Profits at onshore-listed Chinese companies rose 25.7% in the three months to June, the fastest pace in nearly five years, according to data cited in market reports. The headline number initially appeared to confirm that the earnings season was strong. But the market response told a different story: the CSI 300, the main yardstick for Chinese shares, has fallen about 9% since the end of June, while the technology-heavy Star 50 Index has plunged 29% over the same period.
The divergence between record-breaking profit growth and falling stock prices marks a critical shift in investor sentiment. For much of the past year, Chinese equities, particularly science and technology stocks, rose on expectations that artificial intelligence would drive the next wave of corporate expansion. The Star 50 had climbed 76% in the prior quarter, making it one of the best-performing indexes in the world. Once earnings were reported, however, many investors used the strength in profits to sell into rallies, treating rising AI-related capital expenditure as a cost rather than a promise of future earnings.
Why the headline number overstated strength
A closer look at the numbers shows that the 25.7% increase was extraordinarily narrow. UBS Securities estimates that ChiNext, the Shenzhen-based growth board, posted profit growth of 42% in the same quarter. On the Shanghai STAR Market, which hosts semiconductor and advanced-technology companies, profit growth reached 370%. These compare with much slower growth on the main board. The staggering gap demonstrates how dependent the aggregate result was on a small cluster of AI-linked companies.
That concentration has important implications. When a market measures aggregate earnings growth, outliers can obscure underlying weakness. If investors exclude AI hardware and chip producers, the profit picture in China is considerably less bright. Consumer companies, property developers and traditional manufacturers are still facing sluggish demand and pricing pressure. The impressive aggregate figure therefore masked a bifurcated economy: a vibrant technology sector investing aggressively in AI and a domestic economy that has not yet recovered its pre-2021 momentum.
Indeed, the market sold off precisely because much of the optimism was already in the price. A typical AI-linked stock had rallied sharply in the months before the earnings season. When companies delivered strong numbers, there was little additional good news to push shares higher. Attention shifted instead to the balance-sheet damage caused by massive spending on data centers, advanced chips and cloud infrastructure.
Alibaba and Tencent show the cost of AI
The clearest examples of this shift are Alibaba and Tencent. Alibaba shares fell in Hong Kong after the company reported higher revenue but sharply lower profit. Alibaba attributed the decline to the cost of AI projects and computing infrastructure, areas it considers critical to future growth. To fund those ambitions, the company is raising $10.2 billion through convertible notes. The proceeds will be used for similar investments. The move strengthens Alibaba’s ability to compete in AI, but it also creates dilution and increases fixed costs.
Tencent took an even more aggressive approach. The company more than doubled its AI-related spending in the second quarter, a strategic decision that weighed on its share price. Capital expenditure jumped 176% to 52.8 billion yuan, while free cash flow turned negative by 13.8 billion yuan. Negative free cash flow is a conspicuous change for a company that historically generated enormous cash surpluses from gaming, social media and online advertising.
“Strong numbers no longer work for tech,” said Vey-Sern Ling, a managing director at Union Bancaire Privee. He cited uncertainty over AI investment, unclear return on investment and rising financing costs. Ling’s comment captures the