S&P 500 Experiences Decline Amid Economic Concerns
The S&P 500 has seen a nearly 9% drop from its peak in January, mainly due to ongoing geopolitical issues in the Middle East. Investors on Wall Street are increasingly worried about a slowdown in both the U.S. economy and corporate earnings, particularly as oil prices rise.
Some tech stocks have fared worse than the S&P 500. Take Nvidia, for instance. Its value has plummeted 20% from its all-time high. Even though the company might face some economic headwinds in the short term, the demand for data center chips should remain robust in the medium-to-long term, especially given the surge in artificial intelligence (AI) advancements.
Intriguingly, Nvidia’s stock has now become more affordable than the S&P 500 based on its forward price-to-earnings (P/E) ratio for the first time in thirteen years, leading many investors to see it as a rare buying opportunity.
Insights on Nvidia’s Advancements
Nvidia’s flagship data center chip, the graphics processing unit (GPU), is designed for simultaneous processing, making it particularly suited for tasks like AI training and inference. The latest model, the GB300, significantly outperforms its predecessor, boasting up to 50 times the performance under certain setups.
This year, Nvidia introduced the Vera Rubin platform, which integrates Rubin GPUs and Vera CPUs alongside enhanced networking equipment for quicker processing capabilities. As stated by Nvidia, this new platform enables developers to train AI models with 75% fewer GPUs—resulting in a remarkable 90% drop in inference token costs.
To clarify, tokens refer to the units of data produced by AI models in response to prompts. They include words, symbols, images, etc., which require computational power to generate. Consequently, most AI companies charge based on the number of tokens used. Lowering inference token costs could boost adoption rates while enhancing profit margins.
As a result, Nvidia’s CFO, Colette Kress, anticipates that all cloud providers and AI developers will adopt the new Vera Rubin chips, with samples dispatched recently and mass production expected later this year.
Wall Street Projects Significant Growth
Nvidia’s fiscal year 2026 wrapped up on January 25, achieving record revenue of $215.9 billion—up a staggering 65% from the previous year. The data center segment alone brought in $193.7 billion, marking a 68% increase.
Moreover, Wall Street is predicting that Nvidia’s revenue will climb even further in fiscal year 2027—projected to surge by 71%, likely reaching about $370 billion (as noted by Yahoo! Finance). This indicates a substantial demand for the Vera Rubin chips.
Ultimately, Nvidia realized adjusted non-GAAP earnings of $4.77 per share in fiscal year 2026, with forecasts suggesting a jump to $8.29 per share in fiscal 2027. These figures are crucial to understanding the company’s valuation.
Nvidia’s Valuation at a Unique Junction
Currently, based on its fiscal year 2026 earnings, Nvidia’s stock trades at a P/E ratio of 34.7, which is considerably lower than its 10-year average of 61.6. On the other hand, using the earnings forecast for fiscal year 2027, the stock is only at 20.5 times forward earnings. Notably, this makes Nvidia cheaper than the S&P 500, which has a forward P/E of 20.7—something we haven’t seen in over a decade.
This situation might present an exceptional opportunity for investors. For Nvidia’s stock to align with its 10-year average P/E of 61.6 by the end of fiscal year 2027, it would need to experience a 200% increase. Of course, this hinges on whether Wall Street’s profit forecasts hold true.
I have this feeling—call it a hunch—that Nvidia might not only meet but possibly surpass those expectations. CEO Jensen Huang recently pointed out that the world typically allocates about $400 billion each year to traditional computing infrastructure. To put things in perspective, he emphasized that AI workloads require, well, perhaps a thousand times that in computing power.
Because of this, Huang believes that data center operators could end up spending as much as $4 trillion yearly on AI infrastructure by 2030, which, if accurate, opens up an enormous market opportunity.
