Policy Evolution of China’s Critical Metals: An Integrated Analysis of Instruments and Networks

Abstract

Critical metals constitute essential raw materials for clean energy transition, making their policy evolution highly significant for global resource governance. Analyzing policy texts from China (1973–2024), this study develops a three-dimensional analytical framework—Instrument Type, Policy Objective, and Implementation Domain—integrated with social network analysis to investigate the characteristics and drivers of policy evolution. Findings indicate that China’s critical metal governance paradigm has shifted from securing resource supply to pursuing sustainability goals. Policy instruments have transitioned from authority-based dominance to diversified combinations, while the policy network, centered on the Ministry of Industry and Information Technology (MIIT) and the National Development and Reform Commission (NDRC), exhibits increasingly frequent interdepartmental collaboration. The evolution is shown to stem from the dynamic interdependence between policy instruments and network structures. This research provides theoretical and practical insights for optimizing critical metals governance systems.

1. Introduction

With the ongoing advancement of global climate governance, emerging industries such as new energy and new materials have developed rapidly, leading to a surge in global demand for critical metals [1]. Due to their unique metallic properties, critical metals serve as fundamental raw materials for products like new energy batteries and vehicles, playing a pivotal role in the clean energy transition [2]. Consequently, critical metals have become a major focus of policy attention worldwide, drawing significant interest from national governments globally and researchers alike. Many countries are accelerating their energy transitions and industrial structure upgrades, indicating that global demand for critical metals will continue to grow in the future [3,4]. In recent years, scholars have conducted substantial research on relevant policies to better explore governance models for critical metals [5,6]. These studies encompass diverse aspects of critical metals policies, including supply security, pricing, taxation, finance, circular economy, and sustainable development [7,8,9]. The concept of “critical metals” intersects with those of “strategic minerals” and “critical minerals”, yet their meanings vary across policy documents of different countries. Generally, these terms refer to metallic mineral resources that are of significant importance to a nation’s industrial and technological development, yet face challenges in substitution or availability. It is worth noting that in the Chinese policy context, “strategic minerals”, “critical minerals”, “strategic metal minerals”, and “critical metals” are considered equivalent.

In recent years, the global resource market has intensified its focus on critical metals such as lithium, nickel, and cobalt, driven by rapidly growing demand stemming from their irreplaceable roles in energy transition, advanced manufacturing, and future industries [10]. Modern green and high-tech sectors—including electric vehicles, batteries, smart grids, wind power equipment, and data centers—heavily rely on critical metal supplies [11]. Consequently, governance capacity for these resources has become a key metric for assessing national sustainable development performance, profoundly reshaping global industrial patterns and sustainability pathways. From an industrial sustainability perspective, critical metals underpin multidimensional green transitions: lithium, cobalt, and nickel form core components of EVs and energy storage systems; copper enables electrification; rare earth elements are essential for wind turbine permanent magnets and high-efficiency motors; platinum group metals serve as critical catalysts for hydrogen electrolyzers and fuel cells [12]. Studies project long-term demand growth for critical metals. Economically, their supply-demand dynamics significantly influence green economic trajectories. Recycling technologies for metals like indium and gallium demonstrate high recovery rates, yielding substantial economic benefits while enabling circular economy practices and carbon reduction—highlighting dual environmental and economic value. Contemporary technological breakthroughs are reshaping utilization paradigms through green exploration, sustainable mining, material innovation, and recycling advances, collectively building a “mining–processing–recycling–substitution” sustainability framework. This positions critical metals at the confluence of industrial, economic, and technological sustainability, providing foundational material support for global sustainable development [13].

Prior research indicates that the global energy transition and industrial restructuring have imposed unprecedented challenges on the existing critical metals supply system [14]. Rational policy formulation and the optimization of the policy framework represent effective strategies for mitigating this risk [15]. Analyzing the evolutionary characteristics of critical metals policies contributes to a deeper understanding of their macro-level features and driving forces, thereby enabling more targeted recommendations for policy reform. Some studies suggest that the Chinese government is establishing a critical metals reserve system to enhance self-sufficiency [16]. Simultaneously, export controls have strengthened China’s bargaining power in critical metals trade [17]. Through long-term development, China has continuously refined relevant laws and policies, employing fiscal, tax, and other regulatory tools to manage production capacity. However, increasingly stringent environmental regulations have also exerted certain negative impacts on the sustainability of critical metals development [18,19]. Therefore, future policy orientation must achieve a relative balance between ecological environmental protection and critical metals exploitation.

Research on policy texts employs well-established methodological approaches in academia, primarily encompassing quantitative, qualitative, and mixed methods [20,21,22,23]. Within quantitative methods, content analysis achieves the quantitative presentation of textual features through systematic coding, while bibliometrics utilizes techniques like co-word analysis to reveal the evolution of research hotspots. Leveraging tools such as NVivo, both approaches enable efficient processing of large-scale textual data. Qualitative methods focus on the interpretation of meaning. Discourse analysis deconstructs power dynamics within texts, frame analysis identifies the underlying logic defining policy issues, and narrative analysis parses the storied structure of texts, providing deep insights for understanding the ideologies underpinning policies [24]. Mixed methods, integrating quantitative and qualitative approaches, maintain analytical systematicity while ensuring interpretive depth, constituting the dominant paradigm in current research. Technologically, the efficiency and precision of text mining are crucial for policy text research, while visualization enhances the intuitiveness of research findings [25]. However, existing methodologies exhibit limitations. Quantitative analysis, for instance, is susceptible to the “decontextualization dilemma”, and overreliance on AI tools for semantic understanding can introduce bias, potentially compromising the validity of conclusions. Furthermore, insufficient interdisciplinary integration constrains the real-world explanatory power of research findings. Future research must enhance the complementarity between methods and deepen cross-disciplinary synthesis.

Building on this foundation, this study employs an empirical analysis of critical metals policy texts issued in China from 1973 to 2024 to investigate the characteristics and underlying mechanisms of their evolution. Compared to existing literature, the novel contributions of this research are threefold: First, utilizing China’s critical metals policy texts as the entry point, this study establishes a three-dimensional policy instrument analysis framework based on instrument type, policy objective, and implementation domain. This framework is used to explore the evolutionary characteristics of China’s critical metals policy instruments, providing a novel perspective for understanding this policy evolution. Second, leveraging social network analysis (SNA), the research constructs a policy network analysis model. This model identifies core actors within the network and analyzes the intrinsic driving forces behind the co-evolution of instruments and networks, thereby providing empirical evidence for the mechanisms of policy evolution. Third, the findings offer new insights for the government to optimize and refine critical metals policies. They provide a scientific basis for enhancing governmental governance capacity over critical metals, contributing to economic and social sustainability.

This study focuses specifically on China for several compelling reasons. First, China is a leading global producer, consumer, and trader of critical metals, rendering its policy developments highly influential for global supply chain stability and the worldwide energy transition. Second, China’s governance of critical metals exemplifies a distinct “state-led, market-assisted” model, offering a significant case for examining the state’s role in resource governance. Furthermore, China’s critical metals policy system has undergone a decades-long evolution, progressively transforming from early, single-objective control mechanisms into a complex system characterized by multiple objectives, diverse policy instruments, and multi-agency collaboration. Analyzing this evolutionary pathway provides a valuable reference for other resource-based economies. Therefore, a deep dissection of the Chinese case not only elucidates its internal governance logic but also offers an empirical foundation for constructing robust global critical metals governance frameworks.

The subsequent content in this article is arranged as follows: The second part presents the literature review and theoretical analysis; the third part details the research methodology, including data sources, model specification, coding of analysis units, and related procedures; the fourth part presents the empirical results and discussion; the fifth part conducts an extended analysis; and the sixth part concludes the study and proposes targeted recommendations.

2. Literature Review and Theoretical Analysis

2.1. Institutional Complexity in Critical Metals Policy Systems

Critical metals are typically defined as metallic mineral resources indispensable to a nation’s economy, emerging industries, and high-tech sectors [26,27]. Their applications in clean energy technologies, power batteries, specialty materials, and other fields are irreplaceable, with lithium, copper, cobalt, nickel, and rare earth elements (REEs) serving as prominent examples [28]. The specific scope of critical metals exhibits variations across nations, contingent upon resource endowment, industrial structure, and technological development stage. Moreover, their application scenarios and defining boundaries have continuously expanded over time. The recognition of critical metals’ value has evolved considerably throughout history [29]. Since the 1970s, governments have begun prioritizing policy research on critical metals. In recent years, propelled by global climate governance and industrial upgrading, critical raw materials underpinning emerging industries have progressively been incorporated into the critical metals category [30]. The policy evolution in the United States offers a significant reference point. Its critical metals management system transitioned from an early mechanism focused on resource security during special periods towards a globalized model. Since 2010, the U.S. has centered its policy on enhancing supply chain resilience, strengthening critical metals governance capabilities through a series of legislative actions [31]. Following the release of the E.U. Raw Materials Initiative in 2008, the European Union has consistently updated its critical raw materials list and issued multiple reports and action plans, aiming to build a more stable end-to-end industrial chain [32]. Japan, implementing a strategic stockpiling system since the 1960s, has developed a resource security framework characterized by both foresight and operability through dynamic adjustments to its critical metals catalog. Scholarly consensus indicates that the policy landscape for critical metals has undergone significant transformation compared to the past. Governmental governance over these resources now employs more diverse and frequent policy instruments, while governance objectives have grown increasingly complex. Future policies are expected to place greater emphasis on enhancing the systematicity and sustainability of critical metals governance, positioning it as a key driver of global climate governance efforts.

Global climate governance and industrial restructuring present multidimensional challenges to critical metals policy. These challenges stem from both fundamental shifts in demand structures and paradigm shifts in governance logic. The clean energy transition (e.g., electric vehicles, energy storage systems) driven by climate governance, coupled with industrial upgrading (e.g., semiconductors, advanced manufacturing), has triggered exponential growth in demand for critical metals. Concurrently, climate governance mandates adherence to stringent Environmental, Social, and Governance (ESG) standards throughout the mineral life cycle, creating tension with the cost-control imperatives inherent in industrial upgrading. Technological innovation simultaneously creates new opportunities and introduces risks of path dependency. These challenges essentially represent a microcosm of the transition from industrial civilization to an ecological civilization [33]. Consequently, they demand that policymakers transcend traditional resource governance frameworks and construct a novel “Climate-Industry-Security” trinity governance paradigm.

Based on the above analysis, this study proposes Hypothesis 1.

H1. 

The evolution of critical metals policies exhibits progressively intensified multi-objective orientation and institutional complexity.

2.2. Evolution of Multidimensional Policy Instruments

Policy instruments, serving as critical vehicles for governments and governing entities to achieve public objectives, constitute a core bridge connecting policy objectives with governance outcomes. They exhibit distinctive relational attributes and functional diversity. Their relational nature manifests in deep integration with public policies and government programs, functioning as the essential linkage translating objectives into results [34]. For instance, to advance technological innovation goals, governments may deploy instruments such as R&D subsidies or tax credits. The diversity of instruments stems from the complexity of governance contexts, encompassing legal, economic, administrative, and cultural dimensions—with no universally applicable panacea existing across all scenarios. Scholars have classified policy instruments from diverse governance perspectives, with these taxonomies essentially distilling instrument functionality and demonstrating an evolution from unidimensional to multidimensional views. Rothwell categorized instruments based on resource-market relationships into supply-side, demand-side, and environmental types, focusing, respectively, on augmenting supply, stimulating demand, and optimizing external conditions to influence target achievement [35]. Hood’s NATO model emphasized operational forms, proposing eight instrument types including exhortation, legislation, and grants/loans [36]. Schneider classified instruments into five categories (e.g., authority-based, incentive-based) according to their source of influence over target behaviors, while McDonnell and Elmore adopted similar classification logics [37]. However, unidimensional classifications exhibit limitations by focusing solely on instrument form or function, neglecting the critical alignment with policy objectives and implementation domains. For example, relying solely on legal mandates to govern residential waste sorting may prove ineffective due to high enforcement costs, necessitating complementary incentive-based instruments.

The studies by Rothwell, Hood, and Schneider have each made distinct contributions to the classification of policy instruments. Rothwell categorized policy instruments into three types: Supply, Environmental, and Demand. Supply instruments are defined as those that directly provide resources—such as funding or technology—to the targets of policy implementation. Environmental instruments function by shaping the policy context through regulatory measures or planning mechanisms. Demand instruments aim to stimulate market demand via procurement policies or other market-based interventions. This classification emphasizes the dynamic interaction between policy objectives and the specific context in which policies are implemented. Hood introduced the NATO framework (Nodality, Authority, Treasure, and Organization), which classifies policy instruments into four categories based on the core resources of government. Nodality instruments guide behavior through the dissemination of information and data; Authority instruments rely on legal coercion; Treasure instruments leverage financial incentives or sanctions; and Organization instruments involve direct administrative execution by government bodies. This approach reveals the operational mechanisms of policy tools from the perspective of governmental resource allocation. Schneider and Ingram, focusing on the behavior of policy targets, classified policy instruments into five types: Authority Tools, Incentive Tools, Capacity-Building Tools, Symbolic and Hortatory Tools, and Learning Tools. Their framework highlights the need for flexible combinations of instruments tailored to specific policy goals. Previous research has underscored the diversity of policy instrument typologies, providing a valuable reference for contemporary studies on complex policy systems. In contrast, this study adopts a classification approach that is more closely aligned with the characteristics of critical metals. In light of China’s policy formulation logic—which emphasizes long-term planning and top-down administrative guidance—we innovatively incorporate planning-type and guideline-type policy instruments. Furthermore, in accounting for the underlying governance logic of critical metals, this study designs policy objectives based on practical contexts—such as supply security, trade regulation, and industrial development. Finally, considering the full life cycle of critical metals, we introduce the dimension of Implementation Domain, classifying policies based on the specific life cycle stage or sector in which they are applied. This integrated framework helps prevent misalignment between instruments and policy goals and domains, thereby enhancing its applicability to the analysis of complex governance scenarios, such as that of critical metals.

Consequently, this study adopts the integrated “Instrument Type-Policy Objective-Implementation Domain” (T-O-D) three-dimensional analytical framework, synthesizing prior research. By interrelating instrument types, policy objectives, and implementation domains, this framework enables multidimensional calibration. It facilitates more precise instrument selection aligned with actual governance needs, effectively resolving mismatch dilemmas between instruments and objectives/domains, thereby meeting the demands of complex governance scenarios.

Based on the above analysis, this study proposes Hypothesis 2.

H2. 

Employing scientifically multidimensional policy instrument classification frameworks significantly enriches analytical perspectives in policy evolution research.

2.3. Structural Dynamics of Policy Networks

Policy networks serve as a vital theoretical framework for analyzing the mechanisms of public policy formulation and implementation [38]. Their core value lies in revealing the interactive logics and structural characteristics among multiple actors within policy processes. The essence of policy networks resides in the relational structures formed through stable interactions among diverse actors during policy development. A mutually constitutive relationship exists between policies and networks. On the one hand, policy enactment reshapes network structures, influencing interactive relationships among actors and potentially diminishing or elevating specific actors’ roles and positions within the network. Conversely, network structures exert constraining and enabling effects on policy processes. Since the introduction of Social Network Analysis (SNA) into policy network research, the field has experienced a paradigm shift from qualitative description toward quantitative measurement [39,40]. This methodological advancement enables the quantitative parsing of relational attributes through network metrics, uncovering structural features and correlating them with policy outcomes. Such approaches have gained extensive application across energy, industrial, and food security policy domains.

Policy network research not only helps explain the root causes of policy implementation gaps but also illuminates pathways for collaborative governance. Findings derived from integrating SNA offer actionable insights for optimizing policy design and execution, ultimately enhancing governance efficacy.

Based on the above analysis, this study proposes Hypothesis 3.

H3. 

Policy network structures fundamentally drive critical metals policy evolution through dual mechanisms of actor-positional empowerment and constraint.

3. Materials and Methods

3.1. Research Framework

This study examines the evolution of China’s critical metals policies through integrated policy instrument and policy network perspectives, constructing a four-phase analytical sequence: First, systematically collecting and statistically profiling policy texts while demarcating developmental stages based on pivotal historical events; subsequently applying the T-O-D three-dimensional framework to conduct line-by-line textual interpretation, coding, and classification, then statistically tracing evolutionary patterns in instrument types, policy objectives, and implementation domains; concurrently identifying all participating government entities through coded records to construct inter-agency network matrices via SNA, quantifying structural metrics and identifying core nodes; ultimately synthesizing instrument configuration shifts and network dynamics to elucidate macro-level evolutionary characteristics and driving forces, with the complete framework visualized in Figure 1.

3.2. Data Sources

3.2.1. Text and Data

This study utilizes publicly released policy documents from the Chinese government as its research sample. The selected texts are issued by entities including the General Office of the State Council, its constituent departments, special agencies directly under the State Council, institutions directly under the State Council, administrative agencies directly under the State Council, national bureaus, and other government departments. A full-text search was conducted on the Central People’s Government of the People’s Republic of China website (www.gov.cn) using keywords such as “strategic minerals”, “critical minerals”, “strategic metal minerals”, and “critical metals”—which in Chinese policy documents correspond to the critical metals examined in this study. The search included policy texts published up to December 2024.

The initially retrieved policy documents underwent a two-stage screening process. The first stage excluded invalid policies that had been repealed or superseded. The second stage involved a detailed review of the full text of each policy, filtering based on keywords such as “exploration”, “mining”, “reserve”, and “export” to exclude texts with low relevance. This process ensured that each retained policy aligns with the three-dimensional analytical framework (T-O-D) established in this study. Ultimately, 236 valid policy texts were identified, forming the empirical basis for this research.

3.2.2. Analytical Unit Coding

Within the application framework of content analysis, analytical units refer to concrete entity objects designated for description and interpretation, with their delineation criteria determined not by textual length but by the completeness of information conveyed. These units may constitute fundamental semantic elements (e.g., phrases, sentences) or extend to complex textual structures (e.g., sections, full documents). By deconstructing policy texts into quantifiable and comparable analytical units, this approach not only captures the precise configuration of policy instruments but also reveals the logical relationships and functional positioning within policy evolution processes, thereby providing micro-level analytical support for policy change research. Given that core policy provisions embody the essential elements of policy texts, this study defines individual policy provisions as foundational analytical units. Through line-by-line coding and content deconstruction of provisions, we achieve granular interpretation of policy documents. Applying the previously established T-O-D three-dimensional framework, this research ultimately yielded 326 discretely coded units with independent semantic functions for subsequent quantitative statistics and qualitative analysis.

For example, the “Implementation Plan for the Mineral Prospecting Action in the ‘14th Five-Year Plan’ Period” issued by the Chinese government in 2023 serves as a guiding document for the country’s new round of mineral prospecting breakthrough strategy. This policy document was jointly issued by five ministries and commissions, including the Ministry of Finance and the Ministry of Natural Resources, indicating collaboration among these five departments within the policy network. The text sets the goal of adding new reserves of critical metals by 2025, with key breakthroughs in lithium, copper, rare earths, and other critical metals, and the formation of 5–8 large-scale resource bases, explicitly stating that investment will be guided by fiscal resources. Here, the policy instrument types can be identified as Planning-based (as it is a medium- to long-term plan targeting China’s 14th Five-Year Plan period) and Incentive-based (as the government encourages investment), with the policy objective being Resource Supply. The large-scale resource bases correspond to the key implementation domains of Exploration and Mining. Consequently, this policy text can be recorded as involving 5 government departments, 2 instrument types, 1 policy objective, and 2 implementation domains for statistical analysis.

3.3. Model

3.3.1. Policy Instruments Analysis Model

Building on established policy instrument taxonomies while incorporating the distinctive attributes of critical metals, this study advances an innovative T-O-D three-dimensional analytical framework (

Table 1. Three-Dimensional Policy Instrument Analysis Framework (T-O-D).

Dimension	Indicator	Explanation
X-Dimension (Instrument Type)	Authority-based	Enforces or supervises policy implementation through mandatory actions, including laws and implementing regulations.
	Planning-based	Formulates long-term development plans defining strategic directions, typically without detailed operational rules (e.g., outlines, plans).
	Guidance-based	Provides operational directives from higher to lower administrative levels (e.g., guidelines, directives).
	Incentive-based	Influences target groups’ behavior via fiscal measures (e.g., subsidies, tax exemptions).
Y-Dimension (Policy Objective)	National Security	Ensures security through export restrictions and mining volume controls.
	Resource Supply	Secures mineral supply via exploration initiatives and strategic reserves.
	Industrial Development	Guides industrial layout, structural adjustments, and key technology projects.
	Market Regulation	Standardizes industries and regulates market transactions.
	Recycling Utilization	Promotes waste recycling and process improvements.
	Ecological Protection	Enhances environmental governance through legislation and taxation.
Z-Dimension (Implementation Domain)	Exploration	Identifies mineral deposits and evaluates ore types/reserves.
	Mining	Extracts and refines ores from operational deposits.
	Processing	Based on availability, technologies such as smelting, refining, and material processing are employed to transform raw ores and concentrates into high-value-added products, including high-purity metals, alloys, and high-performance materials, thereby addressing resource scarcity and enhancing strategic value.
	Trading	Facilitates domestic and international sales/purchases of mineral products.
	Recycling	Reclaims waste minerals and rehabilitates mining sites (e.g., tailings ponds, abandoned mines).

).

These dimensions exhibit complementary emphases while maintaining intrinsic interconnections, enabling granular analysis of critical metals policy evolution characteristics. The Type dimension (T) classifies instruments according to their operational logics, governmental intervention intensity, resource allocation modalities, or market/society interaction dynamics. The Objective dimension (O) captures the intended purposes, requirements, and outcomes of policy implementation—constituting both fundamental policy conditions and prerequisites for execution. The Domain dimension (D) designates the specific stages or processes engaged during implementation, requiring delineation based on critical metals’ full life-cycle progression. Grounded in the canonical classification models of McDonnell and Elmore and Schneider, this framework represents a domain-specific advancement for critical metals policy analysis [41].

3.3.2. Policy Networks Analysis Model

This study adopts social network analysis methodology, representing actors as nodes and inter-actor relationships as arcs (or edges) to construct a network. Assuming nodes $V$ combined with arcs (or edges) form set $E$, if all elements in $E$ possess weight values, the network is validated. Building on this theoretical foundation, the policy network analysis model is defined as follows:

Definition 1:

An inter-entity coefficient matrix $A={\left[a_{ij}\right]}_{n\mathrm{*}n }$exists, where${ a}_{ij }$denotes the correlation coefficient between the $i$-th and$j$-th entities, and$n$is the total number of entities ($1\le i,j\le n$).

Using $A$ as the adjacency matrix, an undirected network graph $G=\left(V,E\right)$ is derived, where $V$ is the set of all vertices $v_i$; $E$ is the set of all edges $e_{ij}$; the weight of $e_{ij}$ is $a_{ij}$.

Definition 2:

The undirected graph$G$represents the policy network, where${ v}_{i }$ denotes the $i$-th government department;${ e}_{ij }$denotes the association between the $i$-th and $j$-th departments.

Given data availability, this study employs the frequency of co-published policy texts as the criterion for evaluating inter-departmental collaboration.

Definition 3:

$e_{ij}$ equals the number of jointly published policy texts between the $i$-th and $j$-th departments.

Centrality and structural holes are selected as metrics for network structure analysis. Centrality reflects both the core departments in the policy network and the network’s overall agglomeration. Centrality metrics include degree centrality, betweenness centrality, and closeness centrality. Structural holes measure the control capacity of core departments over the entire network during its operation. Structural hole metrics include effective size, efficiency, constraint, and hierarchy.

(1) Degree Centrality: Measures the number of connections a node has with others. A higher number of connections indicates greater centrality. The formula is

$$C_D\left(n_i\right)=d\left(n_i\right)$$

(1)

(2) Betweenness Centrality: Indicates whether a node lies on the shortest path between any two other nodes. Let $g_{jk}$ be the number of shortest paths between nodes $j$ and $k$, and $b_{jk}\left(i\right)$ denote the probability that node $i$ controls interactions between $j$ and $k$. The number of shortest paths between $j$ and $k$ passing through $i$ is $g_{jk}\left(i\right)$. The formula is

$$b_{jk}\left(i\right)={\displaystyle \frac{g_{jk}\left(i\right)}{g_{jk}}}$$

(2)

For an undirected network, the betweenness centrality $C_B$ of node $i$ is calculated as

$$C_B=\sum _j^n\sum _k^n{b}_{jk}\left(i\right),\left(j\ne k\ne i\right),j<k$$

(3)

(3) Closeness Centrality: Measures a node’s proximity to all other nodes in the network. The formula is

$$C_C={\left(\sum _{j-1}^n{d}_{ij}\right)}^{-1}$$

(4)

where $d_{ij}$ is the geodesic distance between nodes $i$ and $j$.

(4) Effective Size: Measures the number of non-redundant connections a node has, representing its unique information sources. The formula is

$${ES}_i=\sum _j\left(1-\sum _q{p}_{iq}{m}_{jq}\right), q\ne i,j$$

(5)

where $j$ represents all nodes connected to $i$; $q$ is any third node excluding $i$ or $j$; $p_{iq}{m}_{jq}$ denotes redundancy between $i$ and $j$; $p_{iq}$ is the proportion of $i$’s connections directed to $q$.

(5) Efficiency: Measures a node’s effectiveness in information transmission. The formula is

$${EF}_i=\sum _j\left[1-\sum _q{p}_{iq}{m}_{jq}\right]/n, q\ne i,j$$

(6)

(6) Constraint: Measures the constraints on a node’s ability to leverage structural holes for accessing information and resources. The formula is

$$C_{ij}=p_{ij}+\sum _q{p}_{iq}{p}_{jq}$$

(7)

where $P_{ij}$ is the ratio of $i$’s connections to $q$ relative to all its connections.

(7) Hierarchy: Reflects power distribution imbalances within the network. The formula is

$$H={\displaystyle \frac{\sum _j\left(\frac{C_{ij}}{C/n}\right)\ln \left(\frac{C_{ij}}{C/n}\right)}{n\ln \left(n\right)}}$$

(8)

where $n$ is the number of nodes connected to $i$, and $C/n$ is the average constraint per node.

4. Results

4.1. Policy Stage Division

Analysis of the publication volume trends of China’s critical metals policies from 1973 to 2024 (see Figure 1), combined with landmark policy events, reveals four distinct developmental stages:

(1) Phase 1 (1973–2000): During China’s initial industrialization, the concept of “critical metals” remained undefined. Although protective measures were implemented for specific minerals (e.g., tungsten, tin, antimony, ion-adsorption rare earths), policy frameworks were underdeveloped. Metal resources were governed under uniform regulatory approaches.

(2) Phase 2 (2001–2005): Policy issuance increased compared to Phase 1. The landmark 2001 introduction of “strategic minerals” (equivalent to critical minerals, encompassing energy/metallic/non-metallic resources) marked a conceptual shift. Subsequent policies referenced critical metals but lacked formal inventories.

(3) Phase 3 (2006–2015): The 2006 “Decision on Strengthening Geological Work” directed exploration of critical metals. In 2012, China imposed total mining quotas and export controls for dominant minerals (e.g., tungsten, ion-adsorption rare earths). Policy output grew quantitatively and thematically.

(4) Phase 4 (2016–Present): A pivotal 2016 policy formally designated 24 strategic minerals (including 14 critical metals), initiating institutionalized governance. The 2021 “Mineral Exploration Breakthrough Initiative” prioritized critical metals exploration. This phase exhibits surging policy volume, expanding regulatory scope, and heightened strategic emphasis.

4.2. Evolution of Policy Instrument

4.2.1. Evolutionary Characteristics of Instrument Types

The evolution of policy instruments reflects China’s transition from a planned economy to a socialist market economy during the early reform era, when synergistic effects of instrument combinations were limited and mineral resource governance prioritized national security objectives (see Figure 2).

The 1996 revision of the Mineral Resources Law established strict state control, cementing authority-based instruments (laws/regulations) as dominant in Phase 1, though the 1998 establishment of the Ministry of Land and Resources and subsequent market activation gradually reduced their prevalence. Post-2001 marketization shifted focus to industrial restructuring, with planning-based (strategic outlines) and guidance-based instruments (operational directives) becoming primary tools in Phase 2 to guide critical metals industry layout, while incentive-based tools remained limited to economic downturns for market stabilization. Following the 2005 campaign to phase out outdated production capacity, Phase 3 saw a resurgence of authority-based instruments targeting ecological governance, exemplified by the 2009 Nonferrous Metals Industry Adjustment and Revitalization Plan prioritizing resource security. After 2016, energy/industrial restructuring accelerated with China’s release of the Strategic Minerals List (encompassing 14 critical metals), where surging downstream demand drove authority-based supply security policies (2018 onward) alongside increased guidance-based instruments; during Phase 4, systematic guidance for critical metals exploration and emerging industries established planning-based and guidance-based instruments as mainstream, while expanding incentive-based tools stimulated downstream product consumption.

China’s critical metals policy instruments demonstrate a pivotal transition from authority-based dominance to a diversified multi-instrument governance paradigm, where marketization progress has driven increasing instrument variety and synergistic effects, enabling enhanced governance efficacy through strategic tool combinations.

4.2.2. Evolutionary Characteristics of Objectives

From the perspective of policy objectives (see Figure 3), the policy objectives of the first stage mainly focus on national security and resource supply.

In the early days of reform and opening up, the government regarded the supply security of key metals as its top priority. Since 1989, with the rapid development of China’s market economy, the trading of key metal-related products has entered the market-oriented era, and the proportion of industrial development and market supervision in policy objectives has increased. In the second stage, the proportion of industrial development and market regulation in policy objectives has further increased. The government enhances the competitiveness of industries through long-term planning and standardized management. With the advancement of global economic integration, international trade in mineral products has become increasingly active. However, industrialization, urbanization and the continuous growth of population have also caused certain ecological and environmental problems. Since 2012, China has strengthened the policy design for ecological and environmental protection. The proportion of ecological protection targets in the third stage has significantly increased. Around 2016, the global mining market entered an upward cycle, with the demand for key metals increasing day by day. In 2017, China vigorously promoted resource exploration and the recycling of scrap metals, and strengthened the supervision of the entire industrial chain of key metals. In recent years, China has continuously promoted high-quality economic development and comprehensively enhanced the influence of its industrial and supply chains through scientific and technological innovation, industrial upgrading, and international cooperation.

The evolution of policy objectives shows an evolutionary feature from simple to complex. From the early focus on national security and resource supply, it has gradually expanded to multiple fields such as industrial development, market regulation, and ecological protection. The evolution characteristics of policy objectives are closely related to the phased features of China’s economic development. The policy objectives at different stages reflect the main contradictions and demands of economic and social development at that time, and also demonstrate the deepening and expansion of China’s governance concepts for key metals.

4.2.3. Evolutionary Characteristics of Implementation Domains

Analysis of policy implementation domains reveals China’s persistent focus on the Processing (see Figure 4), reflecting governmental prioritization of mid- and downstream industrial development. Prior to 1999, relatively low industrialization levels led to mineral resources being primarily exported for foreign exchange earnings, with policies targeting exploration, mining, and trading.

Entering the 21st century, accelerated industrialization drove surging demand for critical metals, shifting policy emphasis toward processing enhancement to meet domestic market needs through improved extraction efficiency and processing technologies.

During Phase 3, policies increasingly emphasized processing and recycling, leveraging technological advancements to drive industrial upgrading while addressing ecological pressures through improved waste recovery. Post-2016, rising global demand intensified supply security concerns, reinvigorating exploration as the foremost priority through technological innovation and policy support.

Current policies demonstrate balanced domain distribution without over-concentration on single stages, reflecting coordinated development across all implementation domains.

4.3. Policy Network Characteristics

4.3.1. Overall Network Features

Analysis of 236 policy documents across 22 government entities (adjusted for institutional restructuring) reveals distinct evolutionary patterns: the National People’s Congress and State Council functioned as apex coordinators with exclusive unilateral issuances, while 171 policies were unilateral and 65 were co-authored (

Table 2. Policy Issuance by China’s Government Agencies.

Abbreviation	Agency	Unilateral	Joint	Total
CBIRC	China Banking and Insurance Regulatory Commission	0	2	2
MOC	Ministry of Commerce	0	5	5
NEB	National Energy Bureau	0	5	5
EMM	Ministry of Emergency Management	1	3	4
GAC	General Administration of Customs	0	2	2
SAMR	State Administration for Market Regulation	2	5	7
SC	State Council	129	0	129
MARA	Ministry of Agriculture and Rural Affairs	4	2	6
MEE	Ministry of Ecology and Environment	3	17	20
MOHURD	Ministry of Housing and Urban-Rural Development	0	3	3
MIIT	Ministry of Industry and Information Technology	7	15	22
MNR	Ministry of Natural Resources	34	14	48
MOE	Ministry of Education	0	3	3
MOF	Ministry of Finance	2	12	14
MOST	Ministry of Science and Technology	2	7	9
MOT	Ministry of Transport	0	2	2
MPS	Ministry of Public Security	0	2	2
NDRC	National Development and Reform Commission	3	14	17
NHC	National Health Commission	0	2	2
NPC	National People’s Congress	8	0	8
PBC	People’s Bank of China	0	3	3
NFGA	National Forestry and Grassland Administration	0	7	7

).

During Phases 1–3, collaborative policymaking was minimal (only 3 joint documents in Phase 2), indicating pre-2016 dominance by single agencies—primarily the State Council and Ministry of Natural Resources (

Table 3. Statistical Indicators of Overall Network Characteristics.

Indicator		Phase 1	Phase 2	Phase 3	Phase 4
Sample Size	Unilateral	23	12	44	90
	Joint	0	3	0	64
	Total	23	15	44	154
Network Scale		-	9	-	20
Number of Relations		-	20	-	92
Connection Frequency		-	30	-	260
Network Density		-	0.583	-	0.489
Network Connectivity		-	1.000	-	1.000
Average Shortest Path		-	1.417	-	1.579
Network Efficiency		-	0.536	-	0.567

). Phase 4 witnessed substantial growth in network scale, relational ties, and connection frequency, demonstrating diversified interagency collaboration. Visualization identifies the National Development and Reform Commission (NDRC), Ministry of Industry and Information Technology (MIIT), and Ministry of Finance (MOF) as core nodes, reflecting their pivotal roles in joint policymaking. The post-2016 surge in co-authored policies correlates with expanded departmental engagement following China’s Strategic Minerals List release. NDRC-MIIT collaborations focused on industrial chain development and green manufacturing transitions, while MOF-led initiatives prioritized fiscal subsidies and tax incentives for critical metals production.

Network visualization (see Figure 5) reveals that NDRC, MIIT, and MOF occupy central positions with larger node sizes, indicating their pivotal roles in joint policymaking. The substantial increase in co-authored policies during Phase 4 demonstrates expanded departmental engagement following the critical metals list release, resulting in intensified interagency collaboration. Textual analysis shows that MIIT- and NDRC-led joint policies primarily address comprehensive industrial chain development, aiming to facilitate critical metals industry transformation, enhance domestic supply chain competitiveness, and accelerate new industrialization. In contrast, MOF-coauthored policies predominantly focus on fiscal measures including subsidies and tax incentives.

4.3.2. Network Metric Characteristics

Analysis of network metrics reveals distinct institutional roles (

Table 4. Network Structure Measurement Results.

Agency	Centrality			Structural Holes
	Degree Centrality	Betweenness Centrality	Closeness Centrality	Effective Size	Efficiency	Constraint	Hierarchy
CBIRC	5.00	0.000	36.000	2.310	0.462	0.453	0.007
MOC	12.00	2.952	26.000	5.562	0.464	0.343	0.167
NEB	13.00	10.884	25.000	7.521	0.579	0.278	0.087
EMM	5.00	0.000	34.000	1.988	0.398	0.493	0.024
GAC	6.00	0.000	33.000	2.337	0.390	0.460	0.039
SAMR	10.00	18.000	28.000	4.956	0.496	0.308	0.081
MARA	5.00	0.000	36.000	2.383	0.477	0.483	0.120
MEE	15.00	12.927	23.000	8.332	0.555	0.312	0.227
MOHURD	12.00	6.370	26.000	6.414	0.534	0.321	0.100
MIIT	16.000	21.357	22.000	9.577	0.599	0.281	0.210
MNR	11.000	5.503	28.000	6.374	0.579	0.348	0.203
MOE	7.000	1.364	32.000	3.693	0.528	0.398	0.068
MOF	16.000	17.857	22.000	8.636	0.540	0.328	0.272
MOST	10.000	5.217	29.000	5.671	0.567	0.327	0.088
MOT	9.000	0.000	29.000	4.082	0.454	0.347	0.020
MPS	9.000	0.000	29.000	4.082	0.454	0.347	0.020
NDRC	14.000	7.569	24.000	6.990	0.499	0.334	0.252
NHC	1.000	0.000	46.000	1.000	1.000	1.000	1.000
PBC	5.000	0.000	36.000	2.310	0.462	0.453	0.007
NFGA	5.000	0.000	36.000	2.331	0.466	0.531	0.306

Computation results retain three significant figures.

).

From the perspective of centrality metrics, both MIIT and MOF exhibit a degree centrality of 16 and a closeness centrality of 22, reflecting broad direct connections and a central role in information dissemination. MIIT demonstrates a betweenness centrality of 21.357, significantly higher than other agencies, underscoring its function as an critical bridge in policy integration, interdepartmental coordination, and information transfer. It plays a leading role in critical mineral supply chain management, technological innovation, and industrial planning. SAMR and MEE show betweenness centrality values of 18 and 12.927, respectively, indicating their substantial influence in shaping policies related to market regulation and environmental oversight. NDRC, with a betweenness centrality of 7.569, retains moderate policy coordination capacity. In contrast, agencies such as CBIRC and EMM score 0 on betweenness centrality, indicating limited proactive influence and a predominantly receptive role in information networks. MIIT, MOF, and MEE each possess an effective size greater than 8, an efficiency above 0.54, and a constraint below 0.33, suggesting extensive non-redundant connections, efficient access to diverse information, and strong leadership in policy formulation, resource mobilization, and bargaining. By comparison, NHC has an effective size of 1 and a constraint of 1, reflecting complete reliance on external information inputs and a peripheral policy role. NFGA shows a constraint of 0.531 and a hierarchy of 0.306, indicating dependence on a limited number of agencies and constrained independent influence.

The efficiency metric further elucidates interagency differences in the ability to integrate information and coordinate resources in critical metals governance. Higher efficiency values reflect a stronger capacity to acquire diversified and independent information through fewer contacts, thereby enhancing strategic flexibility in cross-sector collaboration. This trait is closely linked to industry-specific applications of critical metals. For instance, rare earth elements are used in high-technology sectors including defense, permanent magnet motors, and new energy vehicles, involving multiple agencies such as MIIT, MNR, and MOST. Within this context, MIIT achieves an efficiency of 0.599. Despite a degree centrality of 16—indicating numerous connections—it maintains high non-redundancy, effectively integrating R&D support from MOST, reserve management by MNR, and industry standard-setting by SAMR. This mitigates policy duplication and conflict while increasing supply chain responsiveness. In the case of lithium, which is chiefly used in power batteries and involves MNR, MIIT, NDRC, and MEE, MNR’s efficiency of 0.579 enables rapid incorporation of environmental constraints from MEE and industrial plans from MIIT into mining rights management, reducing informational delays in permitting processes. Conversely, agencies such as MOT and MPS exhibit an efficiency of only 0.454, rendering them prone to inefficient coordination and redundant information flows during battery industry policy adjustments. For critical metals such as tungsten and antimony—used in aerospace, military, and fire-resistant materials—governance requires coordination among MIIT, EMM, and SAMR. MIIT’s high efficiency (0.599) facilitates rapid alignment with EMM’s production safety requirements and SAMR’s quality supervision directives. However, SAMR’s lower efficiency (0.398) suggests a risk of information overload or overreliance on limited channels, potentially leading to delayed responses to emergent safety incidents. In cobalt and nickel supply chain management—key metals for battery cathodes—MIIT’s high efficiency is instrumental, whereas insufficient efficiency at MOF could hinder coordination between international procurement and domestic environmental policies, particularly under high import dependency. High-efficiency agencies such as MIIT, MNR, and MOST maintain informational advantages and decisional agility in cross-sectoral collaboration, contributing to policy synergy and effectiveness. Conversely, low-efficiency agencies are more likely to act as informational bottlenecks within the policy network.

In summary, MEE, MIIT, MOF, and NDRC exhibit high degree centrality, indicating their extensive interdepartmental connections. MIIT, MOF, SAMR, and MEE show elevated betweenness centrality, signifying critical brokerage functions. While CBIRC, EMM, and MARA demonstrate lower degree/betweenness centrality, their high closeness centrality reflects influential positioning; MEE, MIIT, and MOF achieve superior efficiency in information transmission, whereas NFGA and CBIRC exhibit low constraint values, implying greater operational autonomy. Functionally, MEE-dominated policies emphasize ecological protection and resource recycling, aligning with China’s green development priorities; MIIT’s peak betweenness centrality establishes it as the core coordinator for joint policymaking, while NHC’s maximal closeness centrality places it peripherally. Crucially, NDRC and MIIT combine large effective size with low constraint, confirming their roles as central hubs accessing non-redundant information and driving critical metals governance.

4.4. Reliability Validation

Coding reliability serves as a critical foundation for textual quantitative analysis. To ensure research validity, we conducted inter-coder reliability tests using Holsti’s formula:

$$A={\displaystyle \frac{2M}{N_1+N_2}}$$

(9)

where $M$ = number of identical codes, $N_1$ and $N_2$ = total codes per round. Values ≥ 0.8 indicate acceptable reliability. After independent recoding at one-month intervals, results demonstrated robust consistency: Reliability validation results confirm coding consistency rates of 94.8 percent for the X-dimension, 93.9 percent for the Y-dimension, and 92.3 percent for the Z-dimension with government agency coding achieving 99.4 percent consistency, collectively demonstrating that this study has passed reliability testing thresholds.

4.5. Testing of Research Hypotheses

All three hypotheses proposed in this study were supported by the empirical findings. First, H1 posits that the evolution of critical metals policies is characterized by multi-objective orientation and increasing institutional complexity. The policy stage division reveals that China’s critical metals policies have expanded from an initial focus on national security and resource supply to encompass multiple objectives such as industrial development, market regulation, ecological protection, and recycling. The types of policy instruments have also evolved from authority-based dominance to a diversified hybrid model, reflecting an increasing institutional complexity within the policy system and a trend toward synergistic of multidimensional goals. Second, H2 proposes that adopting a scientific multidimensional policy instrument classification framework can significantly enrich the analytical perspective of policy evolution research. The three-dimensional analytical framework (T-O-D) constructed in this study was successfully applied to the coding and quantitative analysis of 326 policy clauses. It revealed dynamic changes in policy instrument combinations, target priorities, and implementation areas across different stages, demonstrating the framework’s strong theoretical adaptability and empirical explanatory power, thereby effectively expanding the analytical dimensions of policy texts. Finally, H3 argues that policy network structures drive policy evolution through dual mechanisms of actor position empowerment and constraint. Social network analysis results show that the MIIT and the NDRC exhibit high degree centrality, betweenness centrality, and structural hole control within the network, indicating their central role in policy coordination and information dissemination, which has facilitated the formation of a multi-agency collaborative governance model. A significant increase in network density and connection frequency in the fourth phase further corroborates the influence of policy network structures on the combination and evolutionary trajectory of policy instruments.

In summary, all three hypotheses are supported by the findings of this study.

5. Discussion

The evolution of China’s critical metals policies reveals distinct governance patterns: during the first three phases, policymaking was predominantly led by single agencies, while Phase 4 witnessed a transformative shift toward multi-departmental collaboration—reflecting expanded management imperatives for critical metals.

Government intervention and market mechanisms jointly drive policy evolution, though state control remains the dominant force [42]. The extensive scope of critical metals governance necessitates integrating cross-sectoral expertise, where interagency collaboration synthesizes specialized knowledge to enhance policy efficacy. Supply security consistently prioritizes policymaking considerations, with rising demand elevating it as a core metric of governance capability, explaining the prevalence of unilateral issuances by apex bodies (e.g., National People’s Congress and State Council). The broadening conceptualization of critical metals governance itself fuels interdepartmental cooperation, serving as an endogenous driver of policy evolution. Market mechanisms complement state efforts by optimizing resource allocation through price signals, stimulating corporate innovation, and advancing industrial upgrading.

Interagency collaboration is central to addressing governance challenges in critical metals. The essence of such cooperation lies in integrating resources, expertise, and administrative authority to overcome the functional limitations of individual agencies. In terms of supply chain security, coordination among MIIT (industrial planning), MNR (resource reserves), and GAC (import regulation) is particularly crucial. Taking China’s rare earth industry as an example, agencies share data on “resource reserves—industrial demand—import dependency” to establish a national rare earth reserve mechanism. Under the coordination of NDRC, MIIT estimates domestic rare earth demand, MNR sets mining quotas based on administrative authority, and GAC regulates import and export flows to mitigate international price fluctuations and ensure raw material supply for China’s new energy vehicle industry. Key policy documents such as the “Rare Earth Management Regulations” and the “Interim Measures for the Control of Total Rare Earth Mining and Smelting Separation” have been jointly issued. In technological innovation, collaboration between MOST (R&D funding) and MIIT (industrial application) helps bridge the gap between technology and industry. To address technical bottlenecks in power battery cathode materials (lithium, cobalt), multiple agencies jointly issued the “Opinions on Improving Insurance Compensation Policies for the First Set of Major Technical Equipment and the First Batch of New Materials”. MOST provides R&D funding to support research institutions in developing solid-state battery materials, while MIIT facilitates pilot platforms with enterprises to rapidly translate lab achievements into mass-production technologies. In 2024, this cooperation increased the energy density of China’s high-nickel ternary materials by 15%, demonstrating the synergistic value of expertise, critical metals, and capital. In environmental regulation, joint enforcement by MEE (environmental standards) and MNR (mining supervision) helps avoid ecological issues in mining development. During lithium mining in Qinghai Salt Lake, MEE formulated ecological protection standards, which MNR incorporated into mining permit approvals. The two agencies shared monitoring data and conducted joint onsite inspections. In 2023, the wastewater reuse rate in Qinghai’s lithium mining reached 92%, reflecting the integration of administrative power (approval and enforcement) and technical expertise (environmental standards). In market regulation, cooperation between MOF (fiscal tools) and SAMR (price supervision) can effectively curb domestic price volatility of critical metals. In 2022, when lithium prices surged, MOF introduced mining subsidies to ease cost pressures on enterprises, while SAMR cracked down on illegal hoarding. The two agencies jointly issued the “Catalog of Key New Materials for First Application Demonstration”, combining fiscal and regulatory measures to stabilize domestic lithium prices. This demonstrates that no single agency can fully cover resource allocation, technical expertise, and regulatory enforcement. Interagency collaboration effectively integrates diverse elements to achieve critical metal governance objectives. Close cooperation among departments is essential to enhancing governance capabilities in this field.

Looking forward, policy formulation will increasingly emphasize synergistic coordination between state control and market forces. Their inherent complementarity—market mechanisms enabling efficient resource allocation while government intervention corrects market failures and safeguards national interests—will refine governance frameworks. Continuous policy adaptation will be essential amid technological and socioeconomic shifts, where innovations in extraction/processing technologies may reshape global supply dynamics, and industrial transitions alter demand patterns. China’s trajectory will maintain state-directed governance with market-assisted coordination, evolving toward a synergistic co-driver model characterized by increasingly specialized interagency collaboration across expanded regulatory domains.

From an international perspective, major economies including China, the United States, the European Union, and Japan have all prioritized ensuring the security of critical metal supplies as a core policy objective. These nations are collectively addressing the restructuring of the global resource landscape and challenges to industrial chain resilience, thereby forming a synthesis of differentiated approaches and common demands. The United States centers its strategy on supply chain security legislation (an authoritative tool), establishing an institutional framework through specialized laws and leveraging market mechanisms to guide resource allocation. This approach emphasizes the use of legal rigidity to clarify the boundaries of responsibility between the government and the market. The European Union focuses on list management and industrial chain stability (a planning tool), identifying key minerals by publishing critical raw materials lists. It promotes diversification of supply sources and recycling, while emphasizing regional coordination and the participation of market entities. China adopts a more state-driven model, with the central government orchestrating top-level design. Long-term goals are set through Five-Year Plans and specialized implementation programs, highlighting inter-ministerial coordination and better accommodating the systematic demands of resource governance in an ultra-large-scale economy.

The findings of this study offer relevant insights for nations and regions beyond China. First, the “planning-led and departmentally coordinated” governance model emerging from China’s critical metals policy evolution presents a potential pathway for other countries grappling with dual challenges of resource security and industrial transformation. Specifically, China’s experience in fostering interdepartmental collaboration and establishing core nodes within policy networks suggests that a strong central coordination mechanism, coupled with specialized departmental functions, can enhance policy implementation efficacy. Second, the observed shift in China’s policy instrument mix—from reliance on authority-based tools toward greater use of incentives and guidance—highlights for other jurisdictions the importance of instrument diversity and contextual fit when addressing complex resource issues. Additionally, China’s ongoing efforts to balance resource security with ecological protection provide a relevant reference for resource-rich yet environmentally vulnerable developing nations. Moving forward, countries can adapt lessons from China’s experience in policy network building and instrument innovation—informed by their own institutional contexts and resource endowments—to develop more resilient and sustainable critical metals governance systems.

6. Conclusions

Our primary conclusions are summarized as follows:

(1) China’s critical metals policy evolution demonstrates four distinct stages: Initial Stage (1973–2000), lacking a clear concept and specific policies; Conceptual Emergence Stage (2001–2005), introducing “critical metals” but without a formal list; System Building Stage (2006–2015), promoting systematic exploration through the Decision on Strengthening Geological Work; and Institutional Maturation Stage (2016–present), marked by the critical metals list and a significant increase in policy volume and coordination.

(2) The policy instrument mix has shifted from authority-based dominance to a hybrid model incorporating planning-based, guidance-based, and incentive-based instruments. Policy objectives have expanded from initial resource and national security to multiple goals including industrial development, market regulation, ecological protection, and recycling, reflecting increasing systemic complexity.

(3) The policy network structure has transitioned from single-agency issuance, led primarily by the State Council and the Ministry of Natural Resources, to a multi-agency collaborative governance model centered on the Ministry of Industry and Information Technology (MIIT) and the National Development and Reform Commission (NDRC). Network metrics such as centrality and structural holes confirm the critical role of MIIT and NDRC in policy coordination, information flow, and resource integration.

(4) Building on the central roles of the MIIT and the NDRC, it is advisable to strengthen the inter-ministerial coordination mechanism for critical metals, co-led by these two bodies, with the MOF providing supporting government funding. This mechanism should conduct regular high-level consultations to address major issues in the critical metals sector. In light of the need for information sharing, it is recommended to establish a national platform for policy information and data sharing on critical metals, thereby breaking down information silos among government agencies.

This research acknowledges certain limitations. First, while thoroughly reviewing prior literature, the text-based analytical approach constrains the three-dimensional framework’s capacity to comprehensively capture all facets of policy evolution, offering instead three focused analytical perspectives. Additionally, some critical metals—particularly specific rare earth elements—may be underrepresented in policy documents due to vague terminology, potentially introducing statistical biases. Future studies will refine this framework by conducting separate textual analyses for distinct critical metal categories. Second, this research utilizes the frequency of jointly issued documents to measure inter-departmental collaborative relationships. While this approach has yielded significant findings, it also reveals the limitations of the method. It is important to note that although joint policy issuance is an important pathway for Chinese government agencies to participate in critical metals governance, it cannot capture informal interactions during the policy formulation process or collaboration during the implementation stage. Due to the non-public nature of the government’s policy-making process, these informal forms of cooperation are more concealed, making it difficult to collect relevant data. Future research could employ interviews, archival analysis, or the incorporation of implementation report data to conduct triangulation, thereby providing a more comprehensive depiction of policy networks. Third, constrained by the research methodology, this study, although capable of illustrating the evolving nature of policy instruments, is unable to delineate coordination or conflicts among policies. It may overlook the interactive effects resulting from different policy provisions. We hope that this work’s demonstrated utility, practical applications, and identified constraints will stimulate scholarly discourse toward advancing critical metals policy research frameworks.