How does a company increase sales in a market where shipments are decreasing?
For those of you who watch Nvidia, you will have taken note of its meteoric rise in share price since September 2015 to date (~$21 to ~$94); the share price has increased about 440%. However, when normalized to Nasdaq, the growth is 323%, as shown in the share-price chart to the right— still damn impressive and better than all of its competitors.
But when you compare the rise in share price to Nvidia’s sales over the same period of time (which are impressive), you see there is a disconnect between sales and share price, as illustrated in the next chart.
Nvidia does, and has, made the bulk of their sales from big GPUs used in gaming and pro visualization, as the following chart indicates.
And if you plot Nvidia’s overall sales, or just the sales from their big GPUs against shipments (or sales) of PCs, you’ll see PCs going down while Nvidia’s sales (in PCs) have gone up. Nvidia’s results are partially due to increased ASPs and increased mix/share of the final product. And although that’s all good news, it’s not something that can likely be continued, and certainly not at the rate Nvidia has been producing.
So why, then, has Wall Street gotten so bullish about Nvidia, and for such an extended period of time? Diversification, son, diversification. Nvidia has been heavily, and steadily, investing in the automotive market, and in various forms of HPC (they lump it all together and call it datacenter, like Intel and AMD do). For example, Nvidia has been investing in deep learning, AI, supercomputing, and server-based remote graphics. None of those markets are even close in size to the big GPU market, but their growth rate and potential are exciting. In a big sweep of the HPC market, IDC recently forecasted it to grow about 6% to 7% in 2016, which would put total HPC revenue at around $24.6 billion dollars for the year. But more importantly, it predicts a 15% to 19% CAGR to 2019, and Nvidia is right in the center of it. The last chart shows Nvidia’s “other” market growth over the past six quarters.
Although the OEM & IP segment is showing growth, it’s a low-margin business for Nvidia.
Nvidia has, in statement and fact, been diversifying the company into emerging markets, applying stuff they’ve learned, and are learning, from AI and deep learning (DL) to automotive for the autonomous car in our future. And according to Gartner, the automotive semiconductor business generated revenue of nearly $30 billion in 2015, up from nearly $15 billion in 2003.
In addition to leveraging their AI and DL learning across multiple products and markets, the company has been able to leverage its architecture across products, taking its basic Pascal GPU design and applying it to every segment it participates in—the hardware equivalent of write once, play everywhere.
As much as I grouse about the sharpshooters on Wall Street, I have to acknowledge they are a pretty smart group, at least the ones I work with and advise. They see these trends, take the temperature of many markets and macroeconomics, and have concluded Nvidia is in the right place at the right time.
I don’t think you’ll see the gains in share price as big as the past 18 months (I can hear Jen Hsun now saying, why the hell not?). But the vectors are in the right place, and unless Nvidia screws up Volta (not likely), the company is going to be a glory company for a while— maybe not a unicorn, but a prince for sure.