The synergy between computing power and electricity drives the demand for green energy; institutions recommend three main strategies.

On March 17, electric power stocks opened lower and then rose, with many hitting the daily limit. By the morning close, the sector had increased by 0.66%, leading industry sectors in gains. Individual stocks such as Huadian Liaoning Energy, Zhaoxin Shares, Xineng Technology, Jiangsu New Energy, Zhejiang New Energy, and Energy-saving Wind Power hit the daily limit. Huadian New Energy rose over 8%, Green Power and Lixin New Energy increased over 5%, and Shimao Energy, Solar Energy, and Xintian Green Energy gained over 3%.

Calculative Power Collaboration Boosts Green Electricity Demand

The 2026 government work report explicitly states: “Implement large-scale intelligent computing clusters, calculative power collaboration, and other new infrastructure projects; strengthen nationwide integrated computing power monitoring and dispatch; support the development of public clouds.” This marks the first time “calculative power collaboration” has been included in the national new infrastructure framework.

As early as 2024, the “Action Plan for Accelerating the Construction of a New Power System (2024–2027)” and the “Special Action Plan for Green and Low-Carbon Data Center Development” were introduced, clearly initiating pilot projects for calculative power collaboration. Notably, by the end of 2025, a preliminary two-way collaborative mechanism between computing power and electricity will be established, with green electricity accounting for over 80% of newly built data center power at national hub nodes.

Against the backdrop of rapid development in AI large models and the digital economy, the explosive growth in computing power demand is significantly driving up data center energy consumption and electricity needs. Data from China Academy of Information and Communications Technology shows that by June 2025, the global computing device capacity reached 4495 EFlops, a 117% increase. Some computing hubs may face power shortages, as existing grid transmission and distribution capacity may struggle to meet the rapidly increasing electricity demand from computing.

Image: Global computing capacity and growth rate, China Academy of Information and Communications Technology

Based on the comparison of power generation and consumption in regions hosting computing hubs in 2025, power supply in Beijing-Tianjin-Hebei, Yangtze River Delta, and Guangdong-Hong Kong-Macao Greater Bay Area all show deficits. Local green electricity supply can no longer meet the needs of computing facilities, with Jiangsu, Zhejiang, and Guangdong experiencing the largest power gaps at 245.8, 234.9, and 249.5 billion kWh respectively.

Data centers are high-energy-consuming industries, with electricity costs being a core part of operational expenses. According to Huachuang Securities, currently, electricity costs account for 56.7% of data center operating costs, making it the largest expense. The low electricity prices offered by green power provide an effective way to reduce costs. As AI-driven data centers expand rapidly, the proportion of intelligent computing power continues to rise. It is expected that by 2026, China’s total computing capacity will reach 767 EFlops, with 73% of it being intelligent computing power.

Three Main Lines of Calculative Power Collaboration

Guosheng Securities notes that recent policies in various computing hub regions are accelerating the implementation of calculative power collaboration. The “East Data, West Computing” project includes eight national computing hub nodes: Beijing-Tianjin-Hebei, Yangtze River Delta, Guangdong-Hong Kong-Macao, Chengdu-Chongqing, and Guizhou, Inner Mongolia, Gansu, Ningxia, with clusters in Zhangjiakou, ecological green integration demonstration zones in the Yangtze River Delta and Wuhu, Shaoguan, Tianfu, Chongqing, Gui’an, Hohhot, Qingyang, and Zhongwei.

Image: Distribution of China’s computing hubs and clusters, International Science and Technology Innovation Center

Main Line 1: Integrated Calculative Power Operators. These companies have the most potential for revaluation under business model upgrades. They possess resources on the power supply side, trading capabilities such as power sales, distribution, and virtual power plants, and are beginning to enter the calculative power leasing and data center services markets. Their revenue models are evolving from traditional “generation volume × electricity price” to “green power supply + energy services + computing services + environmental rights,” potentially offering the greatest profit leap.

Main Line 2: Green Power Operators in Calculative Power Hub Regions. “Calculative power follows electricity” is currently the most mature and visible industry path. With policies from the government and regional authorities requiring higher proportions of clean energy in data centers, the site selection logic has shifted from “near clients and networks” to “prioritizing proximity to low-cost, sustainable, tradable clean energy sources.”

Direct green power connection reduces electricity costs and locks in long-term energy prices. Profitability analysis for data centers must now consider power supply structure, trading capacity, green energy ratio, and peak-shaving ability. Direct green power connection is the most economical mode and has the greatest impact on project returns, especially suitable for western regions rich in wind and solar resources, with lower land and electricity costs, such as Ningxia, Inner Mongolia, Gansu, and Qinghai.

Main Line 3: Grid Upgrades and Intelligent Dispatching. Data centers demand higher standards for cross-regional grid capacity, dispatching, and smart grid technology. Their high, continuous loads require stable power supply, efficient dispatching, and high power quality. Additionally, the distribution of data centers does not always match energy resource locations. Upgrading grid cross-regional capabilities, dispatching, and smart systems—such as microgrid energy storage, virtual power plants, and intelligent grid dispatching—will see significant demand.

In A-shares, driven by multiple favorable factors, green power concept stocks have recently attracted strong investor interest, with share prices rising accordingly. According to Eastmoney Choice data, since the beginning of the year, green power concept stocks have seen a net financing inflow of 9.31 billion yuan, with the latest financing balance at 82.64 billion yuan, up nearly 13% from the start of the year. As of midday March 17, about 90% of these stocks have increased in price this year, with a median rise of 16.5%.

(Article source: Eastmoney Research Center)