CITIC Securities: Computing-Power Synergy Key to High-Quality Domestic Computing Power Expansion
CITIC Securities has identified the synergy between computing power and electricity as a critical infrastructure for expanding domestic computing capacity in China. The firm argues that without this synergy, high-quality development of domestic computing power will be hindered. China's energy system offers substantial, scalable potential for domestic computing power, bypassing grid connection constraints and enabling more efficient conversion of capital expenditure into actual computing power. According to guidance from the National Energy Administration, China's data centers are projected to consume 800 billion kilowatt-hours of electricity by 2030. Based on this, CITIC Securities forecasts an incremental capacity of 78.5 gigawatts for Chinese data centers during the "15th Five-Year Plan" period (2026-2030), requiring approximately 10.7 trillion yuan in capital expenditure. The synergy between computing and power, encompassing both inside and outside data center campuses, is predicted to attract an investment of around 2 trillion yuan. This sector is expected to enter an accelerated growth phase, with a compound annual growth rate (CAGR) of approximately 42% during the "15th Five-Year Plan" period.
The report highlights a strategic imperative for China to enhance its domestic computing power by integrating it with its energy infrastructure. This approach aims to overcome limitations imposed by traditional grid constraints and optimize capital deployment for computing resources. The projected significant investment in data center capacity and the associated computing-power synergy infrastructure underscores a national strategy to bolster technological self-sufficiency. This initiative, driven by projected energy consumption targets and substantial capital allocation, suggests a focus on building resilient and scalable digital infrastructure. The emphasis on synergy indicates a recognition that future advancements in computing, particularly with the rise of AI, will be fundamentally constrained by energy availability and efficient distribution, prompting a proactive, integrated planning approach.
AI-generated to prompt reflection — not editorial opinion, not advice, not a statement of fact. How this works.