Credit where it's due - NVIDIA (NASDAQ:NVDA) (or "Nvidia") has executed a lot better over the past year than I thought it would and the shares reflect that. Nvidia's stock has risen close to 70% over the past year, leaving just about everybody in the dust. It is to management's credit that it has driven performance to a point where concerns about the loss of licensing revenue from Intel (NASDAQ:INTC) and still-significant exposure to gaming are secondary relative to the emerging long-term opportunities in markets like virtual reality, data centers, "deep learning" and autos.
Clearly the company will have to maintain its execution edge, and companies like Intel, Advanced Micro (NASDAQ:AMD), and Mobileye (NYSE:MBLY) have their own growth plans that don't include letting Nvidia beat them. I think Nvidia can continue to leverage its expertise in GPUs to grow and diversify the business, but it's hard to call the stock really cheap at these levels unless you believe auto (or other markets) can drive long-term double-digit free cash flow growth from here - a tall order for a company of Nvidia's size, but not impossible.
Will A Pivot Toward AMD From Intel Hurt Nvidia?
It has yet to be confirmed, but Intel has reportedly been holding discussions with AMD about licensing its graphics technology. As a reminder, Intel pays Nvidia $66 million a quarter under a deal that expires in the first quarter of 2017. Past relations between Intel and Nvidia have certainly been contentious, but I find it interesting all the same. There's no end of debate about which company has the best graphics card for PC gaming, but I think it says something that there is still a debate when Nvidia has continued to take market share from AMD in both discrete desktop and mobile GPU markets.
Intel could want any number of things in licensing AMD's tech. Intel has previously talked about GPUs being important building blocks in the data center and AMD really hasn't been involved in the data center market. Intel could also want to improve its own graphics capabilities and/or better position itself for the upcoming growth in virtual reality applications. Last and by no means least, Intel may expect future litigation from Nvidia after the current deal ends and licensing AMD's technology could help Intel shore itself up for those fights.
I can't say this doesn't matter to Nvidia, but at the risk of being overly dismissive towards AMD, I'd say that AMD's history does speak for itself. Nvidia would certainly like to maintain margin-rich streams of licensing revenue, but it is not as though the company is dependent on that revenue/cash stream. While it is possible that Intel could do more with AMD's technology than AMD has managed, I don't think Nvidia is going to lose a lot of sleep over the competitive ramifications of such a licensing deal.
Queuing Up The Growth Opportunities
That Nvidia is still getting strong growth from gaming (up 25% yoy and 6% qoq in the last quarter) when industry PC shipments are so weak is a pretty strong testament to the resilience of gaming. It's also worth noting that Nvidia gets less than 20% of that revenue from the U.S., with China having risen to somewhere around 40% of the market (which helps explain ASUS' presence as a major customer. Even though gaming revenue seems to have longer legs than previously thought (and management is looking for 8% to 10% growth over the next three years), the company has nevertheless been active in diversifying itself away from that business.
Virtual reality is one such market and one that seems ready to ramp up relatively soon. Given that the demands of virtual reality systems tend to be well beyond that of regular desktop use (I've seen estimates of graphics requirements seven times higher), there's a pretty strong opportunity here for Nvidia assuming VR catches on.
Data center/cloud acceleration and deep learning also remain promising opportunities on the enterprise side. How quickly these opportunities develop is a major question, but if the addressable market sizes that Nvidia management has laid out are credible, the company has relatively minor share in what should be multibillion-dollar opportunities over the longer term. In both cases, though, there is going to be competition. Intel and Xilinx (NASDAQ:XLNX) argue that FPGAs are the way to go, and Microsoft (NASDAQ:MSFT) has apparently seen some good results using FPGAs while Xilinx and IBM (NYSE:IBM) just announced a partnership in high-performance computing. That said, it's not an all-or-nothing situation and it quite likely that CPUs, GPUs, and FPGAs will all have their role to play.
Nvidia's opportunities in the auto sector could ultimately be the largest of all. There is very much an element of "stop me if you've heard this one...," as seemingly every chip company is angling for a piece of the auto market. That said, Nvidia has a more credible platform than many would-be players, given the company's existing Drive CX platform (largely infotainment), its Drive PX platform and the Drive PX2 platform (autonomous driving) that it showcased at the most recent CES in January.
The Drive PX2 is no joke, with two Tegra processors and two discrete GPUs, but I'm curious if they can sell this at a digestible price to auto OEMs and still make attractive margins. It's also well worth noting that there are going to be a lot of competing efforts, including Mobileye's Road Experience Management system, and Texas Instruments (NYSE: TXN) and Intel have made it clear that they intend to be involved in the market as well. This market opportunity is very new, though, and I'm frankly at a loss as to how to handicap it. Clearly any sort of autonomous driving system will have to include imaging technologies, sensors, and controls, as well as a processing platform and software to manage it all. Nvidia's core technology would seem to fit the processing needs, but there's a long way to go before self-driving cars are really ready for primetime.
It's About Profits
The semiconductor market isn't what it used to be, and I don't believe companies can get very far pursuing a strategy of ignoring margins and expecting (hoping?) that growth will make everything okay in the end. With that, those companies that focus on really leveraging unique, defensible assets should do well over the long term. I think Nvidia's GPU-based technology fits and the company has the sort of gross margins you'd like to see in a defensible asset. Now it's about leveraging the business, generating good margins/cash flows, and maintaining that sweet spot between investing enough in R&D to maintain leadership and create new business opportunities while not wasting resources "prospecting" for growth.
I'm only looking for mid-single-digit long-term revenue growth and Nvidia could exceed that if it outperforms in markets like data centers and autos. I'm expecting further improvement in adjusted free cash flow margins (into the high teens), but the discounted cash flows don't get me to an attractive fair value today. On the other hand, the company's GAAP operating margins, even after subtracting the Intel licensing payments would support a price higher into the $30s.
The Bottom Line
Nvidia outperformed my expectations in 2015 and could certainly do so again. If management is correct about the size of the company's addressable markets (and its ability to compete effectively for share) and keeps its eyes on the ball with respect to margins and leveraging the company's true strengths, I can see a path to upward revisions in growth and margin expectations down the line. While I'm still cautious on the shares, I like the company and will continue to follow it.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.