On October 20th, the China Financial 40 Forum (CF40) released its annual report on CF40 research titled "China's New Economic Paradigm" at the 2026 Bund Annual Meeting.

The report shows that in the past five years, the new economy represented by new energy vehicles, clean energy, high-end equipment, and digital technology has rapidly expanded and gradually surpassed the traditional economy represented by real estate at the incremental level. Among them, the technological progress in the field of AI (artificial intelligence) represented by big language models and robots is becoming faster and faster, providing new possibilities for the continuous improvement of productivity, and the related investment demand is beginning to explode. The energy revolution complements the development of AI and advanced manufacturing by reducing electricity costs and expanding clean energy supply.

Guo Kai, the director of the China Institute of Finance Forty, stated that calling it a new paradigm does not necessarily mean that China will experience a strong V-shaped recovery. The mechanisms behind the Chinese economy have undergone significant changes in the past few years, and a new paradigm has been opened.

Guo Kai analyzed that the majority of the contribution of the new economy still comes from exports and investment, with minimal contribution from the demand side. The only continuously growing demand in the new economy comes from new energy vehicles, and it is not new demand. Therefore, despite the significant contribution of the new economy, demand is still insufficient.

In addition to weak demand, there are still many unresolved issues in the current rise of the new economy. For example, can this expansion be sustained when the demand side pull is minimal? Can the rapidly growing AI take over as the next growth engine?

Investment remains the main driving force for new economic growth

China's financial system may also need a process of reshaping. Throughout history, the financial system has been aligned with the economic growth model. The financial system dominated by banks and credit in the past, and the growth model characterized by infrastructure and real estate, are actually highly overlapping urbanization processes, "Guo Kai said in an interview with media including the Economic Observer.

Guo Kai said, "We see that the performance of Chinese new economy enterprises is very different from traditional enterprises. They are highly debt independent and have particularly high R&D investment, which is completely opposite to real estate or infrastructure enterprises. This means that our financing model needs to become a financing model that can manage risks, identify what kind of enterprises can grow better, and start to shift to equity financing in large numbers

According to the sample of CF40 research, enterprises that can represent the new economy are mainly distributed in industries such as computers, machinery and equipment, pharmaceuticals and biology, electronics, power equipment, automobiles, national defense and military industry. These enterprises include both light asset enterprises represented by software and medical equipment, as well as heavy asset enterprises such as semiconductors and batteries that require large-scale capital expenditures.

According to CF40 research, the typical characteristics of new economy enterprises are: placing greater emphasis on research and development expenditures, and creating more revenue and profits with less debt. The report shows that by 2025, the total amount of interest bearing liabilities of all new economy enterprises will only account for 11.9% of the total sample, but the proportion of operating income and net profit will be 17.8% and 29.5% respectively, and the proportion of research and development investment will reach 47.6%. With the increasing proportion of the new economy in the national economy, the growth model of the Chinese economy will gradually change.

And investment has always been the main driving force for the expansion of the new economy. The report shows that since 2018, investment has been the biggest driver of new economic growth in most years. The growth momentum of the new economy peaked around 2022 and has since declined, reflecting the impact of the production capacity cycle in related industries. At the same time, the consumption demand of residents in the new economy is still highly concentrated on new energy vehicles, and a considerable part of this demand is for the replacement of fuel vehicles, rather than truly incremental consumption demand.

Therefore, the sustainability of new economic growth depends on the formation of terminal demand that matches supply capacity. If there is insufficient domestic terminal demand such as household consumption, the production capacity formed by initial investment will be difficult to fully utilize, and the sustainability of subsequent expansion of the new economy will also be constrained. Although the new economy has surpassed real estate in terms of scale and contribution to economic growth, there is still a certain gap in alleviating strong supply and weak demand, increasing household consumption, and promoting domestic circulation.

Bai Chong'en, Dean of the School of Economics and Management at Tsinghua University, stated in an interview with the Economic Observer and other media outlets that it is necessary to address the issue of insufficient total demand, not to lay hidden dangers for long-term development, and to comprehensively and long-term consider and grasp the pace in order to transition to the "new paradigm".

Guo Kai stated that investments that are conducive to medium - and long-term sustainable development, investments that are conducive to reform, and investments that do not increase excessive production capacity are necessary. Investing in people is one of the important focuses. When local governments invest in science and technology innovation, they still need to manage their own hands. If they go too far, they may overreact in risk pricing and industry selection. More importantly, private entities and profit driven entities need to make decisions so that they can better bear risks, identify these industries, and not go too far

Where has AI improved productivity

From a data perspective, the new economy and AI are becoming the main driving factors for investment, and AI is also driving the growth rate of related fields.

The report shows that in the first half of 2026, investment in high-tech industries increased by 4.6% year-on-year, with investment in electronic circuit manufacturing, lithium-ion battery manufacturing, and integrated circuit manufacturing growing by 55.6%, 24.4%, and 8.8% respectively. The development of cutting-edge fields such as AI led to a 15.5% increase in investment in the information service industry. In the first half of 2026, the year-on-year growth rate of AI related imports and exports reached around 60%, contributing more than half of the import and export growth rate.

The report points out that although AI has the embryonic form of a universal technology, there is still a considerable distance to go before it becomes a mature universal technology and widely penetrates into daily economic life. There are four main paths through which AI affects productivity: one is automation, where AI takes over tasks that were originally completed by workers; The second is task complementarity, that is, AI undertakes some of the work processes to improve the efficiency of workers in completing the remaining processes; The third is the deepening of automation, that is, AI further reduces the cost of tasks that have already been completed by machines; The fourth is the creation of new tasks, that is, AI generates new products, positions, and production processes, expanding the scope of producible tasks.

From the perspective of potential demand scale and application scenarios, the future development space of AI in China will not be lower than that in the United States. China has a huge economic scale, manufacturing system, and digital application scenarios, and its top model capabilities are rapidly approaching the international forefront. According to Epoch AI, an artificial intelligence research institution, Chinese models have lagged behind American cutting-edge models by an average of about 7 months since 2023, significantly narrowing the significant generation gap compared to 2023. By 2026, the performance gap between China's top open-source models in mainstream evaluations and the most cutting-edge closed source models in the United States will be quite limited. From the current supply of computing power and the planned investment scale in the future, the gap between China and the United States is still very large.

AI investment is booming, but many people seem to have not yet seen a significant acceleration in productivity. Bai Chong'en believes that when innovation improves people's work or life, it may not always be reflected in GDP (Gross Domestic Product) data. In addition, many applications of AI are still in their early stages, and to realize the value of AI in improving productivity, these applications need to develop better, which is a slow process.

Bai Chong'en said that many people are talking about how AI will affect the organizational structure of enterprises, and many cutting-edge companies are already restructuring their management structures, but most companies have not done so. In addition, a large amount of investment in AI is currently directed towards models and the chips required for model development, with some investments flowing towards the application level. However, the scale of investment in the application level is still far from sufficient.

In the future, I am optimistic about the impact of AI development on productivity, more optimistic about the demand for data center services, but less optimistic about the profitability of model development companies. It is necessary to have a large demand for investment in the future in order to unleash the potential for productivity improvement. ”Bai Chong'en said.

(Author Liu Ying)