Seeking Alpha
Long/short equity, contrarian, independent research, tech
Profile| Send Message|
( followers)  

I was wrong about ARM Holdings' (ARMH) price action over the last 8 months or so, and it's about time that I own up to being far too early (but not necessarily wrong) on that call. In a bull market with a lot of liquidity and a secular growth story, it now appears that the short call on ARMH (which I issued when the stock was at $37 or so) was dreadfully wrong in the near term and could very well turn out to be wrong indefinitely. While it is my personal view that long term the stock is not worth what it's trading at today, my timing, execution, and near term expectation failures are worth examining. I eat humble pie in this article, and you may learn a thing or two from my mistakes.

Mistake #1: Fighting Sentiment...In A Bull Market

Sentiment, at the end of the day, is everything. When the entire market believes something is true, until there is real evidence to the contrary, then you are wrong. Period.

Right now, the market believes that ARM will continue to run essentially unchallenged in the smartphone space, while at the same time making inroads in the traditional client computing space (thanks to a platform/form factor shift). Further, there is a broad belief that ARM's partners will take significant share in the emerging low power server market.

More importantly, while I have believed that the imminent threat from Intel (INTC) in the low power space would lead to a sentiment shift and market share losses, it seems that the Street doesn't agree with me. I don't have a crystal ball (although I have my own views on how things will look in the next year), but it comes down to perception once again. The market thinks there is a massively asymmetric risk/reward at play here in favor of ARM and its partners, and that's what gets priced in. Whether they're right or not down the line doesn't really matter until down the line comes to pass.

The market is driven by headlines, it's driven by analysts, and it's driven ultimately by the "easiest" explanation and a much heavier weighting of what is here today rather than what could happen in the future (no matter how good of an argument you may have for that future), and as a result, shorting ARM in the face of this sentiment has been a very painful ride.

So, the lesson here? If everybody believes something to be true in the market, then it's true. Unless you are trying to time the short right as the inflection point happens (if it does happen), then going against the herd will get you fried more often than not.

Mistake #2: Timing

My thesis is actually somewhat playing out. ARM has been touting that Intel can't do low power, but all of the evidence that I have seen from technical sites and from first hand testing of devices has suggested otherwise. I mistakenly believed that this would be broadly known at this point (it is not), and as a result, it is still "widely known" that Intel makes "inefficient chips" and that "ARM" designs are inherently more power efficient.

However, in the market, all that matters are "dollars and cents", and until there is clear, convincing evidence that there are some legitimate market share shifts going on, then no matter how "clever" you think you are in trying to anticipate the shift, you will get burned. ARM is beating/exceeding analyst estimates, and that's all that matters for a momentum stock. As a result, ARM continues to march on higher, frying short sellers every time.

The sad part is that I knew that "Silvermont" (Intel 22nm Atom) and "Haswell" (high performance/low power "Core" part) would be 2013 events, and these are the parts truly at the center of my thesis, so it would have been wise to actually be long ARM until right about now, starting from ARM Tech Con in which the company started significantly playing up the server opportunity. That strategy would have been much more worry-free, and a lot more profitable at the end of the day, rather than the "losing sleep and watching the share price every day" that shorting the name has brought.

So, what's the second lesson? If you have a thesis involving a major industry/perception shift, then you can't try to anticipate it too early. It's like the people who predicted that Android would start eating Apple's (AAPL) lunch back in 2011. They were right, but getting in before evidence of margin erosion/market share loss was incredibly costly.

Mistake #3: It's Hard To Short Good Companies

ARM is a good company. Really. Their engineers are strong, their management is well focused, and the company does legitimately good things. This isn't a fraud, it isn't a scam, it's just - in my view - highly over-hyped and overvalued. With suspected fraud, it's easy to "short and forget" - it'll be exposed eventually. But trying to simply short market sentiment and/or in your view "unreasonable valuation relative to future prospects" is not going to go well in the short-to-medium term without a catalyst. You could eventually be right - just ask the Rackspace (RAX) shorts whom were recently vindicated - but it's a really unpleasant ride.

So, when shorting common stock, unless you have either evidence of fraud, or you have a suspicion that the thesis of market share shifts/margin erosion/whatever else you're playing for will actually materially affect the top and bottom lines to lead to a "miss" of some sort, then once again, the short will be quite painful.

Conclusion

While I haven't particularly lost my shirt here thanks to opportunistic profit taking/hedging into earnings reports, I have spent a great deal of time/effort on a trade that just isn't ready to work yet. I still believe that within the next year the stock sees a "readjustment" as expectations re: microservers, phone/tablet share, etc. are reevaluated in light of competitive threats from MIPS licensees/Intel, but it's just too soon to try to really make this call. Playing this with perhaps long term puts, or trying to buy puts ahead of each earnings call may be the safest way to do it, but I'm now convinced that the "evidence" needed for such a position to "work" won't really be here until late 2H 2013/early 1H 2014, but by then the share price could be $60/share.

I remain short a position in the common, but I have no delusions that it's going to pay off in any big way for a while to come...and only when I see evidence of such an opportunity will I increase my short exposure meaningfully. I will be opportunistic in selling puts to lower my risk, but this is no longer something that I expect to actually work in the near term as the momentum/sentiment is just too strong.

I've learned a harsh lesson in short selling. Being wrong in the near term when long isn't too bad - you can probably wait it out, perhaps get paid a dividend or two. But being wrong with a short position in a bull market and a secular growth story underlying the name? It requires much greater fortitude - almost as much as it took for me to compose this article essentially admitting defeat.

Source: ARM Holdings: Learning From My Mistakes

Additional disclosure: I am short ARMH.