Nvidia | Q2-24 Earning & Call
The omnipotent.
I do not own any Nvidia shares, sadly, but it is one of the companies on my watchlist and I’ll share my DCA price & argumentation at the end of this write-up. Either way, it is a company any investor has to watch closely, as it clearly is one (if not the) leader of the actual upward trend.
Overview.
Once more, the company has provided a very, very strong quarter, but the market is worried about a few things by now - concerns which have existed for years but weren’t important when the stock was running up… Things are tougher now.
EPS. $0.57 | $0.68 | +19.30% beat
Revenue. $25.59B | $30.04B | +17.39% beat
$15.4B returned to shareholders & new buyback program of $50B.
"Hopper demand remains strong, and the anticipation for Blackwell is incredible. NVIDIA achieved record revenues as global data centers are in full throttle to modernize the entire computing stack with accelerated computing and generative AI."
Growth is still very strong, and this quarter beat on both EPS & revenues.
Business.
Demand for Nvidia’s product is still very strong as many companies are building AI infrastructures, and Nvidia is the only one, or at least the very best, providing the necessary technology to do so.
And it shows, with the datacenter branch being the main source of growth.
“Data Center revenue of $26.3 billion was a record, up 16% sequentially and up 154% year on year, driven by strong demand for NVIDIA Hopper, GPU computing, and our networking platforms. Compute revenue grew more than 2.5x. Networking revenue grew more than 2x from the last year. Cloud service providers represented roughly 45% of our Data Center revenue, and more than 50% stemmed from the consumer Internet and enterprise companies.”
All branches are growing double digits YoY.
This tendency should continue as most companies are still building their infrastructure and we remember Meta’s & Google’s earnings call where both Zuck & Sundar said that they will need a computing power more than 10x what they have now, and that the CAPEX was definitely worth it as not building those infrastructures could have terrible consequences.
This was confirmed by Jenssen today.
“Next-generation models will require 10 to 20 times more compute to train with significantly more data. The trend is expected to continue.”
And everyone needs to spend to get those infrastructures first, as being the first allows you to commercialize it & fidelize customers first.
“And so, the ability to systematically and consistently race to the next plateau and be the first one there is how you establish leadership.”
Hopper & Blackwell.
The GPU solutions are the source of Nvidia’s success, and while the Hopper was the star of the last few years, Blackwell should finally pick up the torch from 2025 - what the company calls “fiscal year 2026.”
“Hopper demand is strong, and Blackwell is widely sampling. Blackwell production ramp is scheduled to begin in the fourth quarter and continue into fiscal year '26. Demand for Blackwell platforms is well above supply, and we expect this to continue into next year.”
Revenues.
Rapid review of revenues before talking about the bull & bear cases on Nvidia.
Another flawless quarter, as we can expect for any tech hardware monopoly. Costs are growing much less rapidly than revenues, while the only expenses growing are R&D, entirely necessary for Nvidia. Everything is perfect.
This perfection continues to cash flow & balance sheet with a $26B net debt for $14.5B of OpCF & $13.5B of FCF - to which only 8% are from stock-based comps.
Guidance.
Q3-24 revenue expectations around $32.5B - up 77% YoY with flat margins QoQ while FY-24 gross margins should stabilize around the mid-70s.
Growth is slowing, but growth is going to slow now that comparison is done after Nvidia’s growth spur - no triple-digit growth can be sustained for long.
The Problematics.
There are four “bear cases” (main ones at least) for Nvidia now - three about its business and one solely about the stock.
Spending Strength.
The question is pretty simple: Will companies continue spending on Nvidia’s products as much as they do at the moment?
The strength of Nvidia is that they’re selling computing power, the second most valuable resource at the moment because it allows companies to retrieve, process & share the one & only most valuable resource of the modern world: data.
When you sell this kind of product, your clients are… everyone.
“Demand for NVIDIA is coming from frontier model makers, consumer Internet services, and tens of thousands of companies and start-ups building generative AI applications for consumers, advertising, education, enterprise and healthcare, and robotics. Developers desire NVIDIA's rich ecosystem and availability in every cloud.”
But it doesn’t mean they’ll continue buying forever, especially with this aggressivity. But there are no answers to this question and we’ll just have to wait & see. The only thing for sure in my opinion, is that the second semester of 2024 & probably 2025 should continue with this trend.
Capex vs ROI.
The second most important question is as simple: Will those investments in Nvidia’s hardware pay returns? And when?
Few answers here from Jensen during the call, all of them being yes & rapidly. The first reason being that Hopper or Blackwell can be seen as energy economizers because any company owning them will be able to treat much more data, faster, for the same or lower consumption. Doing more for less.
“And so, the answer is accelerated computing. We know that accelerated computing, of course, speeds up applications. It also enables you to do computing at a much larger scale, for example, scientific simulations or database processing, but what that translates directly to is lower cost and lower energy consumed. And the reason for that is, of course, you just sped up an application 50x.”
The thing is that this is invisible to almost everyone except companies themselves.
The second payback will come in time, but Jensen isn’t saying it that way and certainly isn’t giving any timeframe because it isn’t his company that’ll do it. Nvidia’s products are for everyone and will be used to build applications of a new standard, Gen AI apps.
“It's a fundamental new way of doing software. Instead of human-engineered algorithms, we now have data. We tell the AI, we tell the model, we tell the computer what are the expected answers. What are our previous observations? And then for it to figure out what the algorithm is, what's the function. AI is a bit of a universal function approximator and it learns the function. And so, you could learn the function of almost anything.”
Other CEOs had exactly the same reflexion, notably Brian from Airbnb said this using almost the same words a few weeks ago.
Those will pay back with much better applications and services with time, and all of those will need Nvidia’s products to be trained.
As to answer the question… Again, no definite answers. My personal opinion is that AI & Gen AI are superb tools and will without any doubts revolutionize our world, like the internet did decades ago. The only difference is that it will very probably be invisible for most of us as, sure, some apps will work differently, but everything will be a matter of optimization, not new products.
Margins.
This is Nvidia’s actual strength, and this is due to its actual monopoly. When you are the only one selling a product everyone wants, you can fix the price pretty high - especially when your clients are Meta, Google, or Amazon, who are, let’s say it, rich as f*ck and ready to spend.
Yet, the market expects Nvidia to keep those margins where they are or even grow them, but that will be a pretty impossible task over the long term… Those are up YoY, compared to the last quarter on which the data center branch was not the beast it is today, but we start to notice a plateau and even a potential decline over the last quarters.
Management’s argumentation is pretty convincing & positive so far.
“GAAP gross margins were 75.1% and non-GAAP gross margins were 75.7%, down sequentially due to a higher mix of new products within Data Center and inventory provisions for low-yielding Blackwell material.”
But we do start to see a stabilization in the results & in the guidance, and this might begin to scare the market, especially at today’s valuation.
Valuation.
Which brings us to the most important part of this write-up, on which I’ll use my usual valuation model, forecasting necessary ratios to return our 11% on our investment.
But Nvidia follows different rules for a few reasons: its monopoly on qualitative GPUs, its pricing power & its strong margins. Stocks have lots of different ratios, but the two I use the most are PER & P/S but when it comes to Nvidia, we have to assume a few more things.
Its pricing power will certainly remain, hence margins should stay strong for the next five years, so everything is about revenue growth. The stock deserves bigger ratios than the semiconductor industry due to its monopoly on its products.
This matters because with net margins above 50%, you cannot find a middle ground between growth & income, as PER will always come very low with the needed growth to reach a proper P/S. So I chose to fix net margins at 55%, the actual value as they shouldn’t go much lower over the next five years.
Again, this has no real value and isn’t meant to project the stock price but to have an idea of what would be needed to reach correct returns.
In both cases, it seems very hard to keep a reasonable P/S at today’s price as it would require 65% CAGR for FY-26 & 35% CAGR for FY-29, which in both cases seem completely unrealistic - it would mean that Nvidia generates as much revenue as Apple in two years.
Reasonable PERs are much easier to reach with a correct growth although for FY-26, a 29% CAGR looks slightly optimistic, not unreachable. I’d personally be a bit more conservative but either way, be it on a two-year or five-year horizon, both those cases end up with a pretty huge P/S.
As to which ratios to look for to find a proper valuation for Nvidia… This is entirely personal, and you could even bring more ratios into the discussion (notably P/FCF or EV/Sales), but I’ll keep those two and use an average for my DCA target.
A target based on expectations of around 20% CAGR FY-29 & 27% CAGR FY-26 with a fixed net margin at 50%, as over the long term. Nvidia might slightly lose some of its pricing power or develop lower margin products - it also allows me to be more conservative.
I’ll be a buyer around $80 with conservative data & 5% of margin of safety.
Conclusion.
Nvidia, the company, is a very easy subject to treat as it is one of the most important & valuable companies in the world at the moment, and much of future innovation relies on its products. I don’t think this is up for argument.
Nvidia, the stock, is a much more complex subject, and many still argue on how to value it - its future growth & margins, its deserved ratios, how much our tech relies on it, and much more…
This here was my take, and I’m sure many will think we will never see Nvidia at $80. I hardly disagree; I’m pretty sure we will. The only question is when & the state of its business & of the market when it goes back to this price.
I’ll blindly buy if we reach $80 in the next year.