If there has been an ultimate winner with the advent of the smartphone and tablet revolutions, it has to be ARM Holdings (ARMH). Not only has the ARM instruction set architecture become the de facto standard for mobile operating systems, including Google's (GOOGL -1.97%) Android and Apple's (AAPL 0.52%), its processor and graphics core designs continue to ubiquitous throughout the mobile world. Further, while ARM's smartphone and tablet applications processors have stolen the spotlight, it's easy to forget that ARM microcontrollers and very-low-power processor cores are at the heart of everything from cellular baseband chips to hard-disk controllers. ARM is the ultimate processor IP firm, but that doesn't necessarily make it the ultimate investment at this point.

The risks to sentiment are very real
While ARM's processor cores truly are ubiquitous across the computing industry, a significant part of its "S-curve" revenue and profit growth comes from the dramatic average selling price uplift from its partners' sale of mobile chips. That is, while 10 years ago ARM's business was all about collecting royalties from chips that may have cost a few cents, today ARM collects royalties on chips that go into phones that cost as much as $50. Further, as ARM now often provides graphics, CPU, and physical IP, its royalty rate per chip has continued to climb, driving substantial growth on the top and bottom lines.

That being said, a good deal of ARM's fairly sky-high valuation appears to be predicated on the following drivers:

  • Continued market share dominance in the smartphone/tablet spaces
  • Even higher royalty rates per chip, particularly as the move to ARMv8 (ARM's 64-bit instruction set) and to the 64-bit processor cores such as the Cortex A57/A53 should provide yet another bump in royalty rate
  • Meaningful share by its partners in the datacenter

While it seems likely that the second point in the above thesis is a certainty, the first and third are nowhere near guaranteed. On the first point, while it's still too early to tell, it seems likely that with the recently announced Bay Trail family of system-on-chip products, Intel (INTC 1.77%) will gain meaningful share in tablets. Further, with the fourth-quarter launch of Merrifield for smartphones, it also seems likely that Intel will finally gain some real share in smartphones. If Intel is actually able to drive meaningful share in both tablets and smartphones over the course of 2014, this will prove to be a significant headwind to ARM's financial position as well as sentiment. Investors should keep an eye on Intel in this space going forward.

Further, it is not at all clear if ARM's partners will make meaningful inroads in the datacenter. While I do expect that there is a real chance that network processor vendors that have traditionally used either MIPS by Imagination, such as Cavium Networks and Broadcom, it's hard to sell a general purpose server CPU story. Intel's recently released, 64-bit Atom C2000 chips offer best-in-class performance per watt for micro-servers, and given that the majority of the general purpose compute in the datacenter is done on X86, it'll be very difficult to convince people to switch – the software recompilation and optimization costs would be a killer!

Granted, if Intel didn't have competitive solutions for this lower-power space, ARM's partners would have a much easier time of it, but with Intel being first to market with competitive products, the optimism there (and the opportunity for significant royalties) seems fairly muted.

What's ARM actually worth?
As of the most recent close, ARM is priced for perfection. In fact, assuming that the company can generate $1.04/share in free cash flow this year, and assuming that the company can grow free cash flow at a 25% 10-year CAGR before leveling off to 11% long-term growth, it's tough to justify a fair value above $50/share – just 8% above the most recent close.

The problem here is this means that the shares are pricing in some fairly substantial growth. While ARM may actually meet – or even actually exceed – these estimates, it's hard to ignore that this seems to be pricing in a scenario under which everything goes right. Any indication that things are actually going south (for example, Imagination Technologies winning back GPU IP share, Intel winning smartphone/tablet sockets, etc.) or simply not as planned (limited penetration in the datacenter), and those growth estimates come down and with it the share price.

The Foolish bottom line
ARM Holdings is a great company that has managed to be a wonderful secular growth story over the last few years, but it looks fairly valued at best with significant risks to its long-term growth projections and. While shares would look more attractive on a meaningful pullback, it's difficult to recommend them today.