The technology rivalry between the United States and China has exposed the pressure facing global companies as Microsoft scales back its operations in China. Microsoft closed at least 15 branches and joint ventures in the country over the past five years but maintains that it has no plans to leave China entirely.
The company’s situation shows that the era in which technology firms could operate at full capacity in both the United States and China has passed. While leaving the Chinese market altogether is not easy commercially or operationally, expanding operations in the country is becoming increasingly difficult because of political and regulatory restrictions.
Pressure from Washington and Beijing
Microsoft’s operations in China are being shaped by two different forms of pressure at the same time. Washington’s export restrictions on advanced chips were converted into a case-by-case licensing system for Nvidia’s H200 model as of January 2026. The arrangement also includes a 25 percent tariff, quantity quotas and mandatory US testing.
The changing rules are creating lasting uncertainty for companies rather than imposing an outright ban. This environment limits Microsoft’s flexibility to expand its artificial intelligence and cloud infrastructure in China.
The other source of pressure in China is the policy of increasing the use of domestic software. A qualification program that directs state-owned companies toward domestic software instead of foreign tools is narrowing Microsoft’s access to the country’s public-sector market.
Microsoft was named in only one of six central government procurement guidelines published between December 2023 and May 2026. In August 2026, Beijing instructed government agencies to switch to Linux-based systems by abandoning their customized version of Windows ahead of schedule.
Microsoft’s model between two markets
Rather than making a definitive choice between the United States and China, Microsoft is retreating into a narrower and more protected operating space. Outside China, the company provides Azure cloud services and Western artificial intelligence models to Chinese companies such as ByteDance and Shein that need infrastructure compliant with global regulations.
This approach allows Microsoft to preserve the connection between China and global technology ecosystems. However, Beijing’s data security regulations and Washington’s US Data Security Program are making cross-border data and cloud traffic more vulnerable.
The absence of a comprehensive bilateral agreement governing this traffic between the two countries also leaves companies more exposed to political decisions. Microsoft is therefore continuing to operate in an area that could face the greatest pressure if tensions escalate.
Hardware is decoupling, but commercial ties continue
The US-China split is becoming particularly clear in hardware and advanced technology. China is increasing its chip production capacity with substantial government support, while Nvidia has announced that its share of the Chinese market has fallen to zero.
The picture is different in the consumer market. Windows continues to account for a large share of desktop internet traffic in China. For this reason, it does not appear easy for Microsoft to eliminate its presence in the end-user market entirely in the short term.
Developments between the two countries point to partial decoupling rather than a complete severing of commercial ties. Licensed trade, equipment sales and the flow of rare earth elements continue, while Beijing is using restrictions on mineral exports as a strategic bargaining tool.
Two separate technological ecosystems are emerging in the digital economy, government systems and advanced computing, but commercial ties are not being completely severed. Companies therefore have to comply with regulations on both sides and continue their operations through limited connections.
Dependence on China increases vulnerability
Under the new system, the companies in a more advantageous position are not those generating the highest revenues from China, but those with a limited presence in the country or the ability to change their production models flexibly. Apple’s ability to shift part of its iPhone production to India within a few quarters is an example of this flexibility.
Advanced chip suppliers and enterprise software companies that derive a significant share of their revenues from sales in China’s domestic market are more exposed. The decline from 62 percent to 52 percent in the share identifying China as the top investment destination shows that companies are reassessing the country’s market.
More Western companies are expected to adopt Microsoft’s controlled withdrawal model in the coming period. The model is based on partially giving up market share in China’s domestic market while preserving cross-border services.
The next 10 years for companies
As long as the US and Chinese governments preserve regulatory and commercial flexibility that they can use when needed, the likelihood of a full-scale rupture will remain low. A complete end to commercial relations in the short term is also considered unlikely.
Over the next 10 years, the central challenge for companies will not be deciding which market to choose, but building resilient structures that can sustain their operations under pressure from the two major economic systems.
