From Data to Dominance: The Economics of Chinese Computing Power and Its Global Implications

Abstract

This in-depth strategic and competitive intelligence analysis examines China's Computing Power Economy (EPC, 算力经济, Suànlì Jīngjì), a core component of the nation's "New Quality Productive Forces" strategy. Leveraging data from the CAICT (2025), the paper quantifies the EPC's massive infrastructure, including 10.431 million standard racks and 748 EFLOPS of intelligent computing power (Zhìsuàn). It details the macro-level GDP spillover effect (0.0426% GDP growth per 1% computing scale increase), the state-driven "East Data, West Computing" strategy (Dōng Shù Xī Suàn), and the crucial role of algorithmic innovation (e.g., DeepSeek R1 series) in mitigating hardware dependence. The analysis emphasizes the Dual-Use implications, highlighting the strategic vulnerabilities (reliance on high-performance chips) and the proactive measures in talent cultivation (e.g., "Computing Power Vouchers") and green energy integration (Electricity-Computing Synergy). It concludes with a structured SWOT analysis and future scenarios for technological self-reliance and global geoeconomic projection.

A high-tech rendering of a central processing unit (CPU) surrounded by holographic data, global projections, and digital elements, with Chinese characters for "New Productive Forces" and "Dual-Use," symbolizing China's strategic Computing Power Economy.

1. The Strategic Concept: The Computing Power Economy (EPC)

1.1. Foundations and Geopolitical Positioning

The Computing Power Economy (EPC, 算力经济, Suànlì Jīngjì) is defined in China as an advanced economic form that transforms raw computing power into a productive force (生产力) through the synergistic collaboration between computing infrastructure and specialized labor. It is the critical engine for realizing the value of the data element and driving both digital transformation (数字化转型) and intelligent upgrading (智能化升级) across all industries. The EPC is officially positioned as a strategic pillar for reshaping national competitive advantage and is regarded as the necessary step from "digital civilization" to "intelligent civilization."

Globally, economic competition has intensified into a "contest centered on computing resources." Computing power has evolved from mere "passive support" to a "core driver that dictates the rules."

1.2. Economic Impact and Operational Mechanisms

The EPC exerts a widespread influence through a significant spillover effect (溢出效应).

1.2.1. Microeconomic Level: Enterprise Optimization

At the firm level, the EPC drives efficiency and innovation:

1. Technological Capability Enhancement. Companies enter a "dynamic cycle" where raw business data is processed, modeled, and converted into "knowledge assets" (知识资产) (e.g., user profile databases, market trend analysis). These assets have a marginal cost that approaches zero, fueling algorithmic upgrades and enhancing core technical innovation capability.

2. Shift in Development Model. The operational logic shifts from traditional economies of scale (规模经济) (based on asset expansion and standardization) toward a model based on "flexible supply" (柔性化供给能力) and "precision delivery." Profit is driven by increased product added value and cross-scenario application capability, rather than relying solely on reducing long-run average costs.

3. Cost Reduction and Efficiency Increase (降本增效). Accessing computing resources, algorithms, and data now has a significantly lower marginal cost. This creates a self-reinforcing positive feedback loop: lower marginal cost of internal knowledge capital leads to reduced internal management costs, justifying continuous business expansion based on transaction cost theory.

1.2.2. Macroeconomic Level: Total Factor Productivity (TFP)

On the macro scale, the EPC is the core mechanism for improving Total Factor Productivity (TFP):

1. Factor Value Release. Computing power fully releases the value of the Data (D) element and promotes the efficient collaboration of Technology (A), Capital (K), and Talent (L) through widespread industry applications, thereby increasing overall resource utilization.

2. Spatial Restructuring. The unified national strategy, centered on the 8 National Hub Nodes (e.g., Beijing-Tianjin-Hebei, Yangtze River Delta, Greater Bay Area), drives the cross-regional scheduling of computing resources (East Data, West Computing). This restructuring optimizes resource allocation efficiency by leveraging the comparative advantages of different regions (e.g., lower power costs in the West).

3. Factor Saving and Green Transition. Technological progress driven by the EPC enables the production of the same output level with fewer factor inputs, as illustrated by the inward shift of the isoquant curve. Furthermore, the intense power consumption of computing production mandates an accelerated push for energy-saving technologies (like water-cooling and "source-grid-load-storage" microgrids), which subsequently diffuse to other high-energy consumption sectors (e.g., manufacturing, construction), promoting a systemic green economic transition.

2. The National Project: Infrastructure, Policy, and Numerical Data

2.1. Infrastructural Backbone and Computing Scale

China's long-term strategy involves continuous investment and policy support for computing infrastructure.

• Total Scale. As of March 2025, China's computing centers reached 10.431 million standard racks.

• Intelligent Computing Power (智算, Zhìsuàn). Crucial for AI development, the scale of intelligent computing reached 748 EFLOPS (FP16) by March 2025, maintaining a five-year compound growth rate of 49%.

• Network Integration. The "Action Plan for Computing Power Interconnection" (2025) aims to unify standards and open interfaces between platforms, promoting efficient national resource scheduling and synergistic utilization.

  
  

2.2. Capital, Talent, and Energy Factors

The EPC is driven by massive investment and coordinated factor mobilization:

• Capital Drivers. The high demand from large AI models has attracted significant capital. Major investments include Alibaba's 380 billion RMB over three years for cloud/AI infrastructure, Tencent's 221% increase in capital expenditure (primarily for computing infrastructure), and substantial spending by state-owned carriers like China Mobile (37.1 billion RMB) and China Telecom (22% increase).

• Talent Drivers (Dual-Use Focus). To address talent shortages, regional authorities implement specific measures:

  • Incentive Programs. Cities like Chengdu offer "Computing Power Vouchers" (算力券) to subsidize the use of computing resources by research institutions and high-level talent.

  • Targeted Education. Qinghai Province focuses on cultivating Green Computing talent (绿色算力人才), including dedicated master’s and doctoral programs in collaboration with clean energy enterprises.

  • Research Platforms: Hami City established the Energy Computing Fusion Research Institute to train personnel in the "Numbers, Computing, and Electricity" (Shù Suàn Diàn) triple domain.

    
    

• Energy Drivers (Electricity-Computing Synergy). Computing centers consumed about 166 billion kWh in 2024. To ensure sustainable growth:

  • China leverages its green power (绿电) advantage, with renewable capacity surpassing coal-fired capacity by late 2024.

  • The "Action Plan for Accelerating the Construction of a New Power System" (2024-2027) promotes "Electricity-Computing Synergy" (电算协同, Diànsuàn Xiétóng) and explores "Green Power Aggregation Supply" (绿电聚合供应) models.

    
    

2.3. GDP Spillover Effect: Empirical Data

The empirical analysis confirms the positive impact of computing scale on GDP.

• Overall Impact. Computing scale growth promotes GDP growth. A 1% increase in computing scale drives a 0.0426% increase in GDP.

• Policy Amplification. The designation of a city as a Computing Hub Node amplifies this effect. In 2021, being a hub node resulted in an additional 0.0109% GDP growth for every 1% increase in computing scale

• Economies of Scale: The impact is strongest in high-GDP cities, where the industry has reached maturity (coefficient).

3. Dual-Use Innovation: From Zhìsuàn to AI Algorithms

The strategic value of the EPC lies in its immediate applicability to AI development (Zhìsuàn), critical for both commercial and military superiority.

3.1. The Critical Role of the Algorithmic Ecosystem (IDC Data)

The IDC report (2025) stresses that algorithms are the core engine driving AI development, determining the upper limit of application intelligence⁴⁵. The speed of innovation in China's AI ecosystem is intense:

• Large Model Competition: 2024 saw aggressive competition among large models, including Tongyi Qianwen (Alibaba), Llama3, and the DeepSeek R1 series.

• Algorithmic Breakthroughs: The DeepSeek R1 series is highlighted for making "significant innovations at the algorithmic level". This is crucial intelligence: algorithmic efficiency can achieve superior AI performance with less reliance on the most powerful, and often restricted, high-performance hardware (e.g., top-tier GPUs). This capability increases strategic resilience against technology export controls.

3.2. Sectoral Integration and Dual-Use Case Studies

The EPC is permeating key sectors, showcasing its potential for civilian-military integration:

1. Automotive and Manufacturing (Chengzhi Group - 辰致集团). The group adopted a flexible "1+N+X" (Group + Regional + Edge) architecture. The deployment of Edge Nodes (X) in production lines not only reduced smart factory construction time (e.g., Ya'an base) from 6 to 3 months but also achieved low PUE (<1.4), saving 20% of electricity. Rapid deployment and high efficiency are direct measures of resilience for a wartime economy.

2. Critical Infrastructure and Water Management (Inspur - 浪潮信息). Using Edge Computing for water conservancy (水利行业). Edge devices connect sensors and controllers (PLCs) in real-time for immediate data analysis, enabling remote intelligent scheduling and AI supervision of water quality and personnel safety. This distributed, low-latency control model is directly transferable to the supervision and security of other critical national infrastructure or military assets.

4. Strategic SWOT Analysis

4.1. Strategic Strengths

1. Strategic Vision and State Guidance. Computing power is positioned as the "New Quality Productive Force". Central planning and massive state-aligned capital ensure coordinated, accelerated development.

2. Quantifiable Zhìsuàn Scale: The 748 EFLOPS scale and high growth rate provide the necessary foundation for training large-scale, advanced AI models, which is a prerequisite for achieving military AI superiority.

3. Accelerated Algorithmic Innovation. Algorithmic breakthroughs, exemplified by the DeepSeek R1 series, offer a strategic hedge against hardware restrictions by optimizing performance on accessible hardware.

4. Market Efficiency and Service Model. The transition to "Computing Power as a Service" (算力服务) and the development of platforms for resource matching (e.g., Shenzhen) enhance market efficiency, promoting ubiquitous and standardized access to the "basic public resource" of computing.

5. Green Energy Synergy. The forced integration with green power (绿电) and Electricity-Computing Synergy increases the sustainability of the EPC, securing energy supply and reducing operating costs (e.g., PUE < 1.4).

4.2. Weaknesses and Strategic Risks

1. Core Technology Dependence ("Chokepoint"). The primary strategic vulnerability is the weak technological self-reliance (自主创新能力) in core segments: high-performance chips (GPUs, AI chips), advanced servers, and foundational software ecosystems. This creates a significant risk exposure (风险敞口) to export controls.

2. Insufficient Sectoral Integration. The deep fusion of computing power with traditional sectors (manufacturing, finance, agriculture) is still at an early stage, limiting the full release of its transformative potential across the economy.

3. Regional Homogenization and Inefficiency. A lack of clear differentiated development paths means less-developed regions often blindly copy successful models, leading to duplicate investments and low resource utilization.

4. Immature Institutional Framework. The supporting institutional framework (fiscal, tax, and financial policies) is still under development, resulting in high systemic transaction costs and hindering optimal market stimulation.

   
   

5. Future Scenarios and Proactive Proposals

The EPC project is a long-term strategic endeavor with profound implications for global power dynamics.

Scenario 1: Accelerated Self-Reliance and Dual-Use Development (Technological Re-routing)

Dynamic. Tightened restrictions on advanced hardware force China to rapidly accelerate self-reliance efforts, prioritizing technological re-routing to mitigate external dependencies.

Dual-Use Focus and Counter-Measures:

• Investment in Bypass Technologies. Focus will increase on technologies less reliant on restricted lithography, such as chiplets, advanced packaging, and the open-source RISC-V architecture. This aims to achieve comparable AI performance using accessible hardware, maximizing the efficiency of the 748 EFLOPS.

• Algorithmic Supremacy. The momentum from DeepSeek R1-like innovation will be amplified. Algorithmic efficiency becomes a critical national security asset, enabling high-level AI capabilities (e.g., advanced simulation, autonomous systems) even if hardware power is capped.

  
  

Proactive Proposal (External Intelligence):

• Monitor Critical Breakpoints: Intelligence should focus on tracking patents, research collaboration (e.g., Hami Research Institute) and investment announcements specifically in chiplet technology and RISC-V adoption as these represent strategic bypass mechanisms.

• Flexible Restriction Strategy. Evaluate targeted controls on enabling production tools (e.g., EDA software, metrology equipment) rather than just end-product chips, as these constrain the ability to achieve self-sufficiency in manufacturing.

Scenario 2: EPC as a Tool for Global Geoeconomic Projection

Dynamic. China actively exports its EPC model, leveraging its "Computing Power as a Service" model and its Green Computing advantage to expand influence along the Digital Silk Road.

Geoeconomic Focus and Digital Dependence:

• Integrated Offering. China offers partner countries (Africa, Southeast Asia) comprehensive packages: network infrastructure + computing power (Cloud/AI services). This creates immediate technological dependence and drives the adoption of Chinese standards and interfaces.

• Green Computing Diplomacy. China leverages its low-PUE (e.g., Chengzhi Group PUE < 1.4) and green energy capabilities to position its EPC as the ideal, sustainable solution for developing nations, gaining diplomatic leverage.

  
  

Proactive Proposal (External Intelligence):

• Alternative Digital Offering. Promote international public-private partnerships to offer alternative Cloud/AI/Edge solutions that adhere to stringent Western standards of data security, transparency, and interoperability in strategic markets, directly competing with the Chinese model.

• Standardization and Data Sovereignty: Advocate for and enforce strong international standards and certifications for data sovereignty in cross-border computing, ensuring that developing nations maintain control over their critical information assets against potentially controlled Chinese platforms.

Conclusion

China's Computing Power Economy (EPC, 算力经济) represents far more than a technological upgrade; it is a centralized, state-driven national strategy aimed at securing long-term economic and military supremacy in the age of AI. The sheer scale of its infrastructure, now featuring 10.431 million standard racks and 748 EFLOPS of intelligent computing (Zhìsuàn) , combined with proven GDP spillover effects (0.0426% growth per 1% computing scale increase), confirms the EPC's efficacy as a geoeconomic engine.

The core intelligence takeaway is the strategic focus on resilience and bypass mechanisms. While China faces a critical vulnerability in core components (the "chokepoint" risk on high-performance chips), it is actively mitigating this through:

1. Algorithmic Innovation: Breakthroughs like the DeepSeek R1 series demonstrate the capacity to optimize AI performance even on restricted hardware.

2. State-Led Factor Mobilization: Massive capital injection from tech giants and strategic talent cultivation (e.g., "Computing Power Vouchers") ensure factor readiness.

3. Green Synergy: The Electricity-Computing Synergy (电算协同) provides a secured, lower-cost energy base, turning an environmental challenge into a strategic advantage.

The EPC's future success hinges on executing a strategic leap from infrastructure maturity to application integration. If China can successfully overcome the technology dependence while embedding its ubiquitous "Computing Power as a Service" model both domestically and along the Digital Silk Road, it will solidify its position as the leading global power in the intelligent civilization era. The EPC is thus a key indicator for monitoring China's progress toward technological self-reliance and its capacity for global projection of influence.