How does environmental regulation affect the cobalt supply chain resilience: empirical evidence from cross-border data

Constrained by resource endowment, the traditional cobalt supply chain model oriented towards efficiency maximization struggles to effectively respond to sudden risks such as supply disruptions. Therefore, enhancing cobalt supply chain resilience (CSCR) is becoming a critical issue for the sustainable development of the mineral industry sector. However, it remains unclear whether environmental regulation (ER) has an impact on CSCR. This study constructs a theoretical framework for analyzing the influence of ER on CSCR and employs econometric methods to conduct an empirical examination using a sample of 22 representative countries (regions) with accessible data. The research findings are as follows: (1) ER can significantly enhance CSCR, and this conclusion remains valid after a series of robustness tests. (2) ER functions through three channels: First, it drives green technological innovation and enhances the internal production capacity of the supply chain. Second, it guides the diversification of import sources and optimize the structure of the supply chain network. Third, it attracts high-quality foreign direct investment, leverage capital spillover effects, and enhance the vitality of the supply chain system. (3) Political stability plays a positive moderating role in the above process: A stable institutional environment can enhance the resilience dividend of ER. (4) the impact of ER is heterogeneous across different economic scales and the locations of supply chain. The research provides a new resilience theoretical perspective for understanding the relationship between environmental policies and resource security, and offers data support for coordinating ecological environment governance and stable mineral supply.

1 Introduction

With the increasing frequency of extreme climate events, the accelerated transition to renewable energy, and the surge in demand from emerging technologies, the stability and security of mineral resources supply chain have drawn a central issue in the geopolitical competition of major powers (Turner, 2022). Cobalt, as a key element in the ternary cathode materials of power batteries, plays an irreplaceable role in improving battery energy density, extending cycle life, and ensuring thermal stability (Gent et al., 2022). It is regarded as a critical resource supporting the global transition to clean energy (Bahini et al., 2024). According to the International Energy Agency (IEA), global electric vehicle sales exceeded 10 million units in 2022 (IEA, 2023), driving an increase in the cobalt consumption share of lithium-ion batteries (Maisel et al., 2023). It is estimated that by 2030, the demand for cobalt from clean energy technologies will grow by 2–4 times compared to 2023 (IEA, 2024). However, the cobalt supply chain exhibits extreme geographical concentration and significant environmental and social risks. Data from the United States Geological Survey (USGS) shows that the Democratic Republic of Congo (DRC) accounts for 73% of global cobalt production (Figure 1), while China produces 76% of the world’s refined cobalt (USGS, 2022), forming a fragile “Africa mining—Asia processing—global consumption” triangle. This highly segmented structure makes the cobalt supply chain more exposed to cross-border regulatory spillovers, geopolitical risks, and trade frictions, thereby rendering its resilience dynamics fundamentally distinct from those of other critical minerals. Existing studies have indicated that cobalt production will be insufficient to meet demand by 2050 (Valero et al., 2018; Sun et al., 2023). To break through this predicament, merely predicting the supply and demand of minerals is no longer sufficient, and the focus must shift to strategies for enhancing supply chain resilience. This is because short-term fluctuations are merely symptomatic, while resilience represents the long-term capacity of anticipate, buffer, and reconfigure in response to risks across the entire chain. Resilience not only determines the rise and falls of individual industries, but also profoundly shapes the global energy landscape, the trajectory of major-power strategic competition, and the ultimate realization of sustainable development goals worldwide (Sovacool et al., 2020).

Figure 1

Figure 1. Global cobalt production distribution in 2024.

Based on the general understanding of supply chain resilience (Pettit et al., 2010), we define cobalt supply chain resilience (CSCR) as the capacity of the cobalt supply system to maintain stable supply of cobalt and achieve rapid recovery, structural reorganization, and continuous adaptation through the synergistic integration of material, information, and capital flows when confronted with internal and external shocks (Liu et al., 2025). This capacity permeates every stage of the cobalt resource life cycle, including exploration, mining, beneficiation, manufacturing, use, and recycling (Liu et al., 2023). Research related to the mineral supply chain will be summarized in detail in Section 2. It can be found that the endogenous influencing factors of supply chain resilience have been discussed extensively. Under the dual pressure of severe resources supply risks and strengthened global environmental governance, it is of vital importance to build a global mineral resources security system (Shi et al., 2025). However, the existing studies on external policy intervention, especially environmental regulation as a key tool for coordinating economic development and ecological protection (Zhao et al., 2022), are largely based on manufacturing industries or relatively integrated mineral supply chains, and their shaping mechanism and comprehensive effects cannot be directly extrapolated to cobalt.

Environmental regulation typically refers to government-imposed constraints on corporate production and operations through the formulation of laws, standards, economic incentives, or information disclosure requirements, aiming to reduce environmental pollution, resources consumption, and ecological degradation. The process of environmental regulation internalizes the external environmental costs of economic activities and promotes the green transformation of entire society (Yang et al., 2021). Under the economic paradigm, environmental regulation research in China usually adopts the quasi-natural experiment of the promulgation of the Environmental Protection tax law (Kong et al., 2025), but the environmental policies of various countries (regions) are diverse and the implementation times are different, thus making comparisons impossible. In this regard, the practices of international research that utilize green tax revenues (Mpofu, 2022) and government environmental protection expenditures (Akdag et al., 2025) offer valuable insights for measuring environmental regulation in cross-national samples. In addition, existing research has explored the impacts of environmental regulation on general supply chain resilience: On the one hand, increased compliance costs may raise production costs, reduce short-term competitiveness, and lead to industrial relocation (Yin et al., 2023). On the other hand, effective environmental regulation is also considered capable of stimulating innovation through the Porter Hypothesis effect, improving resources utilization efficiency, optimizing process flows, and even creating new market opportunities, thus enhancing the green competitiveness and technological resilience of the supply chain in the long term (Peng et al., 2021). Moreover, environmental regulation can improve the overall environmental risk management level of the supply chain by imposing environmental performance requirements on suppliers (Jiang et al., 2023).

However, when we shift our focus to the mineral supply chain, the ultimate effects of environmental regulation become more complex and theoretically contentious. On the one hand, stringent environmental regulation may increase operational costs, prolong permitting processes, or restrict capacity expansion in the resource extraction and processing stages, potentially exacerbating supply shortages and weakening the responsiveness and efficiency of the supply chain in the short term (Nasir et al., 2023). On the other hand, advancing the clean production of mining, reducing community conflicts caused by environmental pollution, enhancing resources recovery rates in smelting, and promoting responsible mineral sourcing practices are key to building a long-term, stable, sustainable, and socially adaptive supply chain resilience (Jia et al., 2024). Although regulatory approaches to addressing mineral supply chain risks have become a current research trend (Franken and Schütte, 2022), whether environmental regulation acts as a cost burden or a driving force for transformation remains empirically untested. Furthermore, given the global material flow characteristics of cobalt resource, the analytical perspective of this study transcends the boundaries of a single country. Therefore, when exploring the impact mechanism of environmental regulation, we should not be confined merely to domestic production capacity, but rather delve deeply into the role they play through domestic foreign trade and attracting foreign direct investment. This is because all links in the global cobalt supply chain are highly interconnected. The adjustment of a country’s environmental policies not only affects its domestic production capacity but also has a linked impact on the resilience of the global cobalt resource supply chain through the international trade pattern and capital flows.

Over the coming decades, competition and collaboration over the acquisition, control, and supply of mineral resource will constitute a significant factor reshaping the international order (Ali et al., 2017). Thus, we focus on the impact of environmental regulation on the resilience of the cobalt supply chain, systematically examines the relationship between the two using econometric models, and quantitatively determines the magnitude and pathways of the impact. The marginal contributions of this article are as follows: First, unlike traditional supply chain research that focuses on efficiency or cost, we have established the resilience enhancement effect of environmental regulation. By deconstructing the triple transmission chain of “technological innovation - trade restructuring - capital flow”, we provide a new theoretical perspective for understanding the dynamic mechanism of environmental regulation. Second, the moderating effect of political stability is verified through cross-national data, suggesting that the quality of the institutional environment is a prerequisite for the generation of resilience dividends. This discovery breaks through the one-way discussion of policy tools themselves in existing literature, emphasizing the crucial role of effective governance capacity in the transformation of resource policies, especially providing a key entry point for developing countries to coordinate environmental goals with resource security. Third, the establishment of a unified quantitative system for the intensity of environmental regulation and supply chain resilience has overcome the comparability challenges brought about by differences in national policies, providing a methodological foundation for subsequent long-term collaborative research on resource and environmental policies, and also offering data support for the implementation of a more targeted governance framework.

The remainder of this paper is structured as follows. Section 2 reviews the relevant literature. Section 3 develops the theoretical framework and research hypotheses. Section 4 outlines the empirical strategy. Section 5 presents the empirical results and discussion. Section 6 is conclusions and policy recommendations.

2 Literature review

2.1 Mineral resource supply chain and resilience

Research on mineral resource supply chain has gradually evolved from an initial emphasis on risk identification and vulnerability assessment toward a more forward-looking focus on resilience building (Brusset and Teller, 2017). Early studies primarily concentrated on assessing the likelihood and potential impacts of supply disruptions (Tisserant and Pauliuk, 2016). They attributed supply chain vulnerability to factors such as the geographical concentration of production and processing activities (Patiño Douce, 2016), geopolitical instability and trade restrictions (Vivoda and Matthews, 2024), market price volatility (Buchholz et al., 2022), and increasing environmental and social governance pressures (Maybee et al., 2023). However, as large-scale “black swan” events have become frequent, merely identifying vulnerabilities and attempting to avoid risks has proven insufficient (Nagy and Szentesi, 2024). Consequently, the concept of supply chain resilience has gained prominence, marking a shift from static risk exposure toward the long-term capacity of supply systems to absorb shocks, adapt to changing environments, recover from disruptions, and reorganize structural configurations over time (Pettit et al., 2019). Existing studies have explored a wide range of determinants of supply chain resilience, including geological and technological conditions (Dominy et al., 2018), market and trade structures (Gnutzmann et al., 2020), political and institutional environments (Mancheri et al., 2019), and organizational coordination within supply networks (Lou et al., 2025). While this body of literature provides a solid foundation for understanding the multidimensional nature of resilience, it has largely emphasized endogenous structural and market-driven factors. By contrast, the role of exogenous policy interventions remains comparatively underexplored.

2.2 Environmental regulation and supply chain resilience

It is noteworthy that policy tools, especially fiscal, industrial, and environmental policies, often act as an invisible hand used by governments to implement macroeconomic control, achieve oversight of critical resource supply chain, and, in turn, enhance domestic resource security and resilience. For example, Pramusinta et al. (2025), based on a literature review, suggested that Indonesia’s nickel mining sector should strengthen tax regulation and transparency to maintain national economic resilience. Meanwhile, a study by Janikowska and Kulczycka (2021) in Finland found that developing appropriate mineral policies not only significantly increases mineral production but also helps reduce carbon dioxide emissions by over 20%, although this relies on substantial fiscal capital support. Furthermore, policy incentives and constraints have a profound impact on corporate behavior. Zhou et al. (2025) discovered a U-shaped relationship between Chinese government subsidies and supply chain resilience of the resource industry. On the international level, Franken and Schütte (2022) reviewed voluntary sustainability initiatives in the mineral supply chain of major countries, pointing out that regulatory pressures such as due diligence have made supply chain more transparent, though environmental risk assessments remain insufficient. Unfortunately, there is limited literature that directly addresses the role of environmental regulation. Sun and Hasi (2024) argued that environmental regulation can mitigate the negative environmental impacts associated with resource extraction, maximizing the economic benefits of mineral wealth while balancing the development of the mining industry with resource management. However, Mancheri et al. (2019), focusing on China’s rare earth elements, presented an almost opposite viewpoint: strict environmental standards undermine China’s downstream industry’s cost advantage in production, impacting the capital and operational expenditures of these companies. It is evident that existing research has not reached a unified conclusion regarding how environmental regulation affects the critical mineral supply chain resilience.

Furthermore, to concretize the concept of supply chain resilience, existing studies have developed several evaluation approaches, which can be broadly classified into three categories. First, qualitative methods, including expert interviews, the Delphi method, and case studies, are commonly used to assess resilience and its driving factors in the mineral supply chain, particularly when data availability is limited or conceptual frameworks are still evolving (Sinaga et al., 2024; Mohammadian et al., 2026). Second, simulation-based approaches such as system dynamics, agent-based modeling, and network analysis are widely employed to examine the dynamic responses and long-term evolution of supply chains under different shock scenarios. These methods offer insights into non-linear system behavior, but often face challenges related to computational complexity and parameter calibration (Lou et al., 2025; Xiang et al., 2025). Third, indicator-based methods construct composite resilience indices by weighting multiple indicators using techniques such as Analytic Hierarchy Process (AHP), Entropy Weight Method (EWM), Improved Entropy Method (IEM), Principal Component Analysis (PCA), or Data Envelopment Analysis (DEA). This approach enables cross-sectional and temporal comparisons of resilience levels, although the results remain sensitive to data availability and the subjectivity of weighting schemes (Aisyah et al., 2024; Liu et al., 2023; Liu et al., 2025). Accordingly, this study adopts an indicator-based evaluation based on IEM and TOPSIS, which enables the systematic weighting of multiple indicators and supports robust cross-country comparisons of CSCR.

2.3 Cobalt supply chain resilience

Finally, regarding cobalt resource, existing research has conducted some investigations. For example, in terms of supply and demand forecasting, Fu et al. (2020) used modeling methods to estimate that the demand for cobalt in 2030 could range from 235 to 430 thousand tons. In the analysis of supply sustainability, Earl et al. (2022) pointed out that enhancing the resilience and environmental sustainability of cobalt supply requires technological advancements, responsible sourcing, policy regulation, and circular economy practices. Regarding supply chain risk identification, Van den Brink et al. (2022) visualized the trade flows among cobalt-producing countries, mines, refineries and downstream companies using geographic data on the cobalt supply chain, and analyzed the supply risks. Furthermore, in the study of trade network resilience, Yu et al. (2023) constructed a directed weighted network of cobalt mining trade using upstream and downstream cobalt products and assessed its static structural resilience evolution. Similarly, Wang X. et al. (2025) developed an impact diffusion model using global cobalt trade network data to comprehensively analyze the effects of two scenarios (complete export disruptions and partial reductions) on demand countries. Sun et al. (2022) established a cascading failure model based on the cobalt trade network and derived an avalanche network to analyze the structural features of crisis propagation dynamics. Guo et al. (2023) explored the impact of geopolitical relationships on the evolution of cobalt trade networks. In terms of resilience quantification and assessment, Liu et al. (2023) applied a system dynamics model to build price, production capacity, and supply-demand subsystems for assessing cobalt supply chain resilience, simulating the impact of electric vehicle demand on resilience. Liu et al. (2025) constructed a comprehensive resilience assessment system for China’s cobalt supply, incorporating the structural features of the cobalt supply chain and various risk indicators into a unified framework.

The academic community has accumulated substantial findings in the field of critical mineral supply chain. However, several limitations remain. First, existing research tend to focus on the resilience challenges, opportunities and economic impacts of the mining sector or the entire resource industry, or studies on traditional bulk commodities such as oil, steel and coal. However, less attention has been paid to emerging strategic minerals such as cobalt. This gap hinders a comprehensive understanding of the resilience mechanisms specific to particular mineral supply chain. Second, while resilience assessment methods are diversifying, a unified and widely accepted set of evaluation indicators has yet to be fully established. Mainstream approaches, such as constructing trade network models or applying system dynamics to simulate resilience levels under trade disruptions and production supply risks, often rely on certain hypothetical scenarios. As a result, their outcomes can only reflect partial aspects of resilience and fail to capture the full complexity of resilience performance in real-world supply chain. Furthermore, although factors influencing resilience have been widely discussed, most studies remain at the level of theoretical correlation analysis, with a lack of empirical research using economic models to make causal inferences. Moreover, in the context of global energy transition, the consideration of environmental regulation factors remains insufficient, and existing viewpoints are even contradictory. This could be due to the exogeneity of environmental policies being overlooked, as well as the special industrial contexts in different countries (regions). Therefore, there is an urgent need to explore the universal laws between environmental regulation and the resilience of key mineral supply chain. To fill these research gaps, this study intends to take cobalt resource as an example to deeply analyze the mechanism by which environmental regulation affects the resilience of its supply chain.

3 Theoretical analysis and research hypotheses

3.1 Environmental regulation and cobalt supply chain resilience

The traditional view regards environmental regulation as a burden on the production cost of enterprises. However, within the framework of the mineral resources supply chain, this paper holds that environmental regulation can drive system optimization and significantly enhance its ability to resist disturbances, adapt and recover. The direct impact of environmental regulation is mainly reflected in the following dimensions.

First, environmental regulation can strengthen the material flow, enhance resource efficiency and recycling. Environmental regulation aims to restrain the negative environmental externalities such as acidic wastewater, cobalt-containing dust, and the risk of tailings pond dam failure that exist in the cobalt mining and refining process. The restrictions on wastewater and waste gas discharge, the safe disposal of tailings and the requirements for land reclamation rate have forced cobalt mining enterprises to abandon the extensive model, invest in advanced processes such as automated mineral processing and efficient smelting technologies, and establish closed-loop water systems to reduce resource consumption and waste production (Lèbre et al., 2017). More importantly, environmental regulation has promoted the establishment of a cobalt recycling and regeneration system. The EU’s New Battery Regulation and other policies clearly stipulate a minimum proportion of recycled cobalt in batteries and encourage the large-scale extraction of cobalt from used batteries, consumer electronics and alloy waste to build local “urban mines” (Hoarau and Lorang, 2022; Zhang et al., 2023). The circular economy has directly reduced the demand for primary cobalt resource and the reliance on resource-rich countries such as DRC. When the supply chain is disrupted, recycled cobalt, as a stable secondary resource, can effectively enhance the material basis and supply autonomy of the supply chain.

Second, environmental regulation can reduce the risk of operational disruptions, manage environmental incidents and compliance barriers. The lack of environmental management is one of the core causes of disruptions in the cobalt supply chain. Under lax supervision, incidents such as tailings pond leakage and acidic water pollution in mining areas occur frequently, directly triggering community protests, government-mandated production suspensions and lawsuits for exorbitant compensation. On the one hand, environmental regulation, through mandatory environmental risk assessment, prompts mining enterprises to take preventive measures, such as real-time monitoring of pollutants and establishing ecological restoration funds, thereby reducing the probability of accidents and the intensity of hazards, and avoiding physical supply disruptions (McLellan and Corder, 2013). On the other hand, the global ESG wave has transformed environmental performance into market access thresholds (Gao et al., 2022). For instance, the EU’s Digital Battery passport requires the disclosure of cobalt carbon footprints and supply chain due diligence management, and the US Inflation Reduction Act also sets strict environmental standards for key mineral sources. At this point, environmental regulation at the national level will be internalized as enterprises’ compliance capabilities, enabling them to avoid trade barrier supply disruptions such as order cancellations, port detention or brand boycotts caused by non-compliance with environmental protection standards, and even gain compliance premiums (Bahizire et al., 2024). Therefore, environmental regulation directly ensures the global circulation of cobalt products and reduces interference from non-market factors.

Third, environmental regulation can stabilize the supply cooperation network and shape environmental reputation and trust capital. In the interconnected and transparent global economy, the environmental performance of enterprises has become an important component of their brand value and trust capital. Strict environmental regulation prompts mining enterprises to voluntarily disclose environmental data, adopt international standards such as ISO 14001, and participate in the Fair Cobalt Alliance (FCA), thereby shaping a reliable environmental reputation. Due to the pressure from investor ratings and consumers’ strict scrutiny of the green purity of products faced by downstream battery manufacturers and vehicle manufacturers, cobalt suppliers with outstanding environmental performance have thus gained a reputation premium, making it easier for them to sign long-term purchase agreements with stability clauses and establish deeper strategic partnerships. At the national level, a sound environmental governance system is conducive to enhancing the credibility of the country’s mineral resource and promoting the formation of strategic alliances based on common environmental values with other countries that attach importance to sustainable development (Raze, 2023). Such cooperative networks based on environmental reputation, compared with pure price game transactions, will demonstrate stronger synergy in information sharing, resource allocation and risk sharing when facing crises, thereby directly enhancing the collective resilience of the supply chain.

Accordingly, we propose Hypothesis 1.

Hypothesis 1

Hypothesis 1. Environmental regulation has a positive impact on the cobalt supply chain resilience.

3.2 The impact mechanisms of environmental regulation

Environmental regulatory pressure can be transformed into the intrinsic driving force for enterprises’ technological leaps and reshape the supply chain resilience. Based on Porter Hypothesis and dynamic capability theory, environmental regulation mainly promotes technological innovation in enterprises through two paths (He et al., 2024; Peng et al., 2021): The one is that mandatory technical standards force enterprises to break the technological rigidity of the existing production mode, the other is that the innovation compensation effect provides incentives for enterprises to continuously invest in research and development.

In the context of the cobalt supply chain, the technological leaps triggered by environmental regulation mainly manifest in two categories: First, technological innovation enhances production efficiency. To achieve zero waste discharge and harmless tailings treatment, cobalt mining enterprises are constantly introducing green processes such as hydrometallurgical ion exchange technology and intelligent monitoring systems for tailings dry stacking, in order to optimize reaction efficiency, reduce material loss, and thereby minimize production fluctuations. A study on closed-loop water systems in mining (with ore sorting and tailings reuse) reports potential water savings of up to 40% in general mining contexts (Kinnunen et al., 2021), directly alleviating the constraints of water resource shortage on continuous production and enhancing the operational stability of the supply chain. Second, product innovation expands the path of resource substitution. Environmental regulation is accelerating breakthroughs in the research and development of cobalt-free or low-cobalt battery materials and efficient cobalt recycling technologies. When the supply from DRC is disrupted, a technological breakthrough to increase the extraction rate of recycled cobalt to over 95% can fill a supply gap of 500,000 tons in the short term (Botelho Junior et al., 2021). The industrial application of high-nickel and low-cobalt cathode materials has reduced the cobalt dependence per unit product, thereby weakening the structural impact brought by the risk of concentrated production areas and providing a more flexible adjustment space for the supply system.

Therefore, environmental regulation has strengthened the production capacity of the cobalt supply chain through green technological innovation: process innovation has consolidated the anti-interference ability of the production system, and product innovation has broadened the flexible space for resource substitution. The two together form the technical support for enhancing resilience. Therefore, it is proposed that:

Hypothesis 2

Hypothesis 2. From the perspective of internal production capacity, environmental regulation has a significant positive impact on the cobalt supply chain resilience by enhancing the level of green technological innovation.

Hypothesis 3

Hypothesis 3. From the perspective of foreign trade dependence, environmental regulation has a significant positive impact on the cobalt supply chain resilience by diversifying import sources.

Hypothesis 4

Hypothesis 4. From the perspective of international capital participation, environmental regulation has a significant positive impact on the cobalt supply chain resilience by increasing foreign direct investment.

3.3 The moderating effect of political stability

The role of environmental regulation is highly dependent on the institutional support provided by the national governance foundation. Among them, political stability, as a core governance element, profoundly regulates the effectiveness of the transformation of environmental regulation towards resilience. It is mainly reflected in:

First, political stability ensures the continuity of regulation. Unstable political power can easily lead to repeated revisions of mining regulation and other policies, increasing the policy uncertainty faced by mining enterprises and suppressing their willingness to make long-term green investments involving huge sunk costs, such as clean technology R&D and the layout of recycling facilities (Gao et al., 2025). On the contrary, high regime stability ensures the long-term continuity of environmental regulation through legislative checks and balances and judicial independence, conveys stable and positive return expectations to enterprises, and thereby stimulates their continuous motivation for resilience building (Sovacool, 2019). Second, political stability enhances the capacity for administrative enforcement. Unstable political power often leads to lax grassroots governance and disorder in the bureaucratic system, which results in environmental standards being undermined by rent-seeking or protectionism at the local level, falling into “paper compliance” and failing to substantially enhance the environmental safety of mine operations. The stable state of the government ensures the strict implementation of technical standards through a vertical environmental supervision system and professional law enforcement teams, thereby reducing environmental accidents from the root (Handoyo, 2024). Third, political stability builds institutionalized channels for resolving conflicts. The environmental externalities of cobalt mining development are highly likely to trigger community protests. In an unstable political situation, the government tends to suppress conflicts through violence, leading to large-scale production halts and even supply chain disruptions. High political stability enables the state to have the ability to bring social conflicts onto the track of institutionalized resolution through legal mechanisms such as environmental courts, community hearings and universal compensation policies, and maintain the social permission for mining activities (Braunstein and Chuchko, 2025).

In conclusion, the political stability transforms environmental regulation from simple textual provisions into sustainable action frameworks by enhancing the credibility of regulation, improving the rigidity of enforcement, and optimizing conflict buffering. This not only reduces the operating costs of the system, but also ensures that the resilience dividends contained in environmental regulation can be transformed from theoretical potential into practical effectiveness, thereby amplifying the positive impact of environmental regulation on the cobalt supply chain resilience. Therefore, it is proposed that:

Hypothesis 5

Hypothesis 5. The political stability has a regulatory effect: the higher the degree of political stability, the stronger the positive impact of environmental regulation on the cobalt supply chain resilience.

Figure 2

Figure 2. Theoretical framework of this study.

4 Research strategy

4.1 Model specification

To explore the impact of environmental regulation on the cobalt supply chain resilience, referring to relevant research, this paper constructs a fixed effects regression model using panel data from 22 countries (regions) that hold key positions in the upstream and downstream of the cobalt industry chain from 2003 to 2024. The specific setting of this model is as shown in Equation 1.

${CSCR}_{i,t}=a_0+{\beta }_0{ER}_{i,t}+{\gamma }_0{X}_{i,t}+{\mu }_i+v_t+{\epsilon }_{i,t}(1)$

${\text{CSCR}}_{\mathrm{i},\mathrm{t}}$ represents the cobalt supply chain resilience level of the country (region) i at year t. ${\text{ER}}_{\mathrm{i},\mathrm{t}}$ represents the intensity of environmental regulation of the country (region) i at year t. ${\mathrm{X}}_{\mathrm{i},\mathrm{t}}$ is the set of control variables. ${\mathrm{\mu }}_{\mathrm{i}}$ represents the country (region) fixed effect. ${\mathrm{v}}_{\mathrm{t}}$ represents the fixed effect of the year. ${\mathrm{\epsilon }}_{\mathrm{i},\mathrm{t}}$ is a random disturbance term. ${\mathrm{a}}_0$, ${\mathrm{\beta }}_0$, and ${\mathrm{\gamma }}_0$ are the parameters to be estimated.

To examine the impact mechanism and transmission path of environmental regulation on the cobalt supply chain resilience, this paper introduces mediating variables based on benchmark regression and constructs a mediating effect model. The specific setting is as shown in Equations 2, 3.

$M_{i,t}={\alpha }_1+{\beta }_1{ER}_{i,t}+{\gamma }_1{X}_{i,t}+{\mu }_i+v_t+{\epsilon }_{i,t}(2)$

${CSCR}_{i,t}={\alpha }_2+{\beta }_2{ER}_{i,t}+{\theta }_2{M}_{i,t}+{\gamma }_2{X}_{i,t}+{\mu }_i+v_t+{\epsilon }_{i,t}(3)$

${\mathrm{M}}_{\mathrm{i},\mathrm{t}}$ is the mediating variable, including the level of green technological innovation ${\text{GTI}}_{\mathrm{i},\mathrm{t}}$, the concentration of import sources ${\mathrm{I}\text{SC}}_{\mathrm{i},\mathrm{t}}$, and the net inflow of foreign direct investment ${\text{FD}\mathrm{I}}_{\mathrm{i},\mathrm{t}}$ of the country (region) i at year t. The meanings of other variables are the same as those in model (1).

In addition, to examine whether political stability changes the impact intensity of environmental regulation on enhancing the cobalt supply chain resilience, this paper further introduces moderating variables and constructs a moderating effect model. To address the multicollinearity issues caused by interaction terms, the moderator and core explanatory variables were mean-centered. The setting is as shown in Equation 4.

${CSCR}_{i,t}={\alpha }_3+{\beta }_3{ER}_{i,t}+{\delta }_3{PS}_{i,t}+{\varphi }_3\left({ER}_{i,t}\times {PS}_{i,t}\right)+{\gamma }_3{X}_{i,t}+{\mu }_i+v_t+{\epsilon }_{i,t}(4)$

${\text{PS}}_{\mathrm{i},\mathrm{t}}$ is the moderating variable, representing the political stability index of the country (region) i at year t. The meanings of other variables are the same as those in model (1).

4.2 Variables definition

4.2.1 Dependent variable

4.2.1.1 Cobalt supply chain resilience (CSCR)

Liu et al. (2025) recently established an assessment system for the China’s cobalt supply chain comprehensive resilience, promoting the scientific application of index methods. Based on the driving force - Pressure - State - Impact - Response (DPSIR) model that emphasizes the interaction between human activities and the environment (Du et al., 2020) and the full life cycle theory that systematically considers each link of the cobalt supply chain, this paper selects 30 indicators, involving the inherent socio-economic, supply and demand, trade and other related factors of the supply chain, as shown in Table 1. To further quantify the resilience level, this study first uses the IEM to determine the weights of various indicators, and then uses the TOPSIS model to measure and rank the comprehensive resilience of the cobalt supply chain. The combination of IEM and TOPSIS has dual advantages: On the one hand, IEM can effectively reflect the differences in information entropy of different indicators while reducing the bias of subjective weighting by humans, thereby enhancing the objectivity and rationality of weight distribution. On the other hand, TOPSIS visually presents the relative proximity of each evaluated object in terms of the comprehensive resilience level by constructing ideal solutions and negative ideal solutions, thereby achieving a comprehensive ranking of multiple indicators. The workflow of IEM-TOPSIS is shown in Figure 3.

Table 1

Table 1. Cobalt supply chain resilience assessment system.

Figure 3

Figure 3. The workflow of IEM-TOPSIS.

4.2.2 Core explanatory variable

4.2.2.1 Environmental regulation (ER)

There are differing methods to measure the intensity of environmental regulation in Chinese research, such as the frequency of environmental protection terms in government work reports, the number of administrative penalty cases and environmental complaints, income from pollution discharge fees, and investments in pollution control, as well as multi-dimensional indices constructed using indicators like wastewater treatment rates, municipal waste treatment rates, and the recycling rates of general industrial solid waste. These methods reflect the uniqueness of China’s environmental regulation system, which emphasizes a highly administrative-driven, top-down policy deployment, while gradually introducing market mechanisms and placing importance on public supervision and social mobilization. However, due to differences in institutional backgrounds, the structure of environmental governance, and statistical standards, particularly in areas such as the distribution of power in environmental governance, the strength of policy enforcement, the degree of market-based mechanisms used, and the level of public participation, the traditional measurement methods based on China’s context are difficult to apply in cross-national comparisons.

Given that international research tends to select indicators with continuity and comparability, this paper adopts the proportion of environmental protection tax revenue to GDP as the proxy variable for the intensity of environmental regulation. The purpose of establishing an environmental protection tax is to internalize external environmental costs by taxing polluting behaviors and resource consumption, thereby guiding enterprises and consumers to adopt more environmentally friendly behaviors. Based on the availability of data, there are three reasons for selection: First, this indicator more directly reflects the intensity of pressure exerted by the government on environmental pollution and resource utilization through economic levers than command-based control measures (such as pollutant discharge permits), and captures the core idea of “pollution pricing”. The higher the proportion, the greater the economic incentives or penalties that a country’s government invests in environmental governance, and the higher the costs that enterprises need to bear to meet environmental protection requirements. Second, the current global environmental governance is trending towards adopting more efficient and flexible market-oriented approaches. As a typical market incentive tool, the environmental protection tax is an important part of this transformation. Therefore, it is more in line with the development trend of international environmental policies and also helps to deeply explore the unique role of market-based environmental regulation tools in enhancing the resilience of the global supply chain. Third, the collection mechanism and statistical scope of environmental protection tax have higher homogeneity and comparability among different countries. Comparing it with the national economic aggregate of each country can effectively eliminate the impact brought by the differences in national economic scale, thereby providing a relatively standardized measurement indicator. In recent years, this method has been widely applied in studies (Ha, 2024; Sackitey, 2023), verifying its effectiveness and reliability in the analysis of cross-border environmental policies.

4.2.3 Mechanism variables

4.2.3.1 Green technology innovation level (GTI)

To measure a country’s (region’s) green technological innovation capacity related to the cobalt industry, this paper constructs an interactive indicator that combines overall innovation capacity with industry specialization degree, as shown in Equation 5.

$GT{I}_{i,t}=\mathit{ln}\left(GI{I}_{i,t}\right)\times z\left(CGP{S}_{i,t}\right)(5)$

First of all, the overall Innovation capacity of a country is measured by the Global Innovation Index (GII). Due to the certain right bias of the sample distribution of GII, the marginal contribution of the score fluctuations of the high-scoring groups (mostly developed economies) to the improvement of innovation capacity is relatively low. In this paper, the natural logarithm of GII is taken to introduce the marginal diminishing feature and improve its distribution characteristics. Second, the degree of industry specialization is measured by the Cobalt Green Patent Share (CGPS), defined as the proportion of international patents within the scope of the IPC Green Inventory related to cobalt mining, smelting, recycling, and alternative materials, relative to the total number of green patents during the same period in that country (region). To eliminate the impact of different national economic scales on patent output, Z-Score processing is carried out for this proportion.

4.2.3.2 Import sources concentration (ISC)

To measure a country’s (region’s) dependence on cobalt imports, this paper constructs an import-weighted supply chain Herfindahl-Hirschman Index (HHI). The specific approach is as follows: First, obtain the annual trade data of cobalt-related products of various countries from the United Nations Commodity Trade Database (UN Comtrade). The product range covers the main links of the cobalt industry chain, including the upstream, midstream and downstream (see Table 2). Second, based on the Cobalt Content Coefficient (CCC) of different sub-products, the imported weight of all kinds of products is uniformly converted into cobalt equivalent (tons of cobalt). On this basis, ISC calculates through HHI, as shown in Equation 6.

$IS{C}_{i,t}={\displaystyle \sum _{c=1}^n}{\left(\frac{{\displaystyle {\sum }_{p=1}^P}{W}_{c,i,t,p}\times {CCC}_p}{{\displaystyle {\sum }_{k=1}^N}{\displaystyle {\sum }_{p=1}^P}{W}_{k,i,t,p}\times {CCC}_p}\right)}^2(6)$

Table 2

Table 2. Selection of products from cobalt industry chain.

$\mathrm{c}$ represents a specific source country of imports. $\mathrm{N}$ is the total number of source countries of imports in that year, p represents a certain category of cobalt-related products. $\mathrm{P}$ is the number of types of cobalt-related products imported in that year. $\mathrm{k}$ is the traversal index. ${\mathrm{W}}_{\mathrm{c},\mathrm{i},\mathrm{t},\mathrm{p}}$ is the weight of products $\mathrm{p}$ imported by the country i from the source country $\mathrm{c}$ at year t. $\text{CC}{\mathrm{C}}_{\mathrm{p}}$ is the cobalt content coefficient of product $\mathrm{p}$. The numerator of Equation 6 represents the proportion of the imported cobalt equivalent of a certain source country c to the total imported cobalt equivalent of that country i, and the denominator is the total imported cobalt equivalent of that country i from all source countries at year t. Finally, ISC is obtained by summing the squares of the import proportions of each source country, with a value range of (0,1]. The larger the value, the higher the degree of cobalt import dependence on a few countries.

4.2.3.3 Foreign direct investment (FDI)

FDI is an important indicator for measuring international capital flows and the participation of multinational enterprises in a country’s production and supply chain layout. Net inflow of investment refers to the total amount of capital flowing in by foreign investors to obtain lasting management benefits in the host country, including equity capital, reinvestment of profits and other capital flows related to investment, and deducting the net outflow caused by foreign investment withdrawal and domestic foreign direct investment. Given the significant differences in economic volume and currency levels among countries (regions) in the cross-border sample, directly comparing the scale of FDI may lead to large economies having an absolute advantage in indicators, and it cannot objectively reflect the relative role of international capital on supply chain resilience. For this purpose, this paper uses net FDI inflow data from each country (region), which is then standardized using Z-Score to avoid the influence of dimensional differences in subsequent regression analysis.

4.2.4 Moderating variable

4.2.4.1 Political stability (PS)

PS not only directly affects a country’s (region’s) economic operation and the cross-border investment environment, but also may change the intensity of the effect of environmental regulation on the cobalt supply chain resilience. We use the Political Stability and Absence of Violence/Terrorism index from the Worldwide Governance Indicators (WGI) released by the World Bank to measure the level of PS. It reflects the possibility of a country’s (region’s) government being overthrown by non-constitutional means, as well as the frequency and risk of suffering from political violence and terrorism. The higher the PS value, the more stable the political environment and the lower the risk of violent conflicts and terrorist activities.

4.2.5 Control variables

This paper introduces a series of control variables to reduce the possibility of missing variable bias: (1) Annual per capita gross domestic product (PGDP). The level of economic development is an important foundation for the capacity of supply chain construction and resource mobilization. (2) Annual total population (POP). The size of the labor force and the potential market capacity have a direct impact on the production capacity and demand stability of the supply chain. (3) Monetary freedom (MF). MF, drawn from the Economic Freedom Index released by the Heritage Foundation, reflects the degree of capital flow freedom and the financial environment. Higher monetary freedom can reduce cross-border transaction costs, facilitate flexible procurement and financing, and thus enhance supply chain adaptability to external shocks. (4) Government Intervention (GS). GS is measured by the annual index of government expenditure in the Economic Freedom Index. Government intervention in economic activities may affect the operational efficiency of supply chain through public investment, subsidies or regulation. (5) Trade Openness (TO). TO is measured as the ratio of the annual merchandise trade volume to GDP. The depth and breadth of participation in global trade affect the dependence on international markets and exposure to external shocks. We apply the natural logarithm transformation to the aforementioned five control variables. (6) Geopolitical Risk (Risk). Risk is obtained by calculating the annual average of the monthly global geopolitical risk index. The intensity of regional political conflicts and uncertain factors has a potential impact on the continuity and security of international supply chain.

4.3 Sample and data

The adoption of a cross-national sample is essential for analyzing cobalt supply chain resilience. Unlike conventional manufacturing industries that are largely embedded within national boundaries, the cobalt supply chain is characterized by a globally dispersed structure, in which upstream mining, midstream refining, and downstream consumption are located in different countries and connected through international trade and cross-border capital flows. In this context, supply chain resilience is inherently a system-level and transnational property, rather than an outcome that can be fully explained by a single country. Moreover, environmental regulation in one country may not only affect domestic production capacity but also reshape global supply chain structures through trade diversion, foreign direct investment reallocation, and multinational firms’ strategic sourcing decisions. These cross-border adjustment processes constitute a core channel through which environmental regulation influences supply chain resilience. Therefore, a cross-national perspective is not merely a matter of empirical convenience, but an analytical necessity dictated by the globalized nature of the cobalt supply chain and the transboundary transmission of environmental policy effects.

The sample selection in this article mainly follows the following considerations: (1) The sample countries must cover the main stages of the global cobalt supply chain, including resource extraction, smelting and processing, as well as downstream applications and consumption. (2) The sample countries must play a significant role in the import and export of cobalt ores, cobalt compounds, and cobalt-related products, with their total trade volume consistently accounting for a large portion of the global total during the study period. (3) The sample countries must have continuous data available from international authoritative databases or open platforms. Based on these principles, this paper ultimately selects 22 countries (regions) as research subjects: Australia, Belgium, Brazil, Canada, Switzerland, China, Germany, Finland, France, the United Kingdom, Hong Kong Special Administrative Region of China, Indonesia, India, Italy, Japan, South Korea, the Netherlands, Norway, the Philippines, Russia, the United States, and South Africa.

The indicator data involved in CSCR mainly come from USGS, UN Comtrade, and the official statistical yearbooks of various countries. The ER intensity data is derived from the World Development Indicators (WDI) released by the World Bank, the database of the Organization for Economic Cooperation and Development (OECD), and the fiscal and national economic accounting statistics of various countries. The GII data required by GTI comes from the annual report of the Global Innovation Index released by the World Intellectual Property Organization, and the cobalt-related green patent data required comes from the WIPO PATENTSCOPE database of the World Intellectual Property Organization. The cobalt trade data required by ISC comes from UN Comtrade. The PS data is from WGI. The FDI, PGDP, POP and TO data are all from WDI. The MF and GS data are taken from the annual report of the Index of Economic Freedom. The Risk data uses the Global geopolitical risk index compiled by Caldara and Iacoviello (2022).

The descriptive statistics of the main variables in this article are shown in Table 3. All variables have considerable variability within the sample, providing a solid data foundation for subsequent empirical tests. The VIF values of all variables are less than 10, and the average VIF value is 2.61 (less than 5), indicating that there is no multicollinearity problem in the model, and the variables are stable. The correlations among the explanatory variables are within an acceptable range.

Table 3

Table 3. Descriptive statistics.

5 Results and discussion

5.1 Baseline regression results

Table 4 reports the benchmark regression results of the impact of environmental regulation on the cobalt supply chain resilience. Columns (1) to (7) adopt the fixed effects model. Column (1) only includes the core explanatory variables. The results show that the estimated coefficient of ER is 0.049 and is significant at the 1% level, indicating that environmental regulation can significantly enhance the resilience of the cobalt supply chain. On this basis, control variables are gradually introduced in columns (2) to (7), and the main effect remain robust. The result in column (7) shows that when the intensity of environmental regulation increases by 1 unit, the resilience level of the cobalt supply chain will increase by 0.040, verifying the positive effect of environmental regulation on enhancing the resilience of the supply chain. In terms of control variables, the coefficients of variables such as PGDP, MF, GS and TO have all remained positive continuously, indicating that the improvement of the macroeconomic environment and the optimization of institutional factors can, to a certain extent, enhance the stability and adaptability of the cobalt supply chain. Furthermore, column (8) presents similar result under the random effects model: the impact coefficient of ER remains positive and passes the 1% significance test, further supporting the establishment of Hypothesis 1.

Table 4

Table 4. Baseline results.

5.2 Robustness test

5.2.1 Adjusting clustering method

The benchmark regression adopts the robust standard error of clustering at the national (regional) level. Although it controls the sequence correlation and heteroscedasticity within the same country (region), it fails to fully address the potential correlation across countries (regions): the sample countries (regions) belong to different continents and may be simultaneously affected by common shocks or global events at the continental level. In addition, time-varying factors that have not been observed within a specific year may also cause the error term to be correlated at the time level. For this purpose, this section introduces the three-way clustering standard error of “country (region) - continent - year”, and the result is shown in column (1) of Table 5. The coefficient value, significance level and standard error size of the core explanatory variable ER are all consistent with the benchmark regression results. This indicates that the empirical findings of this paper on the impact of environmental regulation do not stem from the correlation of error terms in specific dimensions, and the research conclusions have strong credibility.

Table 5

Table 5. Robustness results.

5.2.2 Replacing explanatory variable

Given that the core goal of environmental regulation is to reduce pollution emissions, and the electricity industry, as a major source of greenhouse gas emissions, is a key target for environmental regulation, coupled with the complementary nature of physical emission indicators and economic incentive-based indicators in their measurement dimensions, we select the methane emissions (in million tons of CO₂ equivalent) from the electricity sector in each country (region) as a new proxy variable for ER. Theoretically, stricter environmental regulation should be more effective in curbing methane emissions from the energy sector, and the two are in an inverse relationship. The regression result in column (2) of Table 5 show that the coefficient of the new proxy variable ER is −0.007 and is statistically significant at the 1% level, indicating that the reduction in methane emissions (representing an increase in the actual intensity of environmental regulation) has significantly promoted the resilience of the cobalt supply chain, which is in line with theoretical expectations. Therefore, the main conclusion of this paper regarding the impact of environmental regulation on the cobalt supply chain resilience is robust and does not rely on specific environmental regulation measurement methods.

5.2.3 Replacing dependent variable

According to relevant research on trade network resilience (Van den Brink et al., 2022; Yu et al., 2023), higher network efficiency usually means shorter average paths and smoother resource flows in the trade network, enabling the supply chain to respond more flexibly to external shocks. When some nodes or trade channels are damaged, an efficient trade network can promptly adjust trade routes, ensure the connectivity of the supply chain, and enhance the network’s adaptability to external shocks. Therefore, the resilience of the cobalt trade network can be an alternative indicator of CSCR and is measured by the local trading network efficiency (NE) centered on each key node, as shown in Equation 7.

$NE=\frac{1}{m_i}\sum _{j,h\in P_{N\left(i\right)}^2}\frac{1}{d_{jh}^w}(7)$

${\mathrm{m}}_{\mathrm{i}}$ represents the actual number of neighbor pairs existing in node i. ${\mathrm{P}}_{\mathrm{N}\left(\mathrm{i}\right)}^2$ is the combination of all two nodes in the neighbor set of node i. ${\mathrm{d}}_{\text{jh}}^{\mathrm{w}}$ is the shortest weighted path length between neighboring nodes j and h, and j≠h. The result in column (3) of Table 5 shows that the coefficient of ER to the new proxy variable CSCR is 0.188, which is statistically significant at the 5% level, indicating that environmental regulation can promote the cobalt trade network resilience. Therefore, the main conclusions of this paper do not rely on specific resilience measurement methods either.

5.2.4 Changing sample range

This section adjusts the research sample along both spatial and temporal dimensions: First, at the country level, samples from resource-driven economies are excluded. Indonesia and Russia, as the second and third largest cobalt resource producers globally after DRC, may experience a “resource curse” effect due to their resource endowment advantages. This means that their abundant resource revenues might weaken policy enforcement, amplify short-term profit-driven production behaviors, and distort the true impact of environmental regulation on the resilience of the cobalt supply chain. Therefore, we exclude samples from these two countries. Second, at the time level, we exclude data from periods of major crisis impacts. The global financial crisis (2008-2009) and the COVID-19 pandemic (2020-2021) led to structural disruptions in the cobalt supply chain, which could obscure the long-term effects of environmental regulation under normal economic conditions. Thus, data from both periods are removed. The adjusted regression result is presented in column (4) of Table 5. The coefficient of ER is 0.037, slightly lower than the 0.040 in the benchmark regression, but the direction of the coefficient is consistent, statistical significance still holds, and the economic meaning remains stable. This indicates that after eliminating the potential interference of resource dependence and the transmission blockage caused by crisis periods, the positive effect of environmental regulation on the cobalt supply chain resilience still holds, confirming that the research findings are not driven by specific countries or extreme events.

5.2.5 Adding control variables

Finally, to further reduce the potential for omitted variable bias, this study introduces two additional control variables based on the original ones. The first is infrastructure quality (Infra), measured by the Logistics Performance Index (LPI) published by the World Bank. The LPI comprehensively reflects trade and transport infrastructure, international freight convenience, logistics service quality, cargo tracking capabilities, and delivery timeliness. As the material flow of cobalt resource involves a complex multi-country trade network, excellent infrastructure not only ensures the smooth flow of cobalt from mining areas to processing and final consumption markets, reducing transportation delay costs and improving cross-border efficiency, but also provides multiple transport routes and quick responses in the face of emergencies such as natural disasters or geopolitical conflicts, thereby supporting the continuous stability of the supply chain. The second variable is industrial structure (Indus), measured by the share of industrial added value in GDP, with data sourced from the World Bank. A higher share of industrial added value usually indicates a more mature domestic processing and manufacturing capacity, a more complete industrial support system, and more abundant technological reserves. Focusing on the cobalt supply chain, as a key raw material for downstream industries such as battery manufacturing, its resilience heavily depends on the industrial system’s ability to absorb upstream supply fluctuations. In the face of trade shocks such as resource export controls or drastic price fluctuations, a strong industrial base can ensure the continuity of downstream production by improving raw material utilization, activating recycling technologies for cobalt, seeking substitute materials, and relying on strategic reserves, thereby systematically enhancing the supply chain’s resilience to external risks. Column (5) in Table 5 shows that after introducing the additional control variables Infra and Indus, the coefficient of the core explanatory variable ER still maintains the same direction and statistical significance as in the baseline regression, with only slight changes in the coefficient value, further confirming the reliability of the research conclusion.

5.3 Endogeneity treatment

5.3.1 Instrumental variables method

To address an endogeneity issue caused by a bidirectional causal relationship, this paper selects the lagged one-period environmental regulation variable ${\text{ER}}_{\mathrm{i},\mathrm{t}-1}$ and water resource pressure level (the ratio of freshwater extraction to available freshwater resources) ${\text{WS}}_{\mathrm{i},\mathrm{t}}$ as instrumental variables, and estimates the model using the two-stage least squares (2SLS) method.

Conceptually, these instruments capture sources of variation in ER that are plausibly exogenous to short-term fluctuations in CSCR. ${\text{ER}}_{\mathrm{i},\mathrm{t}-1}$ reflects institutional inertia and policy continuity embedded in administrative and legislative processes, which evolve gradually over time and are unlikely to respond to contemporaneous changes in supply chain resilience. Similarly, ${\text{WS}}_{\mathrm{i},\mathrm{t}}$ influences governments’ environmental policy orientation primarily through ecological constraints, rather than mineral-specific supply chain dynamics. As such, both instruments generate variation in ER driven by institutional persistence and natural resource conditions, rather than by changes in CSCR itself.

From the perspective of relevance, environmental regulation exhibits strong policy inertia, and the environmental regulation intensity of the previous year serves as an important reference and predictor basis for the current year’s policy formulation. Meanwhile, the water resource pressure level reflects the supply and demand situation of freshwater resources. In countries (regions) with resource scarcity, governments tend to implement stricter environmental regulation to strengthen resource protection. Thus, both ${\text{ER}}_{\mathrm{i},\mathrm{t}-1}$ and ${\text{WS}}_{\mathrm{i},\mathrm{t}}$ are strongly positively correlated with ER. Regarding exogeneity, after controlling for the fixed effects of countries (regions) and years, ${\text{ER}}_{\mathrm{i},\mathrm{t}-1}$ does not directly affect the changes in supply chain resilience in the current period, but only functions through the environmental regulation channels in the current period. ${\text{WS}}_{\mathrm{i},\mathrm{t}}$ is mainly determined by long-term climatic conditions, natural endowments and industrial water use structures, and is relatively stable in the short term. Moreover, factors such as economic development levels and population sizes have been controlled in the model, and its direct impact path on the resilience level of the supply chain has been effectively blocked. Therefore, the instrumental variables proposed in this paper are theoretically reasonable.

Columns (1) and (3) of Table 6 respectively report the regression results of the instrumental variables ${\text{ER}}_{\mathrm{i},\mathrm{t}-1}$ and ${\text{WS}}_{\mathrm{i},\mathrm{t}}$ in the first stage. The coefficients of both are significant at the 1% level, and the F-statistic is greater than the empirical value of 10, indicating a strong correlation between the endogenous variables and the instrumental variables in this paper. Columns (2) and (4) report the results of the second stage. The coefficient of ER remains significantly positive, indicating that even when considering endogeneity issues, environmental regulation still significantly promotes the cobalt supply chain resilience. In terms of instrumental variable testing, the Kleibergen-Paap rk LM statistics are both significant at the 1% level, rejecting the null hypothesis of insufficient identification of instrumental variables. The Kleibergen–Paap rk Wald F statistics are 47.707 and 33.219 respectively, both greater than the critical value of 16.38 of Stock–Yogo at the maximum allowable error level of 10%. This further excludes the problem of weak instrumental variables, indicating that the recognition degree of the model is relatively high. Therefore, the instrumental variables are feasible and the benchmark findings are reliable.

Table 6

Table 6. Instrumental variables regression results.

5.3.2 Dynamic panel model

The baseline regression model does not account for the dynamic inertia of the dependent variable, meaning that the current level of cobalt supply chain resilience may be influenced by its historical level. To address this, this paper introduces the first-order lag of the dependent variable L.CSCR, and constructs a dynamic panel model, as specified in Equation 8. Due to the correlation between L.CSCR and the individual fixed effect, which causes the endogeneity problem of Nickell bias in the model, this paper adopts the Least Squares Dummy Variable Corrected (LSDVC) estimator. Compared to GMM-based methods, LSDVC performs better with smaller individual samples and larger time dimensions in long panel data (Shuaibu et al., 2022). Following relevant research practices, the paper uses 500 bootstrap samples to estimate the variance-covariance matrix and employs difference GMM Arellano-Bond and system GMM Blundell-Bond estimators as initial values with precision of O (22⁻¹ × 22⁻¹). Since the LSDVC estimator partially absorbs individual effects, this model controls only for time fixed effects. Table 7 reports the LSDVC estimation results, showing that the coefficient of the explanatory variable ER remains significantly positive, with a coefficient size close to that of the baseline static model. The coefficient of L.CSCR is not significant, indicating that the previous period’s supply chain resilience does not have a significant impact on current resilience fluctuations. This suggests that the bidirectional fixed effects model chosen in this study is superior. Additionally, the LSDVC-estimated coefficient of L.CSCR is larger than that of the fixed effects model but smaller than the mixed regression model’s estimate, indicating the reliability of the LSDVC results. Furthermore, an Arellano-Bond autocorrelation test was conducted, and the results showed significant negative first-order autocorrelation, with no significant second-order autocorrelation, supporting the reasonableness of the model specification. In summary, after controlling for the path dependence of cobalt supply chain resilience, environmental regulation still has a significant positive effect on resilience. This indicates that the resilience-strengthening effect of environmental regulation makes an independent incremental contribution, independent of the historical resilience levels.

$\mathit{CSC}{R}_{i,t}={\alpha }_4+{\rho }_4\mathit{CSC}{R}_{i,t-1}+{\beta }_{4}E{R}_{i,t}+{\gamma }_4{X}_{i,t}+v_t+{\epsilon }_{i,t}(8)$

Table 7

Table 7. LSDVC regression results.

${\mathrm{\rho }}_4$ describes the self-continuity of the resilience level, and ${\mathrm{\beta }}_4$ measures the net effect of environmental regulation on the cobalt supply chain resilience after considering dynamic inertia and time shock. The meanings of other variables are the same as model (1).

5.4 Impact channel analysis

5.4.1 Green innovation effect

According to the coefficient of ER in column (1) of Table 8, when the intensity of environmental regulation increases by one unit, the average level of green technological innovation increases by approximately 18.5%. In the cobalt industry, which is highly dependent on resources and has a heavy pollution load, technological progress can significantly promote its green transformation. The result of Column (2) further reveals that the coefficients of GTI and ER are both positive and at least reach a significance level of 5%, indicating that green technological innovation plays a positive mediating role. The mediating effect of GTI accounts for approximately 21.28% of the total effect, which is substantially larger than other mechanisms. This finding suggests that the primary resilience-enhancing role of environmental regulation operates through strengthening internal production capacity and technological adaptability, rather than relying solely on external adjustment mechanisms. The above results indicate that while countries are raising environmental protection requirements, they will also invest more resources in supporting green research and development, improving production processes, and introducing clean technologies. This not only reduces the environmental pressure on the industrial production but also lessens the capacity fluctuations caused by resource and pollution restrictions. Thus, ER provides technical support for enhancing CSCR. Hypothesis 2 is proved.

Table 8

Table 8. Mechanism test results.

5.4.2 Import diversification effect

Meanwhile, environmental regulation will also reshape the international resource allocation pattern. In column (3) of Table 8, the coefficient of ER is significantly negative, indicating that stricter environmental regulation is conducive to reducing the concentration of cobalt resource imports. In Column (4), the coefficient of ISC is significantly negative, while the coefficient of ER remains significantly positive, indicating that the diversification of import sources plays a negative partial mediating role between environmental regulation and the cobalt supply chain resilience, with the mediating effect accounting for 12.62%. In other words, strict environmental regulation prompts countries to adjust their resource acquisition structure in the international procurement system, reducing their reliance on a single supplier that does not meet environmental protection standards or has a high environmental risk, while expanding alternative supply sources that meet the requirements of sustainable development. This diversified import pattern can effectively disperse the uncertainties brought about by policy changes, environmental shocks and geopolitical risks in specific supply countries, thereby enhancing the stability of the cross-border cobalt supply chain. Hypothesis 3 is proved.

5.4.3 FDI promotion effect

Environmental regulation not only affects domestic technological and trade structures, but also enhances the attractiveness to foreign investment. The results in column (5) of Table 8 show that the ER coefficient is significantly positive, indicating that stricter environmental regulation is conducive to increasing the scale of net foreign capital inflows. In Column (6), the coefficients of both FDI and ER are positive and significant at the 1% level, indicating that foreign direct investment plays a positive partial mediating role between environmental regulation and the cobalt supply chain resilience, with the mediating effect accounting for 11.11%. Environmental regulation not only implys a comprehensive upgrade in emission standards and governance requirements, but also reflect the standardization, transparency and predictability of the system, thereby enhancing the confidence of international capital in investing in the country’s cobalt industry. The foreign capital flowing in is usually accompanied by advanced technology, management experience and a global market synergy network, forming effects of technology spillover, management optimization and market connection. As a result, the inflow of foreign capital has indirectly enhanced the overall competitiveness of the supply chain in the context of global uncertainty. Hypothesis 4 is proved.

5.4.4 Moderating effect analysis

The results of the moderating effect test are reported in Table 9. In column (1), the regression coefficient of the interaction term ER × PS is 0.023, which is significant at the 1% level. This result indicates that political stability has enhanced the promoting effect of environmental regulation on the cobalt supply chain resilience, exerting a positive moderating effect, which is intuitively demonstrated in Figure 4. Hypothesis 5 is thus verified. This finding aligns with the theoretical mechanism that political stability ensures regulatory continuity, strengthens administrative enforcement, and institutionalizes conflict resolution. A stable regime reduces the risks of arbitrary policy changes, thereby bolstering corporate confidence to undertake long-term, high-sunk-cost investments in green innovation and supply chain reconfiguration. Furthermore, it enhances the credibility and enforcement of environmental regulation, turning them from theoretical guidelines into concrete actions that directly reduce operational uncertainties. To further examine whether the political stability will affect the intensity of the mediating effect, we construct the following moderated mediating effect model, as shown in Equation 9.

$M_{i,t}={\alpha }_4+{\beta }_4{ER}_{i,t}+{\delta }_4{PS}_{i,t}+{\varphi }_4\left({ER}_{i,t}\times {PS}_{i,t}\right)+{\gamma }_4{X}_{i,t}+{\mu }_i+v_t+{\epsilon }_{i,t}(9)$

Table 9

Table 9. Moderating effect test results.

Figure 4

Figure 4. The moderating effect diagram of political stability.

According to columns (2) to (4) of Table 9, the coefficients of the interaction term ER × PS are all significant: for GTI, the interaction coefficient is 0.045, indicating that high PS enhances the innovation-driven effect of ER. For ISC, the interaction coefficient is −0.042, indicating that high PS deepens the supply dispersion effect of ER. For FDI, the interaction coefficient is 0.028, indicating that a high PS enhances the foreign investment attraction effect of ER. These findings confirm that political stability positively moderates the first-stage path of the mediating effect, consistent with the total direct moderation in column (1).

5.5 Further discussion: heterogeneity analysis

Given the differences among economies of various scales in terms of industrial structure, market dynamics and policy implementation capabilities, the impact of environmental regulation on the cobalt supply chain resilience may not be uniform. The top 12 countries (regions) in terms of economic aggregate in the sample are classified as large economies, while the remaining 10 are classified as medium and small economies. The results of the grouped regression are reported in columns (1) and (2) of Table 10. In the sample of large economies, the role of environmental regulation in enhancing the resilience of the cobalt supply chain is not clear. In contrast, it is very significant in the sample of small and medium-sized economies. This difference may stem from: First, in large economies, the cobalt supply chain is merely one link in their vast and complex industrial networks. The formulation and implementation of environmental regulation are often macroscopic and universal, and impact may be dispersed across various industries, leading to the dilution of the direct effect on the supply chain of a specific cobalt industry and thus being statistically insignificant. Second, the industrial structure of small and medium-sized economies is relatively more concentrated, and the economies of some countries even have a high degree of dependence on a few key minerals. Therefore, targeted environmental regulatory policy can be more directly and efficiently transmitted to each node enterprise in the cobalt supply chain, quickly guiding them to carry out technological upgrades, optimize procurement strategies and strengthen risk management, thereby significantly enhancing the resilience of the entire chain. Moreover, for small and medium-sized economies, strict environmental regulation may serve as a reverse push mechanism, encouraging local related enterprises to seek new competitive advantages through green technological innovation and management model optimization. In large economies, enterprises have more options and may avoid regulatory pressure through industrial transfer or alternative procurement, rather than directly building internal resilience.

Table 10

Table 10. Heterogeneity results.

The global supply chain of cobalt is a highly specialized chain, from upstream mining, to midstream hydrometallurgy and refining, and then to downstream material manufacturing and terminal application. The industrial characteristics, technical barriers and environmental pressures faced by each link are completely different. Therefore, the impact of environmental regulation on countries (regions) with different supply chain locations may also vary. This article, based on the core roles of each country (region) in the global cobalt supply chain, The samples were divided into three groups: upstream resource countries (Australia, Brazil, Canada, Indonesia, the Philippines, Russia, South Africa), midstream refining countries (Belgium, China, Finland, the Netherlands, Norway), and downstream consuming countries (Switzerland, Germany, France, the United Kingdom, the Hong Kong Special Administrative Region of China, India, Italy, Japan, South Korea, the United States). The results of the grouped regression are reported in columns (3) to (5) of Table 10. The coefficient of ER in the upstream samples is not significant, but significant in the upstream and midstream samples. Moreover, the coefficient in the downstream sample is relatively the largest. These findings indicate that the resilience enhancement effect of environmental regulation shows a significant strengthening trend along the supply chain from upstream to downstream. Possible explanations are as follows: First, for the upstream mineral mining industry, environmental regulation often directly translates into high compliance costs and operational burdens. These cost pressures may increase the uncertainty of mining investment and even lead to the exit of some marginal production capacity, thereby offsetting the operational stability brought about by the improvement of environmental management to a certain extent. Second, the midstream hydrometallurgical process is a high-energy-consuming and high-polluting link and is also a key regulatory target for environmental regulation. Strict environmental protection standards can effectively eliminate enterprises with backward processes and serious pollution, and force the industry to carry out technological upgrading and green transformation. This mechanism of survival of the fittest has enhanced the industrial concentration of the entire refining process, thereby significantly strengthening the supply chain resilience. Third, the downstream consuming countries (regions) are primarily hubs of advanced manufacturing and innovation in cobalt-dependent industries such as electric vehicle batteries, electronics, and renewable energy storage. These nations often impose stringent standards on imported materials and components, creating a ripple effect that incentivizes global suppliers to align with higher sustainability benchmarks. The downstream players can leverage their market power to enforce upstream improvements, turning environmental regulation into a strategic advantage that propagates backward through the chain. In essence, this downstream amplification underscores how green policy integration at the consumption end not only mitigates risks but also drives systemic sustainability, yielding long-term competitive edges in the global transition to a low-carbon economy.

6 Conclusions and recommendations

6.1 Conclusion

This study empirically analyzes the impact of environmental regulation on cobalt supply chain resilience and its mechanisms using cross-national panel data. The main conclusions are as follows: First, environmental regulation can significantly enhance cobalt supply chain resilience. This conclusion remains valid after a series of robustness tests and addressing endogeneity concerns, confirming that environmental policy is not only a tool for environmental protection but also an effective means to strengthen the security of critical mineral resources. Second, mechanism analysis reveals three primary pathways through which environmental regulation operates: by driving green technological innovation to enhance internal production capacity, by reshaping trade flows to diversify import sources and by attracting high-quality foreign direct investment to inject capital and technological vitality. These three pathways collectively form a multi-dimensional transmission network through which environmental regulation enhances supply chain resilience. Third, political stability plays a positive moderating role. A stable institutional environment can amplify the resilience dividend of environmental regulation, indicating that “good governance” is the cornerstone for the effective implementation of environmental policies. Finally, heterogeneity analysis shows that the effect of environmental regulation varies with economic scale and supply chain positioning, with a more pronounced promoting effect on small and medium-sized economies and mid-to downstream countries (regions). This reveals the context-dependent nature of policy effects and the chain-like characteristics of their transmission.

6.2 Policy implications

Based on the robust findings of this study, we propose the following multi-faceted policy recommendations from a global perspective, aimed at translating environmental regulation into tangible supply chain resilience across both short-term risk management and long-term structural transformation.

First, it is necessary to strategically incorporate environmental supervision into the mineral security framework. On the one hand, policymakers in major consuming countries (such as the European Union and the United States) should move beyond viewing environmental standards as a cost burden. On the contrary, environmental regulation such as carbon border adjustment mechanisms and battery passports should be actively designed and implemented as strategic tools to promote a more resilient and sustainable cobalt supply. In addition to their long-term incentive effects, these regulatory instruments can also enhance supply chain transparency and traceability, facilitating timely identification of potential risks and coordination among supply chain actors during periods of disruption. These policies create a “green pull” effect, incentivizing global suppliers to upgrade their ESG performance to maintain market access. On the other hand, international organizations such as OECD and IEA can develop a harmonized set of core environmental and due diligence standards for cobalt extraction and processing, thereby reducing the complexity and cost of adhering to multiple fragmented standards, lowering transaction barriers, promoting the diversification of import sources, and enhancing the efficiency of the green supply chain.

Second, it is necessary to foster collaborative and diversified green supply chain networks. The international community, led by developed economies and multilateral development banks, should provide financial and technical assistance to build environmental governance capacity in key mining countries, such as DRC, Australia, and Indonesia. This includes supporting the implementation of best available techniques (BAT) in mining and refining, establishing robust environmental monitoring systems, and formalizing artisanal mining sectors. This transforms resource endowment from a risk point into a pillar of sustainable supply. At the same time, complementary measures such as strategic stockpiling and coordinated information-sharing mechanisms can be considered to buffer short-term supply shocks while structural adjustments are underway. Moreover, governments should create a stable and transparent policy environment to attract high-quality FDI. This can be achieved through bilateral investment treaties that emphasize environmental protection and by offering “green channels” for projects focused on advanced recycling technologies, cobalt-free battery alternatives, and clean hydrometallurgy. This directly leverages the FDI promotion effect to enhance resilience.

Third, it is necessary to strengthen institutional foundations to maximize regulatory effectiveness. One the one hand, the positive moderating role of political stability is a critical insight. Governments, particularly in resource-rich upstream countries, must prioritize building stable and predictable institutional frameworks. This involves ensuring the continuity of environmental policies beyond electoral cycles, strengthening the rule of law, and combating corruption in the enforcement of regulation. This stability is paramount for giving companies the confidence to make long-term, high-sunk-cost investments in green innovation and supply chain reconfiguration. One the other hand, given that community conflicts are an important source of supply disruption, governments in both producing and consuming countries should support the establishment of transparent grievance redressal mechanisms and community engagement protocols. This enhances the social license to operate for mining and refining activities, directly reducing the risk of operational halts and strengthening the social dimension of supply chain resilience.

Fourth, it is necessary to implement differentiated and targeted policy approaches. For small and medium-sized economies and downstream hubs (such as South Korea, Japan, and several EU countries), actors can leverage their agility and market focus, policies should actively encourage niche leadership in green technologies or in building strategic reserves of refined cobalt, leveraging the strong positive effects of environmental regulation identified for them. For large economies and upstream/midstream players, policies should focus on mitigating the potential short-term cost pressures. For upstream nations, this means designing environmental regulation that are stringent yet predictable, coupled with fiscal support for the adoption of cleaner technologies to maintain operational stability. For midstream refining countries (such as China, Finland, and Belgium), policy should enforce strict environmental standards to drive industry consolidation and technological upgrading, turning regulatory pressure into a competitive advantage for leading enterprises.

6.3 Limitation and future research

Although this study has achieved meaningful findings, there are still several limitations. First, due to the limitation of data availability, the sample only covers 22 major countries (regions) and excludes small resource countries (such as Zambia and Madagascar) which may underestimate ER’s impact on upstream CSCR. In many resource-rich countries, informal mining activities, weak regulatory enforcement, and governance gaps may further decouple formal environmental regulation from actual operational practices. In the future, more emerging resource countries or small economies can be attempted to be included in the analysis to test the universality of the conclusion.

Second, although internationally comparable indicators are adopted for the measurement of environmental regulation, there are differences in the composition and implementation intensity of environmental policy tools among countries (regions). Moreover, formal regulatory intensity may not fully reflect actual enforcement effectiveness, as monitoring capacity and compliance stringency vary substantially across institutional contexts. Future research can further distinguish the heterogeneous effects of different types of environmental regulation, such as command-control type (mandatory emission standards), market incentive type (pollution discharge rights trading), and voluntary participation type (environmental certification).

Finally, this study mainly focused on the macro mechanisms at the national level. In the future, it can delve into the micro level of enterprises, using enterprise-level data to explore how environmental regulation affect the specific behaviors of core enterprises in the supply chain, thereby revealing its mechanism of action at a more refined scale. In addition, with the in-depth development of the circular economy, future research can also focus on the role of environmental regulation in enhancing the efficiency of the cobalt recycling and reuse system.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author contributions

PC: Data curation, Resources, Formal Analysis, Conceptualization, Writing – original draft. WZ: Writing – review and editing, Software, Writing – original draft, Conceptualization, Data curation, Project administration, Methodology. LC: Writing – review and editing, Methodology, Visualization. YB: Supervision, Investigation, Validation, Writing – review and editing. HS: Validation, Supervision, Investigation, Writing – review and editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This research was funded by the National Natural Science Foundation of China (Grant Nos 71991482, 72304255, and 72404083), the National Social Science Foundation Project (Grant No. 24CGL100), and the Fundamental Research Funds for National Universities, China University of Geosciences (Grant No. 2025XLB159).

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Aisyah, N., Sasongko, N. A., Wahyono, Y., Anda, M., and Trench, A. (2024). Estimating a national critical mineral security index in Indonesia using analytical hierarchy process. Environ. Sustainability Indicators 24, 100510. doi:10.1016/j.indic.2024.100510

CrossRef Full Text | Google Scholar

Akdag, S., Yildirim, H., and Alola, A. A. (2025). Comparative benefits of environmental protection expenditures and environmental taxes in driving environmental quality of the European countries. Nat. Resour. Forum 49, 2188–2203. doi:10.1111/1477-8947.12464

CrossRef Full Text | Google Scholar

Ali, S. H., Giurco, D., Arndt, N., Nickless, E., Brown, G., Demetriades, A., et al. (2017). Mineral supply for sustainable development requires resource governance. Nature 543, 367–372. doi:10.1038/nature21359

PubMed Abstract | CrossRef Full Text | Google Scholar

Bahini, Y., Mushtaq, R., and Bahoo, S. (2024). Global energy transition: the vital role of cobalt in renewable energy. J. Clean. Prod. 470, 143306. doi:10.1016/j.jclepro.2024.143306

CrossRef Full Text | Google Scholar

Bahizire, G. M., Sun, H., and Chen, T. (2024). Exploring the role of managerial green commitment in enhancing sustainability in Congo's cobalt supply chain. J. Environ. Manag. 372, 122997. doi:10.1016/j.jenvman.2024.122997

PubMed Abstract | CrossRef Full Text | Google Scholar

Bonnet, T. (2025). Foreign direct investments and energy transition critical minerals. Resour. Policy 105, 105551. doi:10.1016/j.resourpol.2025.105551

CrossRef Full Text | Google Scholar

Botelho Junior, A. B., Stopic, S., Friedrich, B., Tenório, J. A. S., and Espinosa, D. C. R. (2021). Cobalt recovery from Li-ion battery recycling: a critical review. Metals 11, 1999. doi:10.3390/met11121999

CrossRef Full Text | Google Scholar

Braunstein, J., and Chuchko, M. (2025). Resource curse in the age of critical minerals: geopolitical forces and market maturity. Energy Res. and Soc. Sci. 127, 104247. doi:10.1016/j.erss.2025.104247

CrossRef Full Text | Google Scholar

Brusset, X., and Teller, C. (2017). Supply chain capabilities, risks, and resilience. Int. J. Prod. Econ. 184, 59–68. doi:10.1016/j.ijpe.2016.09.008

CrossRef Full Text | Google Scholar

Buchholz, P., Schumacher, A., and Al Barazi, S. (2022). Big data analyses for real-time tracking of risks in the mineral raw material markets: implications for improved supply chain risk management. Mineral. Econ. 35, 701–744. doi:10.1007/s13563-022-00337-z

CrossRef Full Text | Google Scholar

Caldara, D., and Iacoviello, M. (2022). Measuring geopolitical risk. Am. Economic Review 112, 1194–1225. doi:10.1257/aer.20191823

CrossRef Full Text | Google Scholar

Dominy, S. C., O’Connor, L., Parbhakar-Fox, A., Glass, H. J., and Purevgerel, S. (2018). Geometallurgy—A route to more resilient mine operations. Minerals 8, 560. doi:10.3390/min8120560

CrossRef Full Text | Google Scholar

Du, Y., Wang, W., Lu, Q., and Li, Z. (2020). A DPSIR-TODIM model security evaluation of China’s rare earth resources. Int. Journal Environmental Research Public Health 17, 7179. doi:10.3390/ijerph17197179

PubMed Abstract | CrossRef Full Text | Google Scholar

Earl, C., Shah, I. H., Cook, S., and Cheeseman, C. R. (2022). Environmental sustainability and supply resilience of cobalt. Sustainability 14, 4124. doi:10.3390/su14074124

CrossRef Full Text | Google Scholar

Franken, G., and Schütte, P. (2022). Current trends in addressing environmental and social risks in mining and mineral supply chains by regulatory and voluntary approaches. Mineral. Econ. 35, 653–671. doi:10.1007/s13563-022-00309-3

CrossRef Full Text | Google Scholar

Fu, X., Beatty, D. N., Gaustad, G. G., Ceder, G., Roth, R., Kirchain, R. E., et al. (2020). Perspectives on cobalt supply through 2030 in the face of changing demand. Environ. Sci. and Technol. 54, 2985–2993. doi:10.1021/acs.est.9b04975

PubMed Abstract | CrossRef Full Text | Google Scholar

Gao, D., Li, G., and Li, Y. (2022). Government cooperation, market integration, and energy efficiency in urban agglomerations—based on the quasi-natural experiment of the Yangtze River Delta Urban economic coordination committee. Energy and Environ. 33, 1679–1694. doi:10.1177/0958305X211047480

CrossRef Full Text | Google Scholar

Gao, D., Zhang, T., and Liu, X. (2025). The urban renewable energy transition: impact assessment and transmission mechanisms of climate Policy uncertainty. Energies 18, 2089. doi:10.3390/en18082089

CrossRef Full Text | Google Scholar

Gent, W. E., Busse, G. M., and House, K. Z. (2022). The predicted persistence of cobalt in lithium-ion batteries. Nat. Energy 7, 1132–1143. doi:10.1038/s41560-022-01129-z

CrossRef Full Text | Google Scholar

Gnutzmann, H., Kowalewski, O., and Śpiewanowski, P. (2020). Market structure and resilience: evidence from potash mine disasters. Am. J. Agric. Econ. 102, 911–933. doi:10.1093/ajae/aaz041

CrossRef Full Text | Google Scholar

Guo, Y., Li, Y., Liu, Y., and Zhang, H. (2023). The impact of geopolitical relations on the evolution of cobalt trade network from the perspective of industrial chain. Resour. Policy 85, 103778. doi:10.1016/j.resourpol.2023.103778

CrossRef Full Text | Google Scholar

Ha, L. T. (2024). The role of environmental tax in guiding global climate policies to mitigate climate changes in European region. Nat. Resour. Model. 37, e12412. doi:10.1111/nrm.12412

CrossRef Full Text | Google Scholar

Handoyo, S. (2024). Public governance and national environmental performance nexus: evidence from cross-country studies. Heliyon 10 (23), e40637. doi:10.1016/j.heliyon.2024.e40637

PubMed Abstract | CrossRef Full Text | Google Scholar

He, B., Ding, W., and Zhang, D. (2024). Does the digital economy matter for carbon emissions in China? Mechanism and path. Pol. J. Environ. Stud. 34, 193384–197135. doi:10.15244/pjoes/193384

CrossRef Full Text | Google Scholar

Hoarau, Q., and Lorang, E. (2022). An assessment of the European regulation on battery recycling for electric vehicles. Energy Policy 162, 112770. doi:10.1016/j.enpol.2021.112770

CrossRef Full Text | Google Scholar

IEA (2023). Global EV Outlook 2023. Available online at: https://www.iea.org/reports/global-ev-outlook-2023 (Accessed October 20, 2025).

Google Scholar

IEA (2024). Global critical minerals outlook 2024. Available online at: https://www.iea.org/reports/global-critical-minerals-outlook-2024 (Accessed October 20, 2025).

Google Scholar

Janikowska, O., and Kulczycka, J. (2021). Impact of minerals policy on sustainable development of mining sector—a comparative assessment of selected EU countries. Mineral. Econ. 34, 305–314. doi:10.1007/s13563-021-00248-5

CrossRef Full Text | Google Scholar

Jia, M., Lu, G., Yan, Y., and Nazir, S. (2024). Resilience through mineral resource development, oil, and natural resource efficiency: strengthening economies. Resour. Policy 91, 104942. doi:10.1016/j.resourpol.2024.104942

CrossRef Full Text | Google Scholar

Jiang, C., Liu, R., and Han, J. (2023). Does accountability audit of natural resource promote corporate environmental performance? An external supervision perspective. Environ. Dev. Sustain. 25, 9417–9438. doi:10.1007/s10668-022-02441-0

CrossRef Full Text | Google Scholar

Kinnunen, P., Obenaus-Emler, R., Raatikainen, J., Guignot, S., Guimerà, J., Ciroth, A., et al. (2021). Review of closed water loops with ore sorting and tailings valorisation for a more sustainable mining industry. J. Clean. Prod. 278, 123237. doi:10.1016/j.jclepro.2020.123237

CrossRef Full Text | Google Scholar

Kong, L., Yao, Y., and Xu, K. (2025). Can environmental regulation improve the industrial ecology efficiency? Evidence from China's environmental protection tax reform. J. Environ. Manag. 373, 123792. doi:10.1016/j.jenvman.2024.123792

PubMed Abstract | CrossRef Full Text | Google Scholar

Lèbre, É., Corder, G. D., and Golev, A. (2017). Sustainable practices in the management of mining waste: a focus on the mineral resource. Miner. Eng. 107, 34–42. doi:10.1016/j.mineng.2016.11.004

CrossRef Full Text | Google Scholar

Lin, L., Li, M., Hou, X., and Piprani, A. Z. (2024). The resource curse in least developed countries: the roles of foreign direct investment, energy efficiency, and electricity access. Resour. Policy 89, 104564. doi:10.1016/j.resourpol.2023.104564

CrossRef Full Text | Google Scholar

Liu, W., Li, X., Liu, C., Wang, M., and Liu, L. (2023). Resilience assessment of the cobalt supply chain in China under the impact of electric vehicles and geopolitical supply risks. Resour. Policy 80, 103183. doi:10.1016/j.resourpol.2022.103183

CrossRef Full Text | Google Scholar

Liu, W., Chen, L., Luo, F., Zhao, Y., Li, X., and Zeng, X. (2025). Novel assessment of China's cobalt supply chain resilience based on DPSIR model and machine learning. Resour. Conservation Recycl. 215, 108107. doi:10.1016/j.resconrec.2024.108107

CrossRef Full Text | Google Scholar

Lou, G., Wang, H., Tu, X., Lai, Z., Ma, H., and Ying, S. (2025). Resilience assessment of the electric vehicle lithium-ion battery supply chain under supply shortages of upstream mineral enterprises. Resour. Conservation Recycl. 215, 108161. doi:10.1016/j.resconrec.2025.108161

CrossRef Full Text | Google Scholar

Maisel, F., Neef, C., Marscheider-Weidemann, F., and Nissen, N. F. (2023). A forecast on future raw material demand and recycling potential of lithium-ion batteries in electric vehicles. Resour. Conservation Recycl. 192, 106920. doi:10.1016/j.resconrec.2023.106920

CrossRef Full Text | Google Scholar

Mancheri, N. A., Sprecher, B., Bailey, G., Ge, J., and Tukker, A. (2019). Effect of Chinese policies on rare earth supply chain resilience. Resour. Conservation Recycl. 142, 101–112. doi:10.1016/j.resconrec.2018.11.017

CrossRef Full Text | Google Scholar

Mancini, L., Eslava, N. A., Traverso, M., and Mathieux, F. (2021). Assessing impacts of responsible sourcing initiatives for cobalt: insights from a case study. Resour. Policy 71, 102015. doi:10.1016/j.resourpol.2021.102015

CrossRef Full Text | Google Scholar

Maybee, B., Lilford, E., and Hitch, M. (2023). Environmental, social and governance (ESG) risk, uncertainty, and the mining life cycle. Extr. Industries Soc. 14, 101244. doi:10.1016/j.exis.2023.101244

CrossRef Full Text | Google Scholar

McLellan, B. C., and Corder, G. D. (2013). Risk reduction through early assessment and integration of sustainability in design in the minerals industry. J. Clean. Prod. 53, 37–46. doi:10.1016/j.jclepro.2012.02.011

CrossRef Full Text | Google Scholar

Mohammadian, B., Pishdar, M., and Zarei Matin, H. (2026). Transition of mining industry toward sustainable value creation in developing countries. Int. J. Qual. and Reliab. Manag. 43 (1), 277–294. doi:10.1108/IJQRM-04-2024-0127

CrossRef Full Text | Google Scholar

Mpofu, F. Y. (2022). Green taxes in Africa: opportunities and challenges for environmental protection, sustainability, and the attainment of sustainable development goals. Sustainability 14, 10239. doi:10.3390/su141610239

CrossRef Full Text | Google Scholar

Nagy, G., and Szentesi, S. (2024). Navigating uncharted waters: black swan events and global supply chain resilience. Adv. Logist. Syst. Theory Pract. 18, 41–54. doi:10.32971/als.2024.005

CrossRef Full Text | Google Scholar

Nasir, M., Bakker, L., and van Meijl, T. (2023). Environmental management of coal mining areas in Indonesia: the complexity of supervision. Soc. and Nat. Resour. 36, 534–553. doi:10.1080/08941920.2023.2180818

CrossRef Full Text | Google Scholar

Patiño Douce, A. E. (2016). Metallic mineral resources in the twenty-first century: II. Constraints on future supply. Nat. Resour. Res. 25, 97–124. doi:10.1007/s11053-015-9265-0

CrossRef Full Text | Google Scholar

Peng, H., Shen, N., Ying, H., and Wang, Q. (2021). Can environmental regulation directly promote green innovation behavior? based on situation of industrial agglomeration. J. Clean. Prod. 314, 128044. doi:10.1016/j.jclepro.2021.128044

CrossRef Full Text | Google Scholar

Pettit, T. J., Fiksel, J., and Croxton, K. L. (2010). Ensuring supply chain resilience: development of a conceptual framework. J. Bus. Logist. 31, 1–21. doi:10.1002/j.2158-1592.2010.tb00125.x

CrossRef Full Text | Google Scholar

Pettit, T. J., Croxton, K. L., and Fiksel, J. (2019). The evolution of resilience in supply chain management: a retrospective on ensuring supply chain resilience. J. Bus. Logist. 40, 56–65. doi:10.1111/jbl.12202

CrossRef Full Text | Google Scholar

Pramusinta, A. I., Yulivan, I., and Jubaedah, J. (2025). Strategy to strengthen tax supervision and transparency in the nickel mining sector to maintain national economic resilience. Formosa J. Appl. Sci. 4 (7), 2137–2154. doi:10.55927/fjas.v4i7.241

CrossRef Full Text | Google Scholar

Rachidi, N. R., Nwaila, G. T., Zhang, S. E., Bourdeau, J. E., and Ghorbani, Y. (2021). Assessing cobalt supply sustainability through production forecasting and implications for green energy policies. Resour. Policy 74, 102423. doi:10.1016/j.resourpol.2021.102423

CrossRef Full Text | Google Scholar

Raze, R. (2023). Ecolabeling in the multinational mining industry: a method toward environmental sustainability. Vanderbilt J. Transnatl. Law 56, 247. Available online at: https://scholarship.law.vanderbilt.edu/vjtl/vol56/iss1/11 (Accessed October 20, 2025).

Google Scholar

Sackitey, G. L. (2023). Do environmental taxes affect energy consumption and energy intensity? An empirical analysis of OECD countries. Cogent Econ. and Finance 11, 2156094. doi:10.1080/23322039.2022.2156094

CrossRef Full Text | Google Scholar

Shi, H., Heng, J., Duan, H., Li, H., Chen, W., Wang, P., et al. (2025). Critical mineral constraints pressure energy transition and trade toward the paris agreement climate goals. Nat. Commun. 16, 4496. doi:10.1038/s41467-025-59741-y

PubMed Abstract | CrossRef Full Text | Google Scholar

Shuaibu, S., Badayi, S. A., and Musa, B. A. (2022). Analysis of financial development and foreign direct investment inflows nexus in ECOWAS countries: least squares dummy variable approach. Int. J. Finance and Econ. 27, 4599–4606. doi:10.1002/ijfe.2389

CrossRef Full Text | Google Scholar

Sinaga, P. T., Simatupang, T. M., and Basri, M. H. (2024). Enhancing supply chain resilience in the coal mining sector: a qualitative study. Pol. J. Manag. Stud. 30, 282–296. doi:10.17512/pjms.2024.30.2.17

CrossRef Full Text | Google Scholar

Sovacool, B. K. (2019). The precarious political economy of cobalt: balancing prosperity, poverty, and brutality in artisanal and industrial mining in the democratic Republic of the Congo. Extr. Industries Soc. 6, 915–939. doi:10.1016/j.exis.2019.05.018

CrossRef Full Text | Google Scholar

Sovacool, B. K. (2021). When subterranean slavery supports sustainability transitions? Power, patriarchy, and child labor in artisanal Congolese cobalt mining. Extr. Industries Soc. 8, 271–293. doi:10.1016/j.exis.2020.11.018

CrossRef Full Text | Google Scholar

Sovacool, B. K., Ali, S. H., Bazilian, M., Radley, B., Nemery, B., Okatz, J., et al. (2020). Sustainable minerals and metals for a low-carbon future. Science. 367, 30–33. doi:10.1126/science.aaz6003

PubMed Abstract | CrossRef Full Text | Google Scholar

Sun, L., and Hasi, M. (2024). Effects of mining sector FDI, environmental regulations, and economic complexity on mineral resource dependency in selected OECD countries. Resour. Policy 89, 104651. doi:10.1016/j.resourpol.2024.104651

CrossRef Full Text | Google Scholar

Sun, X., Shi, Q., and Hao, X. (2022). Supply crisis propagation in the global cobalt trade network. Resour. Conservation Recycl. 179, 106035. doi:10.1016/j.resconrec.2021.106035

CrossRef Full Text | Google Scholar

Sun, H., Meng, Z., Li, M., Yang, X., and Wang, X. (2023). Global supply sustainability assessment of critical metals for clean energy technology. Resour. Policy 85, 103994. doi:10.1016/j.resourpol.2023.103994

CrossRef Full Text | Google Scholar

Tan, J., and Keiding, J. K. (2024). Mapping the cobalt and lithium supply chains for e-mobility transition: significance of overseas investments and vertical integration in evaluating mineral supply risks. Resour. Conservation Recycl. 209, 107788. doi:10.1016/j.resconrec.2024.107788

CrossRef Full Text | Google Scholar

Tisserant, A., and Pauliuk, S. (2016). Matching global cobalt demand under different scenarios for co-production and mining attractiveness. J. Econ. Struct. 5, 4. doi:10.1186/s40008-016-0035-x

CrossRef Full Text | Google Scholar

Turner, J. M. (2022). The matter of a clean energy future. Science 376 (6600), 1361. doi:10.1126/science.add5094

CrossRef Full Text | Google Scholar

USGS (2022). Mineral commodity summaries. doi:10.3133/mcs2022

CrossRef Full Text | Google Scholar

Valero, A., Valero, A., Calvo, G., and Ortego, A. (2018). Material bottlenecks in the future development of green technologies. Renew. Sustain. Energy Rev. 93, 178–200. doi:10.1016/j.rser.2018.05.041

CrossRef Full Text | Google Scholar

Van den Brink, S., Kleijn, R., Sprecher, B., and Tukker, A. (2022). Identifying supply risks by mapping the cobalt supply chain. Resour. Conservation Recycl. 156, 104743. doi:10.1016/j.resconrec.2020.104743

CrossRef Full Text | Google Scholar

Vivoda, V., and Matthews, R. (2024). “Friend-shoring” as a panacea to Western critical mineral supply chain vulnerabilities. Mineral. Econ. 37, 463–476. doi:10.1007/s13563-023-00402-1

CrossRef Full Text | Google Scholar

Wang, X., Sun, H., Gu, L., Meng, Z., Yang, L., and Cheng, J. (2025a). The impact of the spread of risks in the upstream trade network of the international cobalt industry chain. Sustainability 17 (15), 6711. doi:10.3390/su17156711

CrossRef Full Text | Google Scholar

Wang, Z., He, B., and Fujita, T. (2025b). Does environmental regulation affect inward foreign direct investment? Evidence from China. Appl. Econ. Lett. 32 (19), 2877–2883. doi:10.1080/13504851.2024.2352164

CrossRef Full Text | Google Scholar

Xiang, Y., Li, X., Liu, W., Luo, F., and Wang, M. (2025). Enhancing chromium supply chain security through resilience strategies: decision support based on system dynamics simulations. J. Clean. Prod. 493, 144981. doi:10.1016/j.jclepro.2025.144981

CrossRef Full Text | Google Scholar

Yang, Q., Gao, D., Song, D., and Li, Y. (2021). Environmental regulation, pollution reduction and green innovation: the case of the Chinese water ecological civilization city pilot policy. Econ. Syst. 45 (4), 100911. doi:10.1016/j.ecosys.2021.100911

CrossRef Full Text | Google Scholar

Yin, K., Liu, L., Huang, C., and Xiao, Y. (2023). Can the transfer of polluting industries achieve a win–win situation for both the economy and the environment? Research based on the perspective of environmental regulation. Environ. Dev. Sustain. 25, 8903–8928. doi:10.1007/s10668-022-02384-6

CrossRef Full Text | Google Scholar

Yu, Y., Ma, D., and Zhu, W. (2023). Resilience assessment of international cobalt trade network. Resour. Policy 83, 103636. doi:10.1016/j.resourpol.2023.103636

CrossRef Full Text | Google Scholar

Zhang, J., Yang, Z., and He, B. (2023). Does digital infrastructure improve urban economic resilience? Evidence from the yangtze River economic belt in China. Sustainability 15 (19), 14289. doi:10.3390/su151914289

CrossRef Full Text | Google Scholar

Zhao, X., Mahendru, M., Ma, X., Rao, A., and Shang, Y. (2022). Impacts of environmental regulations on green economic growth in China: new guidelines regarding renewable energy and energy efficiency. Renew. Energy 187, 728–742. doi:10.1016/j.renene.2022.01.076

CrossRef Full Text | Google Scholar

Zhou, W., Song, Y., Xu, D., and Zhang, Y. (2025). Government subsidies and industrial chain resilience: evidence from Chinese resource-based enterprises. Resour. Policy 102, 105525. doi:10.1016/j.resourpol.2025.105525

CrossRef Full Text | Google Scholar