Waves of industrial upgrade: The new 'new three' and the global story of Chinese innovation

Editor's note: Li Lun, a special commentator for CGTN, is an assistant professor of economics at Peking University. The article reflects the author's opinions and not necessarily the views of CGTN.

In July, 1978, Hong Kong businessman Zhang Zimi arrived in Dongguan in south China with HK$2 million ($255,000), a sample handbag, design drawings and raw materials. His partners in the city supplied the factory and workers. Together, they founded the Taiping Handbag Factory, which became the first company to manufacture processed goods on the Chinese mainland and export them to Hong Kong. Ideas and technology came in; labor-intensive goods went out.

Nearly half a century later, China is supplying large artificial intelligence (AI) models to global users, manufacturing industrial robots for use around the world and licensing drug candidates for multinational pharmaceutical companies. What has changed is not simply the value of what China supplies, but the "spillover" benefits China creates: The robots, AI models and drug candidates it provides are inputs that other firms are integrating, adapting or developing further, generating gains across industries and borders.

In 1978, China's total merchandise trade was worth only $20.6 billion, less than 1% of world trade. Reform and opening up transformed workshops into an expanding manufacturing base. Over time, clothing, furniture and household appliances emerged as the "old three" of Chinese foreign trade, and export production became one of the foundations of China's early industrialization and integration into the world economy.

The significance of trade extended beyond the products themselves. Fulfilling large and recurring overseas orders required factories to establish quality control, reliable delivery schedules and close coordination with suppliers; it also supported workforce training, logistics networks and industrial capital accumulation.

Export earnings supplied the foreign exchange needed to import technology and equipment, while expanding trade and foreign investment brought capital, customers and production standards into increasingly dense industrial clusters. The trained workers, specialized suppliers and production routines assembled around the "old three" formed part of the foundation on which Chinese firms later moved into more complex manufacturing.

From that foundation came a second stage of upgrading into industries requiring sustained engineering, capital investment and complex supply-chain coordination. This shift was embodied in the "new three" – electric passenger vehicles, lithium-ion batteries and solar cells – whose exports reached nearly 1.3 trillion yuan ($188.50 billion) in 2025, about three and a half times the 2020 level. They combined greater technological sophistication with manufacturing scale, broadening global access to clean technology.

The International Energy Agency reported that economies of scale and continuous innovation in China contributed to a decline of more than 80% in solar PV costs, helping solar become the most affordable electricity-generation technology in many parts of the world. Yet the main channel through which the "new three" reached overseas economies remained the manufactured product itself: a vehicle, battery or solar cell purchased and put to use.

The new "new three" – AI, industrial robotics and innovative drugs – carry the progression further while changing how Chinese innovation enters other economies. AI models can be adapted to create new services, industrial robots can be incorporated into production systems, and licensed drug candidates can be advanced through research, clinical trials and commercialization. Each is both an output and an input into further work by firms and researchers abroad.

In the first half of 2026, China exported industrial robots worth 6.29 billion yuan, up 18.6% year on year. From July 13 to 19, Chinese-developed models processed 36.11 trillion tokens on US model routing service OpenRouter and ranked first by country of origin for 12 straight week; and Chinese drug developers announced out-licensing deals with a reported headline value of about $110 billion. Together, these developments trace China's movement from contract production toward becoming a source of foundational technologies and scientific discoveries through advanced manufacturing.

China's new "new three" emerge from a common sequence: Investment in research and talent generates ideas; dense industrial networks convert prototypes into dependable products; and predictable, open rules mobilize private and foreign capital for projects with long development cycles.

The sequence begins with research and talent. In 2025, China spent 3.93 trillion yuan on research and development (R&D). Its universities produce more than five million STEM graduates each year. DeepSeek founder Liang Wenfeng and Moonshot AI founder Yang Zhilin are the products of Chinese universities. These trajectories point to China's deepening talent base that spans fundamental science, engineering and technological entrepreneurship.

But building the talent base is only the beginning. Turning a lab prototype into a reliable, affordable product is typically a bottleneck in industrial innovation. Software, sensors, motors and precision components must be integrated and tested repeatedly in actual workplaces. China's connected industrial ecosystem helps firms overcome that bottleneck. Its industries span every category in the United Nations industrial classification and include more than 200 mature industrial clusters.

Proximity among component suppliers, systems integrators and factories enables rapid testing and redesign, while China's large manufacturing base provides working production lines and a broad pool of early customers. This feedback loop turns technical advances into scalable products faster and at lower cost.

Commercialization also requires financing and a stable policy environment. China's first fundamental law dedicated to the private sector covers fair competition, financing and technological innovation. An action plan released in June outlines measures for wider opening of the pharmaceutical and biotechnology sectors, support for foreign R&D centers and reinvestment, and equal access to government support policies. Together, these measures provide greater certainty for long-term investment. In the first half of 2026, utilized foreign direct investment in high-tech industries rose 33.2%.

Consistent, transparent and predictable policymaking, reinforced by long-term planning, reduces the risk of abrupt reversals and gives investors greater confidence to commit to projects that require years of research and clinical trials.

In 2024, then US Treasury Secretary Janet Yellen argued that China's "non-market policies and practices" were producing "overcapacity" that threatened firms and workers in other countries. More recently, a similar charge includes emerging industries as victims as well. A study released by the European Parliament's trade committee listed robotics as one of four case studies of the "concrete risks from overcapacity" to European manufacturing.

However, applying this logic to AI, industrial robots and innovative drugs misread how these industries actually generate value. As productive inputs, the new "new three" help firms across sectors cut costs, automate routine tasks and accelerate research. They make services, production methods and treatments that were previously unaffordable or unattainable commercially viable, expanding demand rather than merely competing for a share of a fixed market. In this sense, the new "new three" have the potential to raise productivity and consumer welfare, generating mutual gains across firms, sectors and countries.

Evidence across the three industries shows two routes to shared gains: Lowering costs and raising productivity for producers, and providing new goods, services and treatments for consumers. The Organisation for Economic Co-operation and Development reported that its quality-adjusted price index for large language models fell nearly 80% between January 2024 and April 2026, and that open-source development significantly lowers usage costs.

Research covering 17 countries found that robot adoption raised labor productivity and total factor productivity while reducing output prices. US drugmaker Pfizer agreed to pay $1.25 billion upfront for overseas rights to a Chinese-developed experimental cancer therapy.

The common thread in all these is that the benefits extend well beyond the original producer: Lower-cost AI reduces barriers to new services, robots raise efficiency, and drug licensing connects discoveries to additional investment, clinical trials and manufacturing capacity. Each channel expands what downstream firms can do and what consumers can ultimately access.

In the 2010s, some economists coined the term "China shock" to describe local labor-market disruptions associated with the surge in Chinese imports after China joined the World Trade Organization. However, by raising productivity and lowering costs, Chinese innovation expands the range of goods and services that other economies can also produce and their people can consume. Stronger competition may accompany this process, but it does not merely reallocate existing demand – it creates new markets, broadens access and improves welfare.

At the 2026 World Artificial Intelligence Conference in Shanghai last month, Chinese President Xi Jinping expressed China's principle: "AI development should not be a solo performance by a single country, but a symphony of international cooperation."

Over the course of China's reform and opening up, the evolution from the "old three" to the "new three" and then to the new "new three" is more than a shift in export categories. It traces China's movement from producing reliably for global markets, through scaling advanced green technologies, to generating research-intensive platforms, equipment and discoveries that can expand productive possibilities abroad. The significance of this latest transition will be measured not only by what China sells, but by what others can create from them, produce and improve, a more substantive expression of shared development.

(If you want to contribute and have specific expertise, please contact us at opinions@cgtn.com. Follow @thouse_opinions on X to discover the latest commentaries in the CGTN Opinion Section.)