I have been fairly bearish on shares of ARM Holdings (NASDAQ:ARMH), primarily due to the fact that I believe that the assumptions that seem to be propping up the DCFs that target a $40+/share price seem to be - in my view - wildly inaccurate. In particular, as I have thoroughly examined a number of sell-side reports, the following themes tend to be the cornerstone of the bullish thesis:
- ARM will capture 10-30% of the traditional clamshell notebook PC market by 2013 (depending on whom you ask)
- EBIT margins expand from ~40% to 60%+, driving significant EPS leverage
- ~30-40% smartphone and tablet unit CAGR through 2015
- ~30%+ increase in smartphone chip ASPs by FY2016e
- ~20% market share in servers by 2016
The tricky part of trying to make a bearish call on ARM at this point is that many of the assumptions that drive the valuation will not become perfectly clear for a while. That being said, I believe that the supremely premium valuation that the stock trades for - roughly 58x FY2013e consensus - makes the stock particularly vulnerable to any cracks in the armor. I believe that we are beginning to see the first signs of said cracks. Investors need to see one or more of the following in order for the stock to see a steep downside move (rather than the trading noise that we are currently seeing):
- By end of year 2013, ARM's notebook share coming in at significantly below the 10% mark - the low end of many sell-side estimates
- Slower-than-expected royalty rate increases, driven by a mix shift towards parts that use far less ARM content. In particular, as Qualcomm (NASDAQ:QCOM) and others that use internally designed graphics architecture or architectures from Imagination Technologies continue to take share.
- Slower than expected smartphone adoption rates, or a mix shift towards much lower cost smartphones which will use much less expensive chips that likely generate royalties more along the lines of traditional feature phone processors rather than the higher end smartphone processors
- The Q3 launch of Intel's (NASDAQ:INTC) "Avoton" micro-server part will likely seriously call into question the viability of the ARM micro-server strategy (and may cause analysts to revise down estimates here accordingly). Details of this part will likely be given at IDF Beijing, which could throw cold water onto the situation sooner rather than later.
- The Q3 launch of Intel's "Bay Trail", along with the Q2/Q3 launch of AMD's (NASDAQ:AMD) "Temash" should help to block the penetration of Windows RT in the traditional "notebook-sized" 10" tablets. At best, Windows RT may ultimately be relegated to 7-8" devices. Even then, Nvidia (NASDAQ:NVDA) appears to be the leader in Windows RT, which strips ARM of any potential GPU core licensing royalties in this space.
Watch for the above signs as an indicator of when it'll be "safe" to short. One big "miss" against investor expectations, and you will likely see the "Apple-effect", which is what caused shares of Apple (NASDAQ:AAPL) to plummet from $700+ to $430 - no matter how good the underlying business is, if a stock that has seen a significant run over a fairly short amount of time starts being anything less than perfect, then it becomes a very ripe candidate to short. Similar things happened with Cirrus Logic (NASDAQ:CRUS), Mellanox (NASDAQ:MLNX), Green Mountain Coffee (NASDAQ:GMCR), and Netflix (NASDAQ:NFLX) over the last several years - and ARM is certainly not immune.
Even if you don't short the name, be very cautious about going long - the risk/reward simply doesn't favor the long side right now at these levels, and the case to "take the money and run" becomes more compelling by the day. If ARM slips up even once, then all of the folks that bought in at lower prices will be scrambling to take profits, and it will ultimately not end well.
Disclosure: I am short ARMH, MLNX. 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.
Additional disclosure: I am long INTC, NVDA, AMD