XPEV closes below $10: what has changed since 2024

A trade below $10, then a close

The shares first changed hands below $10 during regular trading at 10:52 in New York, after a pre-market trade at $9.99. The last time they traded below $10 during regular trading was 24 September 2024, when they touched $9.95 and then closed at $10.70, 12% higher on the day.

A close below $10 is a separate milestone, and it followed: the shares closed at $9.95, down 1.7% on the day. The last close below $10 had been $9.56 on 23 September 2024. In Hong Kong, where XPeng's shares trade during the Asian day, they closed at HK$39.10 earlier on the same day, their lowest close since September 2024.

XPulse had not identified a confirmed explanation for the move at publication.

A round number, not a verdict

$10 is a threshold people notice, not a measure of the company. Crossing it says nothing by itself about how XPeng's business is doing, or whether the shares are cheap or expensive.

It is still rare ground. XPulse's price record covers every session since the New York listing in August 2020, and before this one the shares had closed below $10 in only two broad episodes. Both came after the fall from their highest close, $72.17 in November 2020.

XPEV in New York, daily closes in XPulse's price record as on file on 28 Sep 2026, every session since the listing on 27 Aug 2020. An episode groups spells below $10 that are less than a month apart; between them the shares closed above $10. The full series is on Stock.

The company at the same price

When the shares last traded here, XPeng's latest results were for the second quarter of 2024. The latest now are for the second quarter of 2026. Side by side, they describe a much larger business with a mixed record on profit and cash.

RMB billion except deliveries and margins; each quarter as XPeng's results release states it, each first half as its interim report does, as on file on 28 Sep 2026. Every quarter since 2024 is on Financials.

Scale, margins and reach are the clear gains: more than three times the deliveries, a vehicle margin — the share of car-sales revenue left after the cost of building the cars — almost twice as high, and far more markets. XPeng says its footprint now spans 79 countries and regions. By XPulse's count it is on sale in 73 markets and has distributors signed in six more; when the shares last traded here, it was on sale in 22.

Most of the gain in profit has gone into higher costs. Gross profit, what is left from all sales after their direct costs, rose by about RMB2.95 billion a quarter. Research and development, selling and administrative costs rose by RMB2.37 billion together, so the operating loss narrowed by less than RMB0.5 billion. Below that line, interest, investment returns, exchange rates and tax added RMB0.32 billion two years ago and took away RMB0.19 billion this time, leaving the net loss slightly larger.

The one profitable quarter on file shows how thin the margin still is. XPeng earned RMB383 million in the fourth quarter of 2025, but gross profit less research, selling and administrative costs was a loss of RMB0.92 billion that quarter; other operating income and non-operating gains turned it into a profit.

Cash has followed the losses rather than the scale: after bringing in RMB8.26 billion over 2025, operations used more in the first half of 2026 than two years earlier, mostly paying suppliers and building up inventory.

The direction of sales has turned too. In August 2024, XPeng's guidance meant third-quarter deliveries 2.5% to 12.5% higher than a year earlier; this August it meant a change of −0.9% to +4.3%. Deliveries for January to August were 10.5% below the same months of 2025: 21% fewer in China, while XPulse's estimate of deliveries outside China nearly doubled.

The same share price also means a similar market value for XPeng's shares, because their number has barely moved. At $10 an ADS, each equal to two shares, the 1,916 million shares in issue at the end of August 2026 are worth about $9.6 billion; the 1,899 million at the end of September 2024 were worth about $9.5 billion. That value now sits on revenue of RMB76.72 billion in 2025, against RMB30.68 billion in 2023, the latest full year known then. Whether that makes the shares cheap depends on whether deliveries grow again and losses and cash use narrow.

The market around XPeng

XPeng's slowdown at home is part of a wider one. China's passenger-car association (CPCA) counted domestic retail sales of passenger cars 23.6% lower in August than a year earlier, and of new-energy passenger cars 10.1% lower. Manufacturers' sales, which include exports, tell a different story: figures published by the industry ministry (MIIT) show new-energy vehicle sales up 17.8% in August, and their exports up 134.6%.

Buyers at home also pay more tax this year. In 2024 and 2025 new-energy passenger cars were exempt from purchase tax, up to RMB30,000 a car; in 2026 and 2027 the tax is halved, with the saving capped at RMB15,000. Costs may be rising as well: TrendForce expected contract prices for conventional DRAM memory chips to rise 58% to 63% in the second quarter of 2026 alone, though XPeng has not said how much this adds to its own costs.

Other Chinese carmakers' shares have fallen too. At the Hong Kong close on 28 September, NIO and Li Auto were less than 2% above their lowest close of the past year, and BYD and Leapmotor within 8%. XPeng's fall this year is the deepest of the five: 51% since the start of 2026, against 18% to 31% for the others.

These conditions are background. None of them establishes why the shares fell on this particular day.

What XPulse's options record shows

XPulse also keeps a daily record of options on XPeng's shares: contracts giving the right to buy the shares at a set price (calls) or to sell them (puts). Among the contracts still open in Hong Kong, calls have become far more dominant.

Open contracts at each market's latest close on file on 28 Sep 2026: HKEX's daily report for Hong Kong, OCC for New York. The daily record is on Options.

Hong Kong's ratio of 0.41 is the lowest since XPulse's record began on 2 October 2025. Open calls have roughly tripled since then, while open puts are about 7% fewer. New York's ratio, 0.28, is also close to its lowest of the past year.

These are positions still open at the close, not that day's purchases, and they do not show who holds them or why: a call can be a bet on a rise, a hedge, one leg of a wider strategy, or written by someone expecting no rise at all. Expiry matters too: 36% of Hong Kong's open calls expire on 29 September, so the ratio may move sharply as they lapse or are rolled into later months.

Trading in the Hong Kong shares has been quiet. In the three sessions to 28 September, 9868 averaged 5.39 million shares a session, 66% below its average over the 100 sessions to 31 August, and XPulse's volume meter puts the last 20 sessions 47% below it. Low volume by itself says nothing about confidence or about pressure to buy or sell.

What the next figures will show

The first come on 1 October, when XPeng usually publishes its September deliveries. To reach the bottom of its third-quarter guidance it needs 37,866 cars in September, after 38,027 in July and 39,107 in August. September 2025 brought 41,581, so the month will also show whether deliveries are still below a year earlier.

The third-quarter results follow in November, on a date XPeng has not yet announced. They will show whether the higher margins held while volumes stalled, and whether the cash position kept falling from RMB42.09 billion at the end of March and RMB40.48 billion at the end of June. The quarterly release carries no cash-flow statement, so the full account of cash use waits for the annual report.