to adapt their strategies to manage dependencies amid uncertainties arising from geopolitical tensions.

The RDT posits that organisations are not autonomous but depend on the external environment for survival (Pfeffer and Salancik, 2003). This creates interdependencies and uncertainties, motivating firms to act to reduce constraints (Hillman et al., 2009). Interdependencies refer to the mutual reliance in interfirm relationships, which creates mutual dependencies and power imbalances (Casciaro and Piskorski, 2005), while uncertainties are the unpredictability and complexity of the external environment over which firms lack control and power in securing access to critical resources (Bode et al., 2011).

Governments are crucial actors in firms' environment, as governments develop rules that may restrict the flow, production, distribution and consumption of critical resources (Pfeffer and Salancik, 2003). Geopolitical forces pose a significant source of uncertainty in global SNs, requiring firms to manage dependencies on external resources controlled by others (Zhu et al., 2025). Thus, we investigate how geopolitical forces create environmental uncertainties and how firms adapt their strategies to manage dependencies on the external environment.

Generally, RDT proposes that firms manage dependencies on an uncertain environment by absorbing, avoiding, creating and negotiating with it (Pfeffer and Salancik, 2003). Absorbing the environment entails internalising the source of dependency fully through vertical integration and partially through interorganisational relationships (Hillman et al., 2009); avoiding dependency is bypassing the strategic needs or withdrawing from the current environment (Jiang et al., 2023); creating the environment is adapting the environmental conditions mainly through legal and political means (Hillman et al., 2009); negotiating with the environment involves managing organisational interdependencies through cooperation, communication and coordination (Pfeffer and Salancik, 2003).

This study employs a sequence-structuring theory elaboration approach (Fisher and Aguinis, 2017) to understand how firms in an SN dependent on CRMs develop adaptive responses to geopolitical tensions. The sequence-structuring theory elaboration approach entails explaining the sequence of events that leads to an outcome as a basis for understanding and explanation building (Fisher and Aguinis, 2017). Guided by abduction logic, we move iteratively between empirical observation and theoretical concepts of the RDT and the adaptive cycles of panarchy theory, allowing new insights to emerge from the data (Ketokivi and Choi, 2014).

The study is situated within the Nordic countries (Denmark, Finland, Iceland, Norway, Sweden), which we treat as a unified empirical context given their similar industrial policies and institutional structures. The Nordic countries are among the global frontrunners for advancing the green transition and innovation (European Commission, 2025; Wyman Oliver, 2026), driving strong demand for CRMs used in clean energy, digital technologies and advanced manufacturing. However, they remain highly dependent on third countries for CRM supplies (Nordic Innovation, 2021), creating supply risks amid protectionist measures, including export restrictions, imposed by major CRM-producing countries (OECD, 2025). These vulnerabilities have been amplified by recent shocks, particularly Russia's invasion of Ukraine, subsequent Western sanctions, and Finland's and Sweden's accession to NATO. For instance, before the war and the imposition of sanctions, 84% of Finland's nickel imports originated from Russia (OECD, 2022). Furthermore, as the USA is a major export destination for several Nordic countries (Export Discoveries, 2025), their export-oriented firms are increasingly exposed to the US–China geopolitical rivalry and recent US tariff measures (Bednarski et al., 2026). Growing tensions between Denmark and the USA over Greenland, which holds significant CRM reserves, have further heightened concerns over access to strategically important natural resources. Together, these factors make the Nordic countries a relevant empirical context for understanding how firms in CRM-dependent SNs develop adaptive responses to geopolitical tensions.

To develop a contextual understanding of the phenomenon, we first conducted an on-site focus group with three domain experts (Interviewees N1–N3, Supporting Information A1) from the National Geological Institute in Finland. The experts possessed extensive knowledge of geopolitical risks, regulatory frameworks and Nordic CRM-dependent SNs. The discussion focused on system-level geopolitical tensions affecting CRM supply, including major disruptive events and their implications for Nordic manufacturing firms. Insights from this discussion informed the second-stage sampling and data collection, which involved senior executives from firms operating in CRM-dependent SNs.

Next, we adopted a multistep sampling strategy, consistent with prior studies examining SNs as a case (e.g. Garcia et al., 2024). Firstly, we identified firms headquartered in the Nordic countries that operate in CRM-intensive sectors using the Orbis database. The search string was “Country: Denmark, Finland, Iceland, Norway, Sweden, AND Employees ≥ 500, AND NACE code 24–29”. Firms with at least 500 employees were selected to ensure inclusion of large firms with extensive operations (Roscoe et al., 2020). NACE codes 24–29 captured firms operating in CRM-intensive sectors, including basic metals, computer, electronic, and electrical equipment, and machinery (European Commission, 2010). This produced an initial sample of 489 firms.

Secondly, we conducted systematic desk research to assess each firm's exposure to geopolitical tensions affecting CRM SN between 2017 and 2025. We selected 2017 as the starting point because prior research identifies this period as marked by increased global geopolitical turbulence (Contractor et al., 2026). We reviewed multiple secondary data sources, including company websites, annual reports, blogs, press releases, executive statements, white papers and news articles. Firms were retained when at least one clear and verifiable reference to geopolitical tensions affecting CRM supply was identified. On the other hand, firms that referred only to general supply chain uncertainty, trade tensions or shifts in global trade patterns without explicitly stating the effects on CRM SN were excluded. Of the initial 489 firms, 392 were excluded for insufficient evidence, leaving 97 firms.

Thirdly, we contacted senior executives responsible for procurement and OSCM within the remaining firms. Sixteen senior executives from 13 firms agreed to participate in the study. The list of interviewees and the interview protocol are presented in Supporting Information A1 and A2, respectively. All informants had between 10 and 40 years of professional experience, with an average of 20 years. The interviews were conducted by the two authors via Microsoft Teams between May 2025 and December 2025, lasting an average of 40 min. With the informants' consent, the interviews were recorded for transcription and data analysis. Interview recordings were transcribed verbatim for analysis (Miles and Huberman, 1994).

SN data were collected from Bloomberg SPLC to examine whether structural changes were consistent with the temporal patterns and adaptive dynamics described by interviewed informants. For the firms in the final sample, we extracted yearly data on direct and indirect second-tier buyer–supplier relationships from 2013 to 2025, including firm identifiers, relationship size and industry classification. Although prior research identifies 2017 as the onset of increased global geopolitical turbulence (Contractor et al., 2026), we collected data from 2013 to capture SN conditions preceding major disruptions. Of the 13 firms, 9 had sufficient Bloomberg SPLC data, whereas 4 private firms had none available. Following Culot et al. (2023), only manufacturing firms were retained. We observed that nine firms operate within a highly overlapping SN, which we subsequently treated as a single network. After removing duplicates, the yearly sample comprised 15,094 dyads and 2,326 unique firms. Nodes represent manufacturing firms and ties represent buyer–supplier relationships. We restricted the SN boundary to direct and indirect second-tier upstream supplier relationships because the study focuses on CRM sourcing dependencies rather than downstream customer structures.

The unit of analysis is the adaptive response of firms in an SN to geopolitical tensions. Analysis was initially conducted at the firm level (embedded unit of analysis) and subsequently aggregated to the SN (the case study) using both primary (interviews) and secondary (SN reports) data. We adopted an analytical strategy suitable for process-oriented research (Grimm et al., 2024). The analysis iteratively moved between empirical data, emerging categories and theoretical concepts. Using NVivo 15, we conducted the first cycle of open coding grounded in the data. Codes captured the geopolitical tensions firms encountered and their responses before and after their onset. Recurring codes were condensed into second-order categories through pattern matching. For example, several informants described terminating contractual relationships with CRM suppliers following the Russia–Ukraine war, resulting in the second-order category “exiting politically risky partners”. However, firms exposed to the US–China geopolitical tensions maintained contractual relationships with existing suppliers while adjusting pricing arrangements with their customers, resulting in the second-order category “adjusting pricing models”.

The next coding cycle involved abductive iteration between the emerging categories and RDT. Initially, we interpreted the second-order categories through the RDT responses of bridging and buffering (Bode et al., 2011). However, we later observed that several empirically grounded second-order categories, such as “leveraging regulatory loopholes” and “reshaping national CRM capabilities”, were not fully explained by buffering and bridging strategies. This prompted us to revisit the RDT. Through iterative comparison between data and theory, we refined and regrouped the second-order categories into the aggregate dimensions of absorbing dependencies, avoiding dependencies, creating the environment and negotiating with the environment. For example, the second-order category of “leveraging regulatory loopholes” and “reshaping national CRM capabilities” was grouped under the aggregate dimension of “creating the environment”.

We noted that responses followed a nonlinear pattern and used adaptive cycles as an analytical lens to understand how RDT strategies unfolded over time, based on phase characterisation. We employed temporal bracketing to decompose process data into successive, adjacent phases to develop emergent patterns over time or to explain causal mechanisms (Langley, 1999). Guided by the adaptive cycles' literature and the phase characterisation (Supporting Information A3), we iteratively examined how firms' responses unfolded in response to geopolitical tensions affecting the CRM SN. Adaptive-cycle phases were not imposed on the data but emerged through iterative comparison between observed response sequences and theoretical phase characterisations. The resulting coding tree is reported in Supporting Information A4.

To enhance the analytical rigour, we triangulated the informants' accounts with the focus group discussion involving CRM experts, news outlets and industry reports (e.g. OECD, 2022, 2025), and firms' secondary data sources (annual reports, newsletters, blogs and white papers) to assess the consistency (Moradlou et al., 2025). This analytical move aligns with Schwenk's (1985) assertion that organisational decision processes should be analysed using multiple data sources, and that both participant recollections and archival data may minimise biases arising from reliance on a single data source in processual studies. Temporal accounts were largely consistent across data sources. One discrepancy concerned Firm D's CRM alternative initiative, which was planned for 2026 rather than 2024. This discrepancy was subsequently corrected in the analysis, with publicly available firm accounts used to validate the event sequencing.

SN analysis was conducted alongside the primary analysis and served as complementary evidence, consistent with a methodological bricolage approach (Pratt et al., 2022). Initial visual inspection confirmed a single interconnected network. We then complemented visual analysis with a focused set of SN structural measures adapted from Borgatti and Li (2009), Kim et al. (2011), Castell and Schrenk (2020) and Rogers et al. (2025). The SN was plotted annually from 2013 to 2025, and all measures were calculated in R (Supporting Information A5).

Changes in SN density and degree centralisation were used to assess increasing coordination and structural consolidation during the exploitation and conservation phases; dependency concentration captured the emergence of asymmetric dependencies associated with conservation; and tie churn reflected network instability and relational reconfiguration during release and reorganisation. Rather than treating these measures as definitive indicators of adaptive-cycle phases, we used them to assess whether SN evolution was consistent with conditions inferred from the primary data (e.g. OECD, 2022, 2025).

We integrated the interview and SN analyses through iterative comparison across temporal brackets. Interview accounts were first used to identify provisional adaptive-cycle phases and response patterns. We then examined whether changes in SN measures supported, contradicted or refined these interpretations. When discrepancies emerged, we revisited the interview and secondary data to refine temporal interpretations.

The validity and reliability criteria are reported in Supporting Information A6. To mitigate social desirability bias, the interview protocol avoided priming respondents with specific geopolitical events. Specific events were probed only after respondents mentioned them. The two authors independently coded the dataset and, at each coding round, compared interpretations and discussed discrepancies until consensus was reached (Miles and Huberman, 1994). When inconsistencies emerged, we revisited analytical memos and data sources to ensure consistent interpretation.

Although the interviews combined concurrent and retrospective accounts, retrospective data are subject to recall and hindsight biases (Grimm et al., 2024). We therefore adopted several mitigation procedures. Firstly, the informants were encouraged to describe the sequence of events and responses taken before, during and after major geopolitical disruptions affecting their CRMs SNs. Follow-up questions were used to clarify event chronology and response evolution (Ahlin, 2019). Secondly, the interviews were conducted jointly by two authors to minimise missed probes and clarifications (Grimm et al., 2024). Thirdly, in the few cases where multiple informants were interviewed, we compared their perspectives to identify convergent accounts (Golden, 1992).

Finally, during the SN analysis, we documented notable structural changes over time. For instance, we identified the termination of the relationship between focal firm L and Mechel PJSC, a Russian mining and metals company, in 2014 following Russia's invasion and subsequent annexation of Crimea. These changes were linked to interview findings to validate event sequencing. Bloomberg SPLC occasionally reflected structural disruptions with a temporal lag (Culot et al., 2023), which we have taken into consideration in our analysis.

Our analysis indicates that geopolitical tensions affected firms embedded in the Nordic CRM-dependent SN, triggering transitions across adaptive-cycle phases. Figure 2 summarises the major phases identified in the data. While the release phase was relatively distinct, the remaining boundaries were less clear, reflecting the gradual and overlapping nature of adaptive-cycle transitions. Figure 2 should therefore be interpreted as a heuristic rather than a definitive representation of these transitions.

We then examined the RDT strategies adopted by firms within SN across the adaptive cycle phases. We did not differentiate responses by specific geopolitical disruptions; instead, we focused on firms' broader adaptive responses to geopolitical turbulence. This analytical choice reflects both the empirical material and the study's theoretical orientation, as interviewees consistently described geopolitical disruption as a “new normal” requiring ongoing adaptation rather than isolated responses to discrete events. The following sections present findings based on interviewees' explanations of their firms' adaptive responses; the interviewees and their associated firms are listed in Supporting Information A1.

The conservation phase is characterised by stability and rigidity in dependency structures, increasing connectedness and rigidity in relational and SN structures. However, these dependencies become vulnerable to systemic shocks. Before the geopolitical disruptions, the Nordic CRM-dependent SN operated in a landscape perceived as stable and low risk despite structurally exposed supply bases. Across the interviews, informants described decade-long supplier relationships as largely uninterrupted and rarely reconsidered (e.g. E1, F1).

In the case of the Russia–Ukraine war, several informants described earlier-established supplier bases in Russia for CRMs such as nickel, helium and coking coal (e.g. D1, D2, E1, L1, M1). Informants frequently associated Russian sourcing with lower transportation costs, given Russia's proximity to the Nordics. For example, K2 explained that “European companies have been using Russian coking coal for decades” and “if you don't have to spend that much money for transportation, then it's better for you”. Likewise, L1 stated that “Russia is one big producer of nickel. […] we bought a lot [from Russia] in the past”, while D2 highlighted that “we use helium 3, which is very rare, and it is something that is in quite limited supply globally [… …]. The original source was from Russia”. These established sourcing arrangements reinforced dependency on existing Russian supplier bases and concentrated sourcing structures to secure continuity of supply. Thus, predictability in volumes and long-term supplier stability reinforced the perception that no significant strategic action was required.

A parallel pattern was evident in China. Despite escalating US–China tensions and Chinese government restrictions on CRMs, informants suggest that firms continue to treat China sourcing as business as usual, with no interruption to CRM. Informants reported greater reliance on China's CRMs because it is a global hub for CRM extraction. J1 emphasised the depth of dependency: “If we look at it in the EU context, for instance, rare earth elements, there is no production within the EU, so 100% is imported”. N3 further elaborated that in Nordic CRM SN, “we are far away from security of supply. We are totally dependent on bigger markets. That means 70% of all commodity markets are regulated by China, and we're talking about high monopolies. I mean, they're talking about 95% of market share”.

The interviews revealed that firms are structurally exposed to escalating tensions between China and Taiwan due to their dependence on semiconductor production concentrated in Taiwan. Informants remarked on their limited ability to strategically respond despite the catastrophic consequences of a Chinese invasion of Taiwan. C1 states that “the electronic industry is always worried about what happens between China and Taiwan. If something happens there, you cannot be prepared. You're screwed anyway”. Amidst the uncertainties, the firms and SN overall remain largely in a conservation phase, not because geopolitical tensions are absent at a higher system level, but because firms viewed the potential disruption as unmanageable and fatal to the whole SN.

Overall, we did not observe major RDT strategies related to CRM dependency during the conservation phase. Although some firms adopted a strategy of negotiating the environment through dual or multi-sourcing, these practices were generally viewed as routine for maintaining performance or mitigating everyday uncertainties rather than as measures taken in anticipation of geopolitical disruptions.

The same is observed in Figure 3. The 2013–2017 period was characterised by a relatively stable and persistent SN structure despite emerging geopolitical tensions. Visual inspection, together with the network measures, indicates a stable core network architecture. Global density remained nearly constant at 0.002 between 2013 and 2016, increasing only marginally to 0.003 in 2017, while mean ego density fluctuated within a narrow range (0.223–0.242) and network centralisation remained similarly stable (0.209–0.226). Although Russia's annexation of Crimea in 2014 coincided with selective relational adjustments among some firms, these changes did not substantially alter the overall structure of the SN. At the same time, tie churn increased markedly from 0.242 in 2015 to 0.404 in 2016 and 0.431 in 2017, indicating growing relational reconfiguration beneath an otherwise stable network architecture, potentially reflecting mounting adaptive pressures characteristic of a late conservation phase.

The geopolitical tensions, including the Russia–Ukraine war, escalating US tensions and tightening export restrictions, pushed the Nordic CRM-dependent SN out of its previously stable conservation state and into varying levels of release. Across the interviews, informants described contract and pricing uncertainty, hidden dependency exposure, lead-time uncertainty, market exclusion, material shortages and redirected investments. In response, firms adopted varying strategies by avoiding dependency, creating the environment and negotiating with the environment, depending on the severity of the disruptions.

The Russia–Ukraine war produced the most immediate disruption for the focal firms during the broader release phase, which had already begun to emerge in the SN around 2017, as seen in Figure 3. Several informants described rapidly disengaging from Russian suppliers despite having established sourcing relationships earlier. Firms primarily responded by avoiding dependency by exiting politically risky partners. As J1 emphasised, “we all understand that this connection to Russia is problematic when operating within the EU, and we are doing everything we can to get rid of it”. Firms terminated relationships even when suppliers were not formally sanctioned. J1 noted, “The companies were dealing with Russia, so that itself ended project or two, even though these companies were not sanctioned and still are not”. The disruption of Russian sourcing arrangements created immediate uncertainty in managing dependencies on CRMs. The disruptions led to market shortages and exclusion, prompting firms to adopt a strategy of avoiding dependency by diversifying their supplier base. A1 explained this response strategy: “The biggest thing is that there was a shortage in the market, so smaller players didn't get the material […]. We then find out alternative options from the South of the EU”. The interviews further showed that firms were exposed to hidden upstream dependencies for CRMs originating from Russia, prompting them to negotiate with the environment through market scanning. Following the war, firms increasingly sought to trace the origins of raw materials and components to ensure that their products did not contain Russian inputs. For example, C1 specified “we haven't had any Russian components on our products, but we have needed to check a couple of times that we are not getting any components from Russia”. Moreover, D1 emphasised how globalised networks complicate even basic traceability: “sometimes it's almost mission impossible to find out if the component has material in those forbidden countries or not”.

At the SN level, the earliest signs of release emerged in 2017–2018, coinciding with rising geopolitical turbulence and doubts about globalisation (Contractor et al., 2026). Between 2017 and 2018, nodes declined by approximately 24% and edges by 29%, accompanied by pruning of peripheral actors and more star-like structures. These changes suggest the beginning of dependency restructuring and increasing concentration around core relationships. The Russia–Ukraine war subsequently intensified these release dynamics, with continued contraction observed between 2021 and 2023, and interviewees reporting supplier exits, diversification efforts, market scanning, and increased requirements for supply chain visibility. Notably, the ego-network analysis suggests that the initial release dynamics were concentrated primarily in the broader SN rather than in the focal firms' immediate SNs. While the aggregate SN lost approximately 29% of its ties between 2017 and 2018, the number of direct ties among the focal firms remained largely unchanged (approximately 1% decline). In contrast, between 2021 and 2022, the direct ties of focal firms declined more substantially (approximately 12%), indicating that the Russia–Ukraine war was more directly reflected in the focal firms' immediate relationship structures and aligning more closely with the interviews.

The interviews also revealed that while several firms attempted to reduce exposure to politically exposed dependencies, others sought to create an environment by leveraging regulatory loopholes or expectations. Some firms continued to access Russian-origin CRMs indirectly through intermediary structures that remained operational despite sanctions. For example, J1 explained that “the oligarch owning it is Russian, and it is still not sanctioned within the EU. They can import nickel from Russia to [specific EU location]. The site is still operational. So, in a way, it has not affected, which is a bit of a surprise”. Another observed pattern was actions by powerful actors to influence CRM pricing during times of crises. A1 provided an example of such an incident: “It's a more or less illegal pattern. […] There are ten people sitting together in some places and so on. Everybody knows what the market level, for example, in Finland and Poland or whatever, is, and then suddenly all the prices are about the same level”.

Beyond the Russia–Ukraine war, disruptions from China's export restrictions destabilised firms' SNs, creating pricing uncertainty. These export restrictions led firms to pay higher prices for CRMs. For instance, E1 described the higher price of silicon produced in China, which was affected by the Chinese government's restrictions. The informant described how the response involves maintaining dependency structure and negotiating with the environment through customer-facing price hedging. E1 explained that: “due to the volatility on the market, we have been working […] at quarterly or semester levels [with customers]” and using index market benchmarks “like long-run metal exchange hybrids or SMM [Shanghai Metals Market]”.

Informants described disruptions resulting from escalating US tensions, including industrial policies, US–China tensions and tariff measures, as a persistent source of disruption. For instance, J1 stated that the Inflation Reduction Act (IRA) disrupted CRMs' SN in the Nordics by redirecting green-transition investments. J1 explained: “One of the most significant impacts wasn't actually Trump's administration, but rather Biden's […]. It was the IRA. It was significant and had an impact, maybe the largest one compared to almost anything. They [the US government] tried to expedite investment in green technologies, and the funding was so significant that almost all the projects […] shifted to North America”. In this case, firms primarily experienced disruptions through higher-level political-system interactions that redirected investments within the Nordic CRM-dependent SN.

Furthermore, US–China tensions and US trade tariffs on Nordic countries heightened uncertainty. The informants pointed to hidden dependency exposures and pricing uncertainties during trade tariff exposures. Consequently, firms adopted strategies that negotiated with the environment through market scanning and adjusted pricing models. As tariff-related dependency assessments increased, C1 described that they needed to calculate the “China content” of their products to anticipate future restrictions. Similarly, firms responded to the tariff measures by adjusting pricing models. I1 described a tariff environment in which products were either subject to 0% or 25% tariffs, depending on their classification under the US–Mexico–Canada Agreement. Firms responded by negotiating with the environment, particularly by adjusting pricing models (D2, F1, L2, C1, D1, G1). As I1 puts it, “you can imagine it's 25% of the selling price in the US. It's a lot, right? […] We say, take it or leave it. So, either you pay, or we don't deliver”. From 2024 to 2025, the SN again exhibited signs of selective contraction broadly similar to the earlier release dynamics. The number of nodes declined by approximately 27% and edges by 28%, while global density increased from 0.003 to 0.004 and focal ego density from 0.219 to 0.237, suggesting renewed consolidation around core relationships and increasing local embeddedness amid ongoing geopolitical uncertainty.

Beyond immediate responses to geopolitical tensions, firms continued to adapt through incremental adjustments to CRM SN dependencies as opportunities for reconfiguration emerged. During the reorganisation phase, firms in the Nordic CRM-dependent SN began experimenting with new structures, capabilities and sourcing arrangements to restore system functioning under altered geopolitical conditions. G1 described this transition as: “[Earlier] there have been some strategies, but I would say typically these are contingency plans due to availability or maybe cost-benefit, just for a quick response. Let's say a correction of the situation, but then later on we start with the local actions, for example, regarding competitiveness, to recap the competitiveness”. During the reorganisation phase, firms began implementing strategies of absorbing the environment, avoiding dependency and creating the environment.

One of the predominant strategies during the reorganisation involved avoiding dependency by geographically reconfiguring supplier bases. A1 explained that “Supply chains are going a little bit more local. Everyone has started to question whether it is still better to source within the EU, because goods from China, Asia, or South America are becoming too hard to handle after COVID, the war, and the tariffs”. Many firms adopted an avoiding dependency strategy by substituting constrained CRMs where feasible. Across interviews, recycling emerged strongly. K2 explained that “our scrap ratio for certain products goes up to 96%”, noting that “when nickel batteries come to the end of life, you get nickel units out of battery waste”. For some materials, alternatives were more radical. For example, D2 described a transition in helium sourcing from the moon. As D2 explained, firms also attempted to build additional supplier options, “trying to build a second source”, even if potential suppliers “have not been able to yet reach the quality levels that we need”.

Substituting CRMs with alternative materials frequently intersected with severe capacity constraints, another defining feature of this phase. N1 illustrated challenges in recovering CRMs through recycling: “Take a cup of coffee. Brew your coffee. Put milk in it, put sugar in it, and then recycle it back. OK, I got the sugar, the water, and the milk out. What about 100s of aromatics? Zero to 40% is possible, and over 40% is mostly impossible in most of the critical materials. In the case of elements, only 1%”. Capacity constraints extended beyond materials to production. Localised manufacturing struggled to keep pace with rising demand (K1). B1 noted that local suppliers struggle because they cannot increase their capacity to the extent the buyer desires. Firms responded by creating additional network capacity, conducting market scanning and leveraging weak ties or intermediaries to broaden access to constrained resources.

Firms restructured their dependency on CRM by absorbing the environment through investments in upstream CRM projects to secure future access and by building internal strategic resources and in-house CRM capabilities. Several firms joined the mining development projects to secure future access. L2 described such efforts in Greenland: “once they are up and running, ten years from now, every year we are entitled to have some volumes from there”. Other firms expanded storage to buffer against future disruptions. F1 explained that they were “developing extra storage capacity” to store more CRMs.

Geopolitical tensions between the USA and China prompted firms to adopt measures to avoid dependency by geographically reconfiguring their supplier bases. As the trade war escalated, several informants described diversifying into alternative locations. B1 explained exiting Chinese suppliers as “in the last year, the total spend on China-related items was about 6%. We want to decrease that because the needs come from our customers […] we are trying to focus on the other part of Asia, as we know there are countries like Thailand, Malaysia, and Vietnam”. Similarly, firms, in response to tariff measures, accelerated efforts to regionalise sourcing and production activities to derisk dependency on China. I1 stated: “There has especially been pressure for the last two years, also to get China out of the equation […] we said we're never getting to 100%, but we're getting to 80% level where we produce 80% in the region we sell it”. While 2022–2023 exhibited release characteristics, structural changes in 2024 suggest early reorganisation and partial exploitation within the SN. Nodes and edges increased by approximately 48% and 49%, respectively, while both global density (0.004–0.003) and focal ego density (0.223–0.219) slightly declined. These patterns suggest that firms began expanding and diversifying their relational structures following the earlier contraction period, which corresponds with interview evidence on supplier diversification and the development of alternative sourcing arrangements.

Finally, the reorganisation phase was marked by creating the environment through establishing new norms. This involved a redefinition of sourcing strategies, with informants emphasising rebuilding their supplier portfolios and geographical footprints, and clarified expectations for acceptable sourcing practices. For example, N1 highlighted: “In China, they use very, very nasty chemicals. We would not consider that legal anywhere here in Europe or the States, to that matter”. By the time Nordic firms entered sourcing from Asia, these norms were applied more systematically. As E1 stated, “When we do the human rights due diligence, we do it also in China”.

In the Nordic CRM-dependent SN, the exploitation phase was characterised by expanding material capacity, leveraging accumulated assets for competitive advantage, extending operations into additional geographies and selectively growing with high-value suppliers. Across interviews, the most visible strategies were creating the environment and negotiating with the environment.

A central feature of exploitation was the scaling up of material capacity (e.g. I1, J1, B1). Firms moved beyond emergency responses and began building higher-volume operations on top of the new supply structures established during reorganisation. As J1 noted, their objective is to scale across the Nordic region and to reshape national CRM capabilities.

Another defining element of exploitation involved leveraging existing assets to gain a market advantage. Informants described how, once the turbulence of release and the experimentation of reorganisation had subsided, firms focused on optimising flows, consolidating orders, and using their financial and operational strengths to negotiate more favourable terms (e.g. B1, A1, K1, C1). As B1 explained, after the reorganisation that led to production moving to the USA, they now choose global suppliers that “can provide the same parts to the US and Finland” to gain a price advantage.

Selective growth with high-value suppliers was also mentioned (e.g. A1, L1, D1, D2). Once firms had established new relationships during reorganisation, exploitation involved deepening these ties and expanding business with those partners that met strategic criteria. This frequently involved negotiating the environment through long-term contracting and resource planning, complemented by creating an environment aimed at developing network capacity. For example, L1 emphasised the importance of maintaining relationships with key CRM producers: “We need to have a good relationship with everybody who is producing or trading”. This sometimes required maintaining volumes even when suppliers were the cheapest option. As L1 put it, firms prefer to keep suppliers “in the loop and as your friends”, while D2 highlighted that the objective was to support suppliers “to be able to grow with us”.

This study aimed to uncover the adaptive response of firms in a CRM-dependent SN amid environmental uncertainties arising from geopolitical tensions. We elaborated on the RDT strategies (Pfeffer and Salancik, 2003) by integrating them with the adaptive cycles (Holling, 2001; Wieland, 2021) to explore how adaptive responses unfold over time.

Firstly, our findings reveal that the stability and rigidity of established sourcing arrangements predispose firms to geopolitical disruptions. However, in some cases under higher levels of dependency, firms tend to normalise geopolitical risks rather than respond adaptively. For instance, firms exposed to vulnerabilities stemming from China–Taiwan tensions acknowledge the potentially catastrophic consequences of escalation yet continue to rely on their supply dependency. As Taiwan is a critical node in the global SN of semiconductor chip production, accounting for 60% of chip production and 90% of advanced chip output (The Economist, 2023), firms are structurally locked in and unable to develop adaptive responses to escalating geopolitical tensions. We contend that the normalisation of geopolitical risks reflects a lack of alternative strategic options by firms to reduce their dependencies. This finding aligns with Bode et al. (2011), who argued that when dependencies on exchange partners are high, they are likely to pursue stability during disruptions. Thus, we suggest that the normalisation of geopolitical disruptions is contingent on the availability of viable alternative sourcing options.

While Holling (2001) and Wieland (2021) emphasise the release phase as a point at which accumulated relational and structural capitals are quickly dissolved, our findings provide partial support for this. Firms adopted different strategies to mitigate geopolitical disruptions. When disruptions involved sanctions, firms avoided dependencies by terminating supplier relationships. In contrast, when disruptions involved protectionist measures, such as China's CRM export restrictions and US trade tensions, firms negotiated with the environment through pricing adjustments and enhanced market scanning. These findings suggest two release dynamics: avoiding dependencies in response to sanctions and negotiating with the environment in response to protectionist measures. This insight aligns with Free et al. (2025), who found that the release dynamics are asynchronous and selective, reflecting partial release rather than complete dissolution.

Interestingly, while many firms unhook their supplier bases from CRMs in Russia due to sanctions (Srai et al., 2023), our findings revealed that other actors continue to leverage regulatory loopholes or expectations to import CRMs from Russia into the EU. This finding provides evidence of infiltration, whereby firms exploit loopholes in government oversight to engage in illegal operations (Sodhi et al., 2025).

After responding to immediate geopolitical disruptions, firms continually restructured their dependencies by substituting constrained CRMs. The findings revealed that substitutions for constrained CRMs were attempts by firms to circumvent the strategic need for primary CRMs (Jiang et al., 2023). This aligns with Kalaitzi et al.’s (2018) findings that when scarce natural resources are important, firms alter or reduce their dependency on them. However, the findings revealed that avoiding dependencies on primary CRMs led to capacity constraints, as it creates emergent sources of uncertainty and coordination challenges for the substituted CRMs. Firms respond to these by creating additional network capacity, enhancing market scanning and using intermediaries. These insights suggest that the effectiveness of avoidance strategies depends on the availability of resource substitutes.

Our findings also highlighted that firms avoid dependencies by withdrawing from established supplier bases through geographic reconfiguration. Firms redesigned their SNs in response to tariff measures on China, prioritising diversification to other locations such as Malaysia, Taiwan and Vietnam, and regionalising sourcing and production through nearshoring in Mexico or localisation in the USA to mitigate persistent tariff-related uncertainties. Our findings support recent studies suggesting that firms reduce their structural arrangements by reconfiguring their SN to mitigate tariff exposures (Chakkol et al., 2024; Moradlou et al., 2025).

Further, the evidence showed that firms' absorption strategies become apparent during the reorganisation phase. These included investment in upstream mines. As prior studies have highlighted, firms reduce dependency constraints by absorbing their dependencies, either fully or partially (Casciaro and Piskorski, 2005; Hillman et al., 2009). Therefore, we contend that firms facing geopolitical disruptions partially mitigate their dependencies by forming partnerships to ensure access to CRMs. These findings are at odds with Kalaitzi et al.’s (2018) proposition that discretion over resources, arising from geopolitical risks and government regulations, is negatively associated with buffering and bridging strategies.

In the exploitation phase, firms renewed their growth trajectory to mitigate uncertainties in their CRM and exchange partner dependencies by creating and negotiating with the environment. As Pfeffer and Salancik (2003) posit, when dependencies are beyond firms' control, rather than adapting to environmental uncertainties, firms alter the environment to meet their demands. Consequently, we observed that firms create the environment by reshaping national CRM capabilities and intensifying efforts to scale up local extraction and production. In another pattern, firms negotiate the environment by enhancing their boundary-spanning roles, such as selectively growing with high-value suppliers with whom the uncertainties in the dependency relationship will be minimised.

Based on our findings, we developed an empirically informed framework (Figure 4) that conceptualises resource dependency strategies as phase-contingent at the firm level and illustrates the implications for SN structural adaptation. During the conservation phase, as dependencies are deeply embedded and switching costs are extremely high, firms rely on routine, low-intensity responses, resulting in a uniformly weak deployment of strategies. At the SN level, during periods of relative stability, the SN is likely to exhibit dense and cohesive characteristics, reflecting persistence among actors. While this increased connectedness, coordination and structural consolidation, it also heightened the network's vulnerability to disruption. At the release phase, firms tend to prioritise negotiating with the environment and avoiding dependencies, as the main objective is to restore short-term flexibility under high uncertainty. Negotiating with the environment allows firms to rebalance dependencies without irreversible commitment, while avoiding dependencies enables rapid disengagement from risky SN ties without presupposing a stable future configuration. At the SN level, these responses contribute to the pruning of network ties and the erosion of established network configurations. As a result, SN tends to converge on a more centralised, star-like structure, in which powerful actors attempt to gain greater control and visibility within their network. During reorganisation, firms opt to absorb the environment and avoid dependencies, as this stage centres on redesigning dependency structures to regain control over critical CRMs. Absorption strategies support uncertainty reduction, while avoidance strategies eliminate exposure to politically risky inputs and regions. At the SN level, these responses drive diversification, particularly in 2024, when the network expanded through the formation of new ties, alternative sourcing arrangements and increased decentralisation. Finally, in the exploitation stage, having re-established a degree of control over CRMs, firms increasingly emphasise creating and negotiating the environment to shape favourable institutional, geographic and relational conditions. Creating the environment allows firms to influence the structure of resource availability, while negotiating the environment enables selective growth with partners that minimises future uncertainty. Absorbing strategies remain present but moderate, as firms balance internalisation with scalability. The SN entered a consolidation phase characterised by selective clustering around key actors and the reinforcement of core relationships reflecting strategic growth and expansion.

This study makes several contributions to theory. Firstly, we advance RDT strategies for managing uncertainty amid geopolitical shocks. While scholars have long suggested that firms manage dependencies on the external environment by adopting two distinct strategies of bridging and buffering (Bode et al., 2011; Kalaitzi et al., 2018), these strategies remain homogeneous without much consideration for the heterogeneity of the external environment and their implications for firms' dependencies on CRMs and exchange partners (Jiang et al., 2023). Our findings elaborate on the RDT strategies by showing that firms develop adaptive responses of absorbing the environment, avoiding dependencies, creating the environment and negotiating with the environment in response to geopolitical shocks that pose uncertainties in managing dependencies.

Secondly, we advance calls to integrate RDT with other theoretical lenses (Hillman et al., 2009; Jiang et al., 2023) by combining RDT with the adaptive cycles of panarchy theory to explain the evolution of SN structures. We found that at the release stage, depending on the severity of disruptions, firms engage in partial release (Free et al., 2025). Firms impacted by the Russia–Ukraine war fully terminated their dependency structures, emphasising the dissolution of accumulated relational lock-ins, reducing rigidity and increasing adaptive capacity to move to the reorganisation stage. However, firms affected by US–China tensions first maintain their relational structures and negotiate with the environment before considering restructuring dependencies by reconfiguring the SN.

Thirdly, while the literature portrays adaptation as an expected and desirable organisational response to geopolitical disruptions (e.g. Roscoe et al., 2022; Zheng et al., 2025), our findings reveal important limits to this assumption. We show that firms do not adapt uniformly, continuously or automatically to geopolitical disruptions and tensions. Instead, under conditions of extreme dependency and structural constraint, firms may rationally choose non-adaptation, normalising geopolitical risk rather than pursuing costly, uncertain or potentially ineffective SN reconfiguration. This challenges prevailing resilience-oriented assumptions that equate exposure to disruption with adaptive action and calls for a more socio-politically grounded understanding of SN design in fragmented geopolitical environments (e.g. Golgeci et al., 2025; Paulraj et al., 2024). More broadly, our findings suggest that adaptation is selective, temporally uneven and at times strategically avoided. Firms may delay adaptation when geopolitical trajectories remain ambiguous, refrain from adaptation when dependencies are structurally unavoidable or actively avoid reconfiguration when adaptation would undermine legitimacy. In this sense, non-adaptation emerges as a rational and consequential strategic response under conditions of extreme dependency.

Our findings offer relevant insights for firms operating in CRM-dependent SNs. Firms' reliance on a single supplier and maintaining a stable SN in a single geographic location are likely to predispose firms to geopolitical shocks. Our framework provides temporal responses for each phase of adaptation and could, thus, benefit managers. Firstly, when geopolitical disruptions arise, we recommend that, in the short term, firms should seek to avoid dependencies by rapidly disengaging from risky SN arrangements, particularly when the severity of the disruptions entails sanctions. When firms respond to tariff-induced disruptions, managers can negotiate the environment to adjust pricing models rather than immediately seek to fully restructure dependencies. Secondly, we encourage firms to adapt their dependencies in the medium term by absorbing the environment and avoiding them. Firms can absorb the environment by substituting constrained CRMs with secondary materials, identifying alternative CRM sources and developing new network capabilities. Additionally, firms may avoid dependencies by reconfiguring SN structures to alleviate vulnerabilities to shocks stemming from geopolitical disruptions. Thirdly, once managers establish controllability following the reconfiguration of dependencies, they can, in the long term, transition towards a phase of strategic renewal and growth by creating and negotiating the environment. Managers may create the environment by reshaping efforts to scale localised CRM capabilities and negotiate the environment by strengthening relational governance with high-value partners.

Beyond firm-level implications, our findings inform government regulations and policymakers' decision-making. Given the strategic importance of CRMs for advancing nations’ economies and the recent geopolitical competition to dominate their SNs, policymakers' roles in shaping CRM capability become salient. We recommend that policymakers at the national and regional levels engage closely with firms to identify essential CRM needs and support coordinated investment to scale up local efforts in extraction, refining, production and secondary-material recycling. Our findings show that although sanctions and restrictions on the flow of CRMs are formally in place, material flows may continue through semi-legal channels in some cases. While such practices may be justified by national strategic objectives, they are perceived very negatively by firms that adhere strictly to regulatory frameworks. This perceived inconsistency undermines trust, distorts competition and weakens the credibility of sanctions regimes. Greater policy coherence, transparency and explicit acknowledgement of these practices, rather than their implicit toleration, are critical.

This study has limitations and offers avenues for future research. We collected data from informants' retrospective and concurrent accounts and examined temporally overlapping geopolitical tensions rather than isolating specific geopolitical events, which is consistent with prior research (Moradlou et al., 2025; Roscoe et al., 2022). While we employed procedures to mitigate retrospective bias and triangulated interview insights, we call on future studies to adopt a forward-looking, real-time process research design (Grimm et al., 2024) to capture geopolitical tensions and organisational responses as they unfold. Longitudinal case study designs can be useful for uncovering mechanisms and contingencies that shape adaptive responses during geopolitical tensions. When applying the adaptive cycle lens, longitudinal designs may provide deeper insights into how quickly firms persist, adapt and transform across the adaptive cycle's phases and the mechanisms that influence the speed of these transitions. In this regard, a useful pathway for future research is to explore how mechanisms such as organisational path dependencies (Sydow et al., 2009) and inertia (Barker et al., 2026) shape adaptive response.

Our findings are limited to manufacturing firms operating in CRM-intensive sectors within the Nordic region. Thus, the findings should be interpreted within their contextual boundaries rather than as statistically generalisable claims. As SNs dependent on CRMs are exposed to geopolitical tensions, we concur that our findings may be transferable to CRM dependent SNs in other empirical or industrial contexts. We therefore encourage future research to conduct studies across other contexts or industrial sectors to validate, refute or extend our findings. An example for future studies pertains to the geopolitical tensions concerning US–Iran tensions over the Strait of Hormuz, which are affecting other strategically important raw materials such as crude oil, fertilisers and helium (Geldard, 2026).

Finally, as we collected data from senior executives responsible for global procurement and OSCM, we recognise that executive risk cognition could influence strategic adaptation or non-adaptive responses. The threat rigidity thesis shows that managerial threat-interpretation bias can limit operational resilience during disruptions (Dankyira et al., 2024). Could the same be true for geopolitical tensions? We encourage future studies to investigate how managerial cognitive interpretations of geopolitical tensions may shape their strategy selection and its consequences on adaptive responses.

The supplementary material for this article can be found online.