I recently wrote an article asserting that buying ARM Holdings' (ARMH) stock wasn't the best way to go about buying into the mobile revolution's silicon. In the article, I gave some brief valuation metrics and then proceeded to discuss alternative investments that I believed were more correctly valued and had ample room to run. However, I would like take the opportunity to do a more in depth analysis to help potential investors understand this high valuation. Despite being a well run company that is seeing healthy growth, investors need to be extremely careful to understand just what they're paying a premium for and what the risks to these sky-high multiples are.
Let's Look At The Basics
When it comes to valuing a growth stock, I care that the company is able to grow revenues while keeping gross margins intact and profitability strong.
In 2009, 2010, and 2011, the company took in $451M, $596M, and $721M in revenues, representing roughly 17% compounded annual growth rate. Since ARM operates by licensing intellectual property rather than building, marketing, and selling microprocessors, it maintains gross margins well in excess of 90%.
More interestingly, though, as ARM has branched out to design more sophisticated processor cores, operating expenses have, unsurprisingly, increased as revenues have shot up. However, the company has managed to keep operating expenses under control and grew operating income at a staggering 48% compounded annual rate.
ARM is a healthy company that's here to stay, but in order to know whether the price is right for the company, it's imperative that we examine the valuation.
Valuation Unattractive Compared To Peers
As of the most recent close, ARM Holdings was worth $12.62B. This represents a price-to-earnings ratio of 56.76 and a price-to-sales ratio of 15.22. To get a feeling for just how high these values are, let's take a look at these metrics for some other well known semiconductor companies:
- Qualcomm (QCOM): P/E = 21.3, P/S = 5.87
- Nvidia (NVDA): P/E = 19.30, P/S = 2.27
- Marvell (MRVL): P/E = 13.37, P/S = 1.79
- Texas Instruments (TXN): P/E = 21.63, P/S 2.55
Now, it's important to realize that the companies above are significant beneficiaries of the ARM ecosystem. While ARM supplies core designs and/or a license to develop CPUs that are compatible with the ARM instruction set, the above companies are the ones actually designing the complete chip solutions that end up being sold to the phone/tablet/micro-server vendors.
While ARM benefits by collecting a license fee from each system-on-chip sold regardless of the particular vendor, it seems that the "explosive" growth from the mobile sector has already happened, with steady growth to follow from here. However, the growth could come from elsewhere.
This is where things actually get interesting. Qualcomm has made it clear that they're aiming to get their Snapdragon S4 into laptops running Windows RT. Nvidia is looking to get into the high performance CPU space with its internally-designed, ARM-compatible "Project Denver" CPU, targeting PCs as well as servers. Marvell already has its own ARM-compatible quad core SoC for server use (in fact, Dell (DELL) has announced that it is going to be using it)
Further, on the tablet I expect that if Windows RT tablets come in significantly cheaper than Intel (INTC) "Clover Trail" tablets running Windows 8 Pro, then this could be a large win for the ARM ecosystem and ARM Holdings itself. However, this, like the ARM server push, is very much a "wait and see". It is critical to note that ARM's designs will likely dominate the Android tablet ecosystem since Intel is apparently sitting this one out for now.
ARM Holdings stock is expensive. There's no denying that there's a significant amount of future growth priced into the share price. The key to successfully growing into this valuation will be to expand into new territory. Luckily, ARM licenses instruction sets and, if necessary, reference core designs; the SoC vendors are the ones burdened with the task of convincing OEMs to use their designs. But that also means ARM's fate is in others' hands.
So that's the bottom line: if you think the ARM ecosystem will become a lot more prevalent in areas outside of tablets and smartphones (and the traditional embedded/low power uses that ARM chips have always had), then ARM could very well grow earnings to bring those price-to-earnings and price-to-sales ratios back to earth, making the stock a buy on a dip. But if you're buying ARM to play smartphone and tablet growth, then I highly suggest you think twice; the SoC vendors seem to be the better bet here (since these other companies generally have other core businesses to back them up in case tablets aren't as popular as people speculate).