A billion here, a billion there, and pretty soon you're talking real money, the old joke goes. For Nvidia (NVDA), a billion or two in guidance when it reports fiscal third-quarter results Wednesday is no joke because it might help determine whether the stock's rally can roll on after recent record highs.
Until last quarter when its outlook disappointed investors, the AI chip giant exceeded analysts' expectations by $1.5 billion to $2 billion in quarterly revenue guidance. This regular pattern of beating Wall Street's estimates suggests the market for Nvidia's AI chips continues to grow even faster than analysts can project. Chip stocks, not including Nvidia, flagged recently amid concerns that demand could be slowing, but Nvidia's guidance could be influential. Another big jump would likely signal that Wall Street underestimates the sector's growth potential, and with it, Nvidia's profit potential.
Some have questioned how long the AI-spending spree from hyperscalers can be sustained, but there haven't been any signs of a let up, at least not yet. That's good news for Nvidia and suggests it could surprise analysts again when it reports Wednesday after the close. Nvidia's earnings come soon after the stock was added to the Dow Jones Industrial Average® ($DJI), replacing rival Intel (INTC) in a symbolic passing of the torch.
Spending spree continues
So-called hyperscalers that spend billions on AI for their data center, internet advertising, and search businesses recently unveiled earnings and guidance that showed no hint of plans to slow down their capital expenditures. Meta Platforms (META), Amazon (AMZN), Alphabet (GOOGL), and others keep snapping up AI chips from Nvidia and others about as fast as they're made, and that appears likely to carry on into next year.
"We continue to expect significant capital expenditures growth in 2025," Meta said in its earnings press release late last month, raising its capital expenditure projection to a range between $38 billion and $40 billion.
When Nvidia reports, analysts and investors will closely watch to see if hyperscalers' spending plans filter through into the type of $1 or $2 billion revenue guidance beats that the market has come to expect.
"Nvidia's revenue guidance beat analysts' estimates by 'only' $500 million last quarter, when investors had become accustomed to a $1.5 to $2.0 billion beat, which is the amount some 'whisper numbers' suggested." said Nathan Peterson, director of derivatives analysis at the Schwab Center for Financial Research. "The AI story, which is the main driver of the chip narrative, looks really healthy. We haven't yet seen any indication of slowdown in spending from hyperscalers, which suggests demand remains strong."
Evidence in favor of AI growth mounted even after the mega-cap's earnings reports. For instance, Palantir (PLTR), which reported earlier this month, said its quarter was "driven by unrelenting AI demand that won't slow down." In a letter to shareholders, Palantir expanded on that, saying it's seen "an unwavering demand for the most advanced artificial intelligence technologies from our U.S. government and commercial customers."
This is evidence that AI is bearing fruit for companies that offer technologies driven by it. In other words, return on AI investments is something market participants want to see from the companies spending big on AI.
In addition, recent moves by Microsoft (MSFT) and Amazon to drive their data centers with power from nuclear energy firms are anecdotal indications of customer AI demand. The thought on Wall Street is that these giants wouldn't go to the trouble of partnering with nuclear plants if the underlying demand wasn't there. AI requires massive power inputs.
Industry data points to strength
Monitoring hyperscaler spending plans is one way to get a sense of what investors might hear from Nvidia. Another is to check the performance of Taiwan Semiconductor Manufacturing (TSM), which builds chips for Nvidia and other firms at its foundries.
Taiwan Semiconductor revenue in U.S. dollars climbed 36% year over year in the third quarter to $23.5 billion, and it guided for fourth-quarter revenue of between $26.1 billion and $26.9 billion. The company said in its earnings call that it saw "extremely robust AI-related demand," Barron's reported, and it now expects the revenue contribution percentage from AI processors to more than triple this year.
Beyond TSM, Palantir, and recent mega-cap earnings and spending guidance, statements from Nvidia and other AI-related firms also weigh in favor of AI bulls. Nvidia CEO Jensen Huang told CNBC last month that demand is "insane" for Nvidia's new Blackwell AI chip, which companies like OpenAI, Microsoft, Meta, and others are using to build AI data centers. "Everybody wants to have the most, and everybody wants to be first," Huang said during the interview.
Blackwell costs $30,000 to $40,000 per unit.
"Bottom line, the AI story looks intact," Schwab's Peterson said. "Palantir and other secondary beneficiaries are validating the story."
Nvidia caught the AI wave and rode it to share gains of more than 900% since the beginning of last year. It's up more than 200% so far in 2024. Fiscal second-quarter revenue, which Nvidia reported in late August, rose 122% year over year to $30.04 billion, and Nvidia guided for revenue of $32.5 billion in its fiscal third quarter that it will report on Wednesday.
Whatever Nvidia's third-quarter revenue ends up being Wednesday (and analysts now expect $32.94 billion, according to Yahoo Finance), it's fourth-quarter revenue guidance probably needs to be above expectations to appease investors. The average estimate on Wall Street now is around $36.7 billion. Last quarter, Nvidia only guided $500 million above estimates and the stock sold off modestly, so investors would likely want to see a larger guidance beat than that.
"The forward-looking revenue guidance in relation to analyst expectations is likely the best gauge of current demand for Nvidia," Schwab's Peterson said. "While reported and guided EPS metrics are important, you can't fake revenue. "If Nvidia provides in-line or slightly above fourth-quarter revenue guidance, it will likely be poorly received by investors."
Margins under microscope
Nvidia's impressive performance story stepped back slightly in the second quarter when gross margin on a generally accepted accounting principles (GAAP) basis slipped sequentially to 75.1% from 78.4%. It was still up from a year earlier but raised questions about possible competitive pressures. Other firms like Advanced Micro Devices (AMD) and Broadcom (AVGO) have their own AI offerings, which could bring pressure on Nvidia to reduce prices, perhaps hitting margins. And the heavy capital spending to bring Blackwell to market could be another margin challenge.
Last time out, Nvidia said it expects fiscal third-quarter gross margins in the 74.4% to 75% area, plus or minus 50 basis points, and in the mid-70% range for the full fiscal 2025 year. Operating expenses are expected to grow in the mid-to-upper 40% range in fiscal 2025. Keep a close eye on margin and expenses guidance for any changes Wednesday because these could help shape the earnings perspective heading into coming quarters.
Another element to monitor beyond guidance is any update on the rollout of Blackwell architecture, which the company calls "a new class of AI superchip" and has attracted interest from Amazon, Alphabet, Oracle (ORCL), and Microsoft. The chips cost about the peak price of the company's H100 line.
Getting Blackwell to market hasn't been without headaches. Last month, Huang noted that Nvidia had fixed a design flaw, Reuters reported. The launch was delayed earlier this year after Nvidia announced the new chip in March. Blackwell samples were shipping last quarter, the company said, and Blackwell is expected to generate several billion dollars in revenue in the company's current fiscal fourth quarter, Nvidia said on its last earnings call.
"Demand for Blackwell platforms is well above supply, and we expect this to continue into next year," said Colette Kress, CFO at Nvidia, in a transcript of the call. Production was expected to start this quarter. Any updates on the production and rollout could get attention Wednesday.
For the quarter, analysts expect Nvidia to report earnings per share (EPS) of $0.74, up from $0.37 a year ago, according to Yahoo Finance. That's a 100% increase. Revenue is seen at $32.94 billion, up 82% year over year.
Data center results will likely be the cornerstone after second-quarter revenue from that segment hit a record $26.3 billion, up 16% from the previous quarter and up 154% from a year earlier. Both of those gains were sequentially lower, suggesting Nvidia may be running into the so-called law of large numbers. Even so, its growth surpasses competitors.
While the data center is the main earnings driver, investors might want to monitor other segments like automotive, gaming, and professional visualization. All these rose year over year in the second quarter.
One challenge for Nvidia is China. Earlier this year, a possible new China problem arose for Nvidia when The Wall Street Journal reported that Chinese firm Huawei Technologies is close to introducing a new AI chip comparable to Nvidia's H100. That's a chip the U.S. government prohibits Nvidia from selling to China. Last month, a media report said Huawei is sending out samples of that chip to large Chinese internet firms that are also major Nvidia customers.
The China challenge potentially got bigger on November 5 with the election of Donald Trump. His promise of even heavier trade barriers against China are a wild card for the entire technology sector, though President Biden's administration has also been tough with tariff policy. The impact could be a discussion item on Wednesday's Nvidia call.
Depending on Nvidia's results, the semiconductor sector could move up or down.
"Sentiment around Nvidia leads semiconductors," Schwab's Peterson said. "AI is a primary growth driver for the chip space, and Nvidia is the obvious leader. I suspect semis may undergo somewhat of a valuation reset if and when mega-cap tech guides capital expenditures lower year over year, because this may send a signal to the markets that a near-term saturation point has been reached. Until then, we assume the AI infrastructure build-out phase still has legs."
The information provided here is for general informational purposes only and should not be considered an individualized recommendation or personalized investment advice. The investment strategies mentioned here may not be suitable for everyone. Each investor needs to review an investment strategy for his or her own particular situation before making any investment decision.
All expressions of opinion are subject to change without notice in reaction to shifting market conditions. Data contained herein from third-party providers is obtained from what are considered reliable sources. However, its accuracy, completeness, or reliability cannot be guaranteed.
Examples provided are for illustrative purposes only and not intended to be reflective of results you can expect to achieve.
Investing involves risk, including loss of principal.
Past performance is no guarantee of future results.
The Schwab Center for Financial Research is a division of Charles Schwab & Co., Inc.
All corporate names and market data shown above are for illustrative purposes only and are not a recommendation, offer to sell, or a solicitation of an offer to buy any security. Supporting documentation for any claims or statistical information is available upon request.
1124-5UCZ
A message from Advisor Perspectives and VettaFi: To learn more about this and other topics, check out some of our webcasts.
© Charles Schwab
More Artificial Intelligence Topics >
