Is this a buying opportunity? Is there a moment here where investors should say, actually, let's run into this fire and let's buy this up?
There's a buying opportunity in semis, but Nvidia is now one of the highest-risk semis. There are other semis where the risk is lower and the reward is also high: Broadcom, AMD, Micron. Yes, maybe there will be further restrictions on sales to China, but they're not going to have to deal with this kind of episode.
Welcome to Prof G Markets. I'm Ed Elson. It is November 11th. Let's check in on yesterday's market vitals. The major indices declined following the inflation report. Treasury yields rose on bets that the Fed will raise interest rates in December. On Kalshi, the odds of a rate hike hit 86%. The dollar gained. And finally, Nvidia shares dropped another 5% as the fallout from Jensen Huang's affair continued.
The Jensen Huang scandal has sent Nvidia stock plummeting for a second day in a row. New reports emerged yesterday alleging Nvidia purposefully intervened in an investigation into Mega Speed International, the company run by Alice Huang. Miss Huang is the woman with whom Jensen Huang was having an affair. The stock is now down more than 10% since news of this scandal broke, erasing nearly half a trillion dollars in market cap. It is also off more than 30% from its record high in early October. More than $1.5 trillion in market value has been erased in less than 5 weeks. So, here to break down what is happening to Nvidia, what it means for the rest of the chip market, we are speaking with Gil Luria, head of technology research at D.A. Davidson. Gil, this has just been a fascinating few days. I mean, I don't think anyone would have expected that the thing that would bring Nvidia to its knees is basically an affair, a scandal between Jensen Huang and an alleged Chinese spy. I want to get your views on the stock and where it's trading at this point, but let's just start with your initial reactions. What do you make of everything that has unfolded this week?
Well, it's very disappointing, to say the least, and I would argue that just this story has cost Nvidia a trillion dollars, just the story, because all the other chip stocks were up 10%.
And so if Nvidia is down 10% when all the others are up, that means that Nvidia is really down 20%, which is a trillion dollars. That's a very costly mistake. The risks here are that the investigation does find wrongdoing. That Nvidia knowingly sold chips to China that it shouldn't have. That Nvidia knowingly sold chips to compute providers they knew would provide access to Chinese companies. That's the big risk because that would limit Nvidia's business going forward. Nvidia still has a healthy Asia business outside of China. And if it turns out that a lot of that Asia business, or at least some of that Asia business, was just funneling to China, that would be an actual risk to the numbers. Beyond that, if the investigation concludes that there was more unethical behavior from Mr. Huang than already disclosed, he will likely have to step down. And while Nvidia can survive and still succeed without Jensen Huang, and the AI ecosystem can survive and succeed without Jensen Huang, let's not forget that Jensen Huang is a big reason we have this AI ecosystem. They were using GPUs for AI in 2006. He delivered the first GPUs to Elon Musk and Sam Altman at OpenAI at the time in 2016. We wouldn't have what we have today if it wasn't for everything that Jensen Huang's done since then to build this ecosystem. Doesn't mean it can't succeed without him, but it would be a very disruptive move for Nvidia and the ecosystem if he had to step down.
Is it possible to put numbers on the risk at this point? Because that's kind of the investor's job right now. I mean, on the one hand, you've got the risk that they knew what was going on and that they knew that they were smuggling chips to China, if that is indeed what was going on. And so, you might have maybe some changes to export policy from the government. It might mean a risk to their Asia business. But I think one of the harder things to sort of ascertain how important it is, is how essential Jensen Huang is to Nvidia. And I feel like investors are divided on this question. I mean, how much of a problem is it for Jensen Huang to be out as CEO? And is it possible to assess in dollar terms what that means for the valuation of the company? I know I sort of thrown a lot of questions at you there. But can you tackle them?
Well, that's what we have to do now, unfortunately. So we have to think about at least a portion of the Asia sales if they were funneled to China. Let's say order of magnitude up to 20% of Nvidia's revenue backward looking may have been funneled to China. Now that doesn't mean that we'd have to reduce our numbers by 20% because the demand environment right now is so healthy that you could probably still sell all the chips that Nvidia can make next year without selling anything into Asia.
And just to clarify your comments there, Gil, just for our listeners, the allegation is that Nvidia was selling their chips to Mega Speed International, which is this Singaporean essentially shell company, and that those chips were being smuggled into China. And you're suggesting that perhaps it's larger than this. Perhaps there are other companies outside of China, but in Asia where a similar thing was going on. That is what you're suggesting might be a risk, right?
Correct. That is possible. And in fact, there have been arrests and allegations around people smuggling GPUs into China that are not associated with Mega Speed. So it wouldn't be completely out of left field. And in fact, the demand from Asia was probably healthier than you would think based on the fact that the vast majority of the AI is happening in the US and in China and not in many other regions. So this has been a concern for investors that's been dormant, but now this story raises that to the surface. And then again, much harder to quantify the impact on Nvidia: Jensen Huang leaving. Again, there wouldn't be no Nvidia, certainly not this big, without Jensen Huang. And again, I would argue we wouldn't have this enormous and growing AI ecosystem without Jensen Huang. But as we sit here today, there are stewards to this industry that are well beyond Jensen Huang. I would include Andy Jassy and Satya Nadella and Sundar Pichai and to some extent Sam Altman, Dario Amodei, Lisa Su. There are enough stewards that can carry the AI ecosystem forward. There's no succession at Nvidia. So that's a problem they have to solve immediately. They have to create a succession plan immediately because even if the investigation concludes that there was no further harm done by Mr. Huang and that he can stay as CEO, even if there's just a small chance that they may have to replace him, they should start the succession program now internally and externally because this is a fast-moving industry and Nvidia is still the largest company in the world. They would need to move fast if they indeed need to take sanctions against Mr. Huang.
So, just looking at the stock, it has gotten destroyed on this news. As I said, down more than 30% from its highs in October. It's down around 15% on the year. And just looking at the multiples here, on a forward PE basis, it's trading at a multiple of earnings of 12 right now. So, the multiple's fallen more than 50% over the past year. On a trailing PE basis, trading at 22 times earnings. So you've got Nvidia here which is trading at a lower multiple than companies like Walmart, Costco, Procter & Gamble, Coca-Cola. These are names that you would never expect to be trading at a premium to Nvidia. And yet that is what is happening right now, which really has to beg the question of: is this a buying opportunity? Is there a moment here where investors should say, actually, let's run into this fire and let's buy this up?
What do you think? There's a buying opportunity in semis, but Nvidia is now one of the highest-risk semis, and the risk-reward now is quite... the risk is high, the reward is high. There are other semis where the risk is lower and the reward is also high: Broadcom, AMD, Micron. These are also stocks trading at a very big premium to the market in spite of the fact that they're going to grow 50 to 100% next year in a healthy market and therefore have a very low multiple if you believe that the data center buildout will continue past that, and they don't have this to deal with. Yes, maybe there will be further restrictions on sales to China, but they're not going to have to deal with this kind of episode, or it's pretty unlikely that they'll have to deal with this kind of episode. And so the risk there is much lower, the reward there is still high, and Nvidia just has to go down the priority list based on the fact that you have this much uncertainty right now.
Is it your expectation that one of the consequences of this would be that those chip makers take up more market share now and that we'll see a lot of customers basically saying, okay, I don't want to be a part of this, I'm going to shift over and find a new supplier?
I wouldn't necessarily include that in the risks apart from the fact that in China they're going to buy less chips. In China it's going to get harder to get your hands on GPUs, but in the rest of the world the customers trust Jensen Huang. Even within these events, there's a lot of trust that's been developed over the last few years. Again, he's orchestrated this whole ecosystem. There's been a lot of cooperation and so there's some level of stickiness to Jensen Huang. Now, whoever the successor is will have to maintain those relationships and maintain the product roadmap because it's a very aggressive product roadmap. Nvidia's secret sauce is their products are always better than everybody else's, right? Everybody else can just argue that their product is cheaper. Nvidia's product has continuously been better and continues to maintain a lead. That would have to be something the next leader, to the extent that Mr. Huang has to be replaced, the next leader would have to maintain that. But short of that, I wouldn't expect their big customers to shift based on this story or based on the developments in China.
Switching topics slightly, CoreWeave reported earnings. The stock popped as much as 8% in after-hours trading, though it did come back down after. CoreWeave has been one of those companies that you have been flagging for a long time now as sort of one of the naughtier players in AI. And it seems that investors have started to recognize that. Obviously, we've seen that the stock has gotten hammered. The spreads on its credit default swaps have exploded. The yields have exploded as well on the bonds. But this looked like maybe a moment where investors say maybe there's hope for CoreWeave. What did you make of the earnings? What does it tell us about the rest of the AI trade?
So, the market for selling compute is absolutely the best it could be. There's so much shortage of compute, the demand is so high that this is the ideal environment for CoreWeave. And yet if you take their adjusted operating income times 4 divided by net PP&E, which will give you a return metric, they return 3% on their investment. Their cost of capital, if you take their interest expense, multiply by 4, divide by the debt, was more than 9%. This is a negative carry trade even in the best market for compute imaginable. If they can't get a sufficient return right now, I don't know when they ever will. If they can't raise capital, which is what they said — we can't go to the debt markets anymore for unsecured debt — they basically just said that on the call. They have to raise equity now. They have to ask for their customers to prepay up to the full level of capex. And even in this environment, that tells you that when things get even a little tougher, they're not going to make it.
Is CoreWeave just one of those stocks that you just have to stay away from at this point? I mean, the risk is too high, as you point out, like they have to go out and raise more money. And it seems as though in this capital environment where investors are very, very anxious, certainly more anxious than they've been in the past several years, going out and raising tens of billions of dollars, hundreds of billions of dollars, is just going to be a lot harder to do.
Oh, absolutely. And to tie the two stories together, part of what CoreWeave shareholders felt was downside protection was this notion that Nvidia would bail them out. Nvidia's already bailed them out three or four times, and there was this notion that they'll just keep bailing them out.
Nvidia didn't bail them out. Jensen Huang bailed them out. If he's not there, that supposed input goes away. So, yes, absolutely stay away from CoreWeave.
