China’s CSI 300 Hits a Bear Market on a Growth Beat

China’s CSI 300 index fell as much as 1.7% on Thursday and 20% below its June peak, crossing into a bear market. The drop extended a run of 10 consecutive weekly losses, the longest streak since the gauge launched in April 2005, and left it 2.6% lower for the week, according to LSEG data.

Three days earlier the National Bureau of Statistics said output grew 4.5 percent in the third quarter, inside the 4.5% to 5% annual target. The stock market did not treat that print as a floor.

A Record 10-Week Slide Finally Tags 20%

A 20% fall from a recent high is the usual line for a bear market. The CSI 300, which tracks the 300 largest and most liquid A-shares in Shanghai and Shenzhen, spent the summer and early autumn getting there the slow way: one down week after another, rather than a 2015-style air pocket.

Hong Kong’s gauge of big Chinese firms dropped as much as 3% in the same session. Optical names that had led the spring AI run, including Eoptolink Technology and Zhongji Innolight, were among the weakest mainland stocks. The CSI 300 was already at a 14-month low on Wednesday after Contemporary Amperex Technology, better known as CATL, slumped about 12% in Shenzhen.

THE PATH TO 20%

1. April 8, 2005: China Securities Index Company launches the CSI 300, set to a base of 1,000 on December 31, 2004.
2. June 2026: The index peaks as AI hardware and battery shares lead the tape.
3. July to mid-October 2026: The CSI 300 posts 10 consecutive weekly losses, the longest such run since the launch.
4. October 19, 2026: The statistics bureau reports third-quarter GDP up 4.5% and year-to-date growth of 4.6%.
5. October 20, 2026: CATL files third-quarter results that trail the sales jump and miss analyst profit estimates.
6. October 22, 2026: The CSI 300 falls as much as 1.7%, 20% below the June peak.

Motilal Oswal noted that the STAR 50, a proxy for Shanghai’s tech board, was down nearly 40% in the current half-year, and that more than 100 Chinese tech stocks had lost at least half their value from their peaks. The blue-chip index is catching up with a drawdown that growth boards already took.

Q3 Growth of 4.5% Left Stimulus on a Tight Leash

The second quarter’s 4.3% rise was the slowest in more than three years. The third quarter picked up to 4.5%, or 1.2% from the prior quarter, and the first nine months came in at 4.6% on output of 107,107.1 billion yuan (about $15.94 trillion). Mao Shengyong, deputy head of the statistics bureau, said that addition of 5.7 trillion yuan in output was the largest for the same stretch in five years, and that new growth drivers accounted for more than 40% of the expansion.

September’s trade figures were the cleanest part of the report. Mainland exports rose 29.6% from a year earlier to 3,034.4 billion yuan, and total goods trade for the month rose 27.5%. Industrial output grew 5.5% in September and 5.3% in the first three quarters, with equipment manufacturing up 10.2% and high-tech manufacturing up 14.3%. Lithium-ion battery output jumped 39.5%.

Home demand did not match that pace. Retail sales of consumer goods rose 1.1% in September and 1.1% in the first nine months. Fixed-asset investment fell 7.2%. Real estate development investment dropped 20.6%, while the floor space of newly built commercial buildings sold fell 13.2% and the value of those sales fell 13.9%. Online second-hand housing transactions, by floor space, rose 10.7%, a sign that turnover in existing homes is firmer than the new-build market.

Profits at industrial firms above designated size were still up 15.7% in the first eight months, at 5,272.0 billion yuan. That YTD print did not stop the tape from treating listed-company margins as the live risk. Because year-to-date GDP sits inside the official target, Beijing can keep stimulus targeted and incremental, which is the policy mix that has so far failed to end the weekly grind.

OFFICIAL PRINTS VERSUS THE CSI 300

The surveyed urban unemployment rate was 5.2% in September, and consumer prices were up 0.9% year on year. Those are not crash readings. They also are not the prints that refill a stock market whose buyers are mostly households.

CATL’s Profit Rise Trailed a Fast Sales Jump

CATL is the world’s largest EV battery maker and a top CSI 300 holding. Its Shenzhen filing showed third-quarter profit of 21.25 billion yuan, up 14.54% from a year earlier, while revenue rose 58.58% to 165.22 billion yuan. Profit fell 5.77% from 22.55 billion yuan in the second quarter. LSEG estimates had pointed to profit growth of about 41.2%.

For the first nine months, net profit rose 31.60% to 64.53 billion yuan and revenue rose 56.19% to 442.14 billion yuan. Gross margin for that stretch was 22.72%, down from 25.31% a year earlier. On its earnings call the company pointed to product mix, a consumption tax on lithium batteries restored on September 1 at a 2% transitional rate, and smaller export tax rebates. Shenzhen-listed shares fell about 12% on Wednesday; the Hong Kong line dropped more than 10%.

CATL still took 39.4% of global EV battery installations in the first eight months, up from 37.7% a year earlier, with BYD at 15.1%, according to SNE Research. Scale is not the dispute. The dispute is how much of that scale reaches earnings after years of extra plant and price cutting, the same overbuild Jordan Cvetanovski, chairman and chief investment officer at Pella Funds, puts at the centre of the equity problem.

I think China is really struggling with reining in the problems that they’ve had in the past from the overbuild of capacity and other things.

Jordan Cvetanovski, chairman and CIO, Pella Funds

He still sees select battery names as usable. The market’s immediate answer was to knock the national champion and, with it, an index that no longer looks like a bank-and-liquor tape. China Securities Index Company’s end-September factsheet listed Zhongji Innolight as the top CSI 300 weight at 3.64%, ahead of Kweichow Moutai at 3.18% and CATL at 3.11%. When those optical and battery weights miss, the “blue chip” index misses with them.

Chinese Households Added 9.39 Trillion Yuan to Deposits

People’s Bank of China figures show household deposits rose 9.39 trillion yuan in the first three quarters. That was a slower add than the 12.73 trillion yuan gain a year earlier, and still a huge cash pile sitting off the exchange. Total yuan deposits rose 20.76 trillion yuan. Broad money, M2, was up 7.2%; narrow money, M1, was up 3.2%. Brokerage notes on the September breakdown said new money from households and firms tilted toward time deposits after the equity slide.

HSBC data put retail investors at about 90% of daily mainland turnover, against about 20% to 25% on the New York Stock Exchange. Shaun Rein, managing director at China Market Research Group, said that mix makes selling faster in a confidence scare, and that households are leaning into wealth preservation after the property slump.

WHERE THE HOUSEHOLD CASH WENT

• Bank books: Household deposits still rose 9.39 trillion yuan in the first three quarters, even after slowing from 12.73 trillion yuan a year earlier.
• Time deposits: September’s split, in brokerage read-throughs of the central bank data, favoured locked-up savings over demand deposits and wealth products.
• Property share: Goldman Sachs estimates household assets near 730 trillion yuan in the first quarter, with housing down to 52% from 67% at the 2021 peak and cash and deposits up to 25% from 16%.
• Direct stocks: The same Goldman mix puts household equity at about 6%, with only about 25% of adults in the market, against housing ownership above 90%.

That is why a 10-week grind is more damaging here than a two-day crash that then bounces. The people who set the price are the same people who can park the next yuan in a deposit and wait.

It’s a crisis of confidence right now not a crisis in the economy.

Shaun Rein, managing director, China Market Research Group

Rein still calls the sell-off excessive, citing resilient exports and firmer property in the largest cities, and he expects a strong rebound in Chinese equities in 2027. He also expects more selling first, from geopolitics, worry over an AI bubble, and weak consumer confidence.

What Earlier CSI 300 Bears Cost Investors?

This is not the first time the index has given back a fifth of its value. Calendar-year losses of 25.31% in 2018 and 21.63% in 2022 already sit in the record. The 2021-22 stretch, as Invesco measured it in 2022, ran 64 weeks and cut about 33%, against an average CSI 300 bear from 2005 through 2018 of about 40% over about 58 weeks. A 20% tag is the start of that definition, not the historical midpoint.

The two years before this slide were the easy comparison. The CSI 300 rose 14.68% in 2024 and 17.66% in 2025. Through October 21 it was down 11.19% in 2026. Those back-to-back gains pulled in households that had just watched property stop compounding. The 10 consecutive weekly losses are how that group leaves.

CSI 300 CALENDAR YEARS

Authorities have already tried to slow the selling with state-linked buying. Onshore tape-watchers said E Fund’s CSI 300 exchange-traded fund took in more than 20 billion yuan over four days even as the index kept falling. Cash arriving into a falling benchmark is the national team’s usual footprint. It has not, on this run, been a weekly close in the green.

Index ETFs Took Cash While Prices Kept Falling

Foreign desks were already light, so systematic selling at the 20% line thins the bid just when local funds need a counterparty. Daniel Gerard, senior multi-asset strategist at State Street Markets, argues that China’s AI build-out can still pull global money back before listed earnings turn. That is a 2027-shaped claim, and it sits beside Cvetanovski’s warning that even globally competitive plants are living with too much capacity.

The official economy, as printed on October 19, has a floor: 4.5% growth, 29.6% export growth, and industrial output that is still expanding. The CSI 300 is a different machine. Ninety percent of its daily volume is retail, its largest weights now include the optical and battery names that led the spring rally, and household deposits are still rising by the trillion. Until those buyers stop choosing time deposits over the next downtick, a 20% line is a description of the last 10 consecutive weeks, not a timer for the next rally.

Rein’s 2027 rebound only works if that confidence switch flips. Thursday’s 1.7% drop said it has not flipped yet.

Disclaimer: This article is news reporting and market analysis for information only. It is not investment advice, a recommendation to buy or sell any security, index product, or deposit, or a forecast of returns on the CSI 300, CATL, or any other listed name. Readers should consult a licensed financial adviser or other qualified professional who knows their objectives and risk limits before making any investment decision. Index levels, company figures, and policy settings are those published by the cited official sources as of the article’s date and may change with later trading, filings, and data revisions.