We don’t care at which price Apple trades. When it was at $165 per share not that long ago, we thought the Club’s original “own it, don’t trade it” stock was momentarily expensive. Then nothing happened, and we had an ascent based on a momentary belief that the iPhone 16 would have some good artificial intelligence in it. The stock market seemed ho-hum on the device; nothing worth buying the stock for. Then came Apple’s annual Worldwide Developers Conference (WWDC) with the confirmation of lots of different, disparate AI features, and we had a lot of instant, negative, analysis that day, a day where the market didn’t yawn; it just went plain down. I remember watching the presentation and deciding right there, I had to upgrade. But investors, prodded by news reports and analysts spouting negatives like machine gun bullets headed to any bull that popped its head up, just dumped and dumped. WWDC was never supposed to be a major, market-moving event as it was that afternoon, and the vicious decline shocked many of us. I talked to Apple CEO Tim Cook late in the trading session. Before I asked my first questions, I said I was very surprised about the reaction given that many potential buyers had been waiting for so many different features before they pulled the trigger, and this phone seemed to address every one of them, including that of competitors. I was thinking of that last call from portfolio holding Best Buy when Samsung’s phones were well-received. Tim was surprised at the reaction, too. He was talking about how there were so many more features than were highlighted. Sure enough, the next day began a historic ascent of Apple stock, an in-your-face rally that many thought made no sense. As it went up, a lot of bears turned into bulls, many price targets were raised, and a lot of tepid-but-pro Apple analysts got more positive. Where were they when everything looked so dim? The stock’s ascent is all considered rational, every darned point of it even as the stock sells at 32 times earnings, much much more than its historic value. Microsoft , also a Club company, brought out a Copilot artificial intelligence assistant that garnered instant love even as most of the answers are soporific and sound like what we used to call “fill ‘er up” fluff in my old newspaper days. So much of an answer reads like a term paper: 700 words long, only 400 of them relevant. Then this past week, we got a new Microsoft personal computer filled with things we can ask it and make it so it’s a smart PC, as if our old ones are idiots. I continue to believe this will be a huge upgrade cycle, benefitting Dell and HP Inc. , although, somehow, it’s accruing entirely to Microsoft. It doesn’t even matter what chip is in the PC. Micron was just thrown back a bit and Advanced Micro Devices is down huge from its high. For all of these new features, Microsoft sells at a historically high 38 times earnings. It’s all considered rational, every darned point of it. Then there’s Nvidia . Right now every company that makes any hardware, every company that connects to the internet, any company that has multiple data centers to address everything from ChatGPT (Microsoft-backed OpenIA) and Claude ( Amazon -backed Anthropic), and Gemini ( Alphabet ‘s Google) to massive video files, and omniverse robots, needs Nvidia chips — and of whatever vintage, because they all work together, or else they cannot compete. They will be left behind, and become the hyper-scaler that is all hype, no scale. Its machines come trained, and you just add the inference and cut through it with “analysis interrogating” the data; the phrase now being used. You need the Nvidia Super Computers that can make it so everything you can do, a robot can do better. It will shock us how unadvanced we are now. There will be nothing from Intel or even AMD that will get near its speed and training. You need both: Intel, despite its bogus, sad claims, has neither. AMD has only speed; no training but that’s coming. Who knows when, though, and every minute that you are reading this is one minute closer to the new Nvidia chip. There’s a coalescence of two themes, high speed, accelerated computing and increasingly intelligent AI, and neither works without Nvidia chips. Therefore every central processing unit (CPU)-based machine will see its CPUs replaced with GPUs (graphics processing units) that are much more powerful, I mean really powerful and will burn less hot than many of the chips before it; at least as a ratio to output. This changeover will occur over multiple years: You can’t upgrade everything that needs to be upgraded in one year or two or even three. How necessary is this upgrade? I regard the fastest CPU-based machines as 286-based, while the new GPUs are the Pentium IIs; to use an Intel metaphor. Remember, by the way, that Intel is still trying to add transistors to their chips with the hope that Moore’s law, power doubling every couple of years, hasn’t run out of time. If we aren’t tempted to upgrade because we use our PCs just as typewriters, then maybe the upgrade won’t be worth it. But that’s a sorely minority position. Most users will regard our current PCs as typewriters once we see Nvidia-powered anything, especially the PC. The only thing holding Nvidia back is the possible max out of its partner, Taiwan Semiconductor Manufacturing Co. , unlike gigantic Intel putting up incredibly expensive factories adhering to the diminishing Moore theory, Nvidia is fabless. And for all of that, Nvidia’s ascent is all considered irrational, every darned point of it. Perhaps that’s because, to the naked eye it sells at 46 times earnings, but, unlike Microsoft and Apple, every time Nvidia has sold at a high level if you check back one year later it turned out to be very cheap. When I say very cheap it looks like the forward price-to-earnings (P/E) was just plain wrong. Or to put it another way, we will find out, a year from now, calendar year 2025, that Nvidia at these prices, was selling for a market multiple. Sure it could turn out that you are paying an overvalued price if anyone comes near Nvidia, if anyone catches them, but right now you would have to have something that doesn’t use traditional semiconductors, something like we thought Elon Musk was going to do but ended up capitulating and going with Nvidia. I don’t see anyone else creative enough to catch the company any time in the foreseeable future. Sure you can sell and bet that the “unforeseeable” future is going to be cataclysmically bad, but that’s a risk I don’t want to take. Now does that make Nvidia over or undervalued? Arguably overvalued now but undervalued a year from now when we find out once again that Nvidia is making more money than we thought. Again, if you think they will make less then you have to abandon ship now. I just think that, too, is a very risky bet. Nonetheless, everything I have written here comes under the category of trying to explain why Apple and Microsoft’s valuation don’t seem to upset anyone while Nvidia’s seems to upset everyone. Most people didn’t know what Nvidia was a year ago, it maintains this weird gate-crashing parvenu status. I was actually discussing this very point with my daughters during our trip out West. I said I couldn’t figure out why people didn’t know it. My eldest said I had been knee-deep in Nvidia (sometimes neck-deep) and that how dare I think that people should know it. “Did the viewers get to go with you to Nvidia?” she asked. “Have they met Jensen Huang like you have?” Jensen is, of course, the founder and CEO of Nvidia. I have to admit that if you haven’t gone to Jensen’s headquarters or spent a lot of time with the man you would still be somewhat baffled by what they do and what they will do. I grew up in Philadelphia where we would go to the Franklin Institute all of the time and we would see what Ben Franklin discovered and invented. Or if you want, go to the Da Vinci museum in Rome where you can see more than 50 of Leonardo’s inventions. These are the people who are like Jensen. You couldn’t buy stock in what the former two had going, but you can buy stock in Jensen by buying shares of Nvidia. I think that if you could get either Franklin or Da Vinci when they just started to dazzle you wouldn’t be thinking, “When is this run going to end.” You would be saying, “OK, I am not early and the darned thing can and probably will go down from here after this historic run” — but you wouldn’t be baffled or mystified — you would just be in awe. But because you have probably only seen the Nvidia nameplate on things connected with gaming and not housing, not factories, not health care, not powerful computers that you will never see close up that belong to Oracle and Club names Amazon, Google, and Microsoft — and, the purpose of Nvidia seems lost. In fact, I thought that Apple would tie up with Nvidia to commercialize its new, pricey Vision Pro mixed reality headset, something that would have benefitted both companies — Jensen had talked about it and showed it at his GTC Conference. It didn’t happen. In this sense, it feels like Cisco before the crash. No one could figure out quite what it did except the “visionaries” that were tech geniuses. Almost all turned out to be fools. I know that it seems impossible to ponder Nvidia’s stock going any higher or passing Apple and Microsoft and staying ahead of it. But it is cheaper than either on 2025 earnings — and it doesn’t have the kind of competition that either has when it comes to foreign phones or data center build-outs. One last thought. When I got in the stock picking business more than 40 years ago, a wise man once told me that the most expensive words are “it’s different this time,” as if everything has occurred and don’t think this time will be different. It will be the same. In a four-decade retrospect, I think the five most dangerous words are “it’s not different this time” because if you believed that, you would have sold Nvidia’s stock at a $1 trillion market value or $2 trillion because both were so outlandish if you didn’t know the stock or the company. Last week, if only briefly, Nvidia topped the U.S. $3 trillion-plus market cap club of which there are only three members. Microsoft and Apple are the others. The whole point of the best part of the market is to reward innovation, gross margins, initiative and dazzling brilliance. Ask yourself, does any company match Jensen and Nvidia right now? The answer is no. This time remains different. As tempting as it is to ring the register, I think you need real reasons to jump off. Right now, the only one I have is that there is only one Jensen. But there is only one Musk, there was only one Steve Jobs, so I have a tough time trimming the stock based on the one-and- only-great-man-theory. I say let others be baffled over the Apple versus Microsoft versus Nvidia conundrum. To me, only one is inexpensive on what will turn out to be 2025 earnings. That one is Nvidia. (See here for a full list of the stocks in Jim Cramer’s Charitable Trust.) As a subscriber to the CNBC Investing Club with Jim Cramer, you will receive a trade alert before Jim makes a trade. Jim waits 45 minutes after sending a trade alert before buying or selling a stock in his charitable trust’s portfolio. If Jim has talked about a stock on CNBC TV, he waits 72 hours after issuing the trade alert before executing the trade. THE ABOVE INVESTING CLUB INFORMATION IS SUBJECT TO OUR TERMS AND CONDITIONS AND PRIVACY POLICY , TOGETHER WITH OUR DISCLAIMER . NO FIDUCIARY OBLIGATION OR DUTY EXISTS, OR IS CREATED, BY VIRTUE OF YOUR RECEIPT OF ANY INFORMATION PROVIDED IN CONNECTION WITH THE INVESTING CLUB. NO SPECIFIC OUTCOME OR PROFIT IS GUARANTEED.
CNBC Investing Club with Jim Cramer
Rob Kim | NBCUniversal
We don’t care at which price Apple trades. When it was at $165 per share not that long ago, we thought the Club’s original “own it, don’t trade it” stock was momentarily expensive.
