As Cirrus Logic (CRUS) saw its stock price plummet on Thursday following CEO Jason Rhode's comments at the Barclays' conference reducing long-term margin guidance from 50-52% to the mid-40s, we were reminded of Apple's (AAPL) massive clout in the marketplace, despite its diminished stock price. With almost 40% of the tablet market and 20% of the smartphone market, Apple wields incredible power. However as Cirrus was cratering, we were also reminded of the cliche, what comes around goes around. Due to the fact that Samsung (OTC:SSNLF), the largest memory company in the world, is also Apple's largest and most aggressive competitor, Apple stands to face intense margin pressure caused by rising memory costs. Ironically, Micron (MU), a long-term Samsung rival, stands to uniquely benefit from Samsung's new stance towards capacity growth, and from the overall battle for market share in the smartphone and tablet segments.
Lest anyone think we're late to the Micron bandwagon, in writing today's positive piece on the stock, we'd point readers to our vocal buy call on November 5th, when shares had closed the previous trading day at $5.71. At the time, we suggested a price target of $9. On March 7th, with shares at $8.94, we revised our price target to $13.
We now expect shares to rise to $16-18 by year-end.
The Smartphone/Tablet Wars driving memory tightness
While Apple has demonstrated an ability to exert significant price pressure on some of its largest suppliers, such as Cirrus, which garners 85% of sales from Apple, we expect them to have little ability to pressure broader-based commodity suppliers. The fact is that Samsung, Huawei, LG, Nokia (NOK), Sony (SNE) and others are attempting to make a huge land-grab in the smartphone market, with Amazon (AMZN), Asus, Microsoft (MSFT), Samsung and others attacking the iPad in the tablet market. Both smartphones and tablets should enjoy dramatic growth in 2013 which has benefited, and will continue to benefit, component suppliers.
Notably, memory pricing has been firm, with widely reported price increases for DRAM earlier this year, and stable pricing for NAND. SDRAM pricing is expected to rise 10% in May, and according to recent reports, NAND is supposed to rise over the summer.
What is particularly shocking about the price environment for memory, is that all of this is happening in a seasonally soft time of the year. In addition, in this particular year, demand has been further reduced due to Apple's widely reported inventory overhang and 1H production slowdown. Despite this, component supply is extremely tight.
With Samsung prepping a 2H onslaught of new phones to go with its April Galaxy launch, an array of new offerings from large players, and Apple resuming mass production with new launches of the 5s and a low-end phone, we can envision significant tightness and rising ASPs, if not profound shortages for memory, as well as potentially for LCD displays, driver-ICs and other components. This tightness should be further exacerbated by a resumption in growth in the notebook space prompted by Intel's (INTC) introduction of the Haswell processor, which buoyed by its improved performance characteristics, should revive interest in the moribund category. Typical 2H PC seasonality and the Haswell introduction should drive a resumption in DRAM ASP increases in 2H.
Expectations of rising DRAM, mobile/specialty DRAM, and NAND pricing all bode very well for Micron's FY14 earnings to move significantly above the current street high estimate of $1.24 - we believe estimates will exceed $2 within the next 3-6 months.
Samsung with a new memory strategy - this time it's different.
Despite the many positives entering 2H, typically we'd be far less sanguine about Micron's prospects than we are today, given the memory industry's historic propensity for imploding by adding too much capacity just when things were getting good. We are confident that this time it's different. Specifically, three things stand out:
1) Samsung realizes it is a superior strategy to make moderate semiconductor capacity additions to keep component prices high.
2) The memory market has become an oligopoly.
3) Capacity adds are becoming more cost prohibitive and more challenging technologically.
Historically, Samsung would aggressively attack the memory market by adding capacity to maintain/grow high market share. With profitability increasingly being driven by smartphone/tablet sales rather than by semiconductor sales, it is less important to maximize memory market share. It has now been widely reported to be sourcing devices from outside suppliers. In addition, increased market share does not necessarily lead to increased profits, and capacity additions directly benefits smartphone competitors.
As a hypothetical example, let's say we believe NAND ASPs may rise as much as 20% by year-end. However, if Samsung increases its NAND output by 10% (which we acknowledge is not possible given equipment lead times) ASPs would stay flat. In this example, Samsung makes fewer dollars, and has to outlay more cash towards cap-ex. In addition, it lowers the price that competitors need to pay for components.
The truth is, probably the smartest thing Samsung can do competitively is to dramatically slow its capacity expansion. Under such a scenario, the majority of Samsung's components would still be sourced internally, at increasingly lower cost (due to die shrinks, etc), while competitors would see pricing rise and further compress margins. And this is what we expect will play out.
In such a scenario, in our view, a likely scenario, Apple will see further gross margin degradation (not in sell-side models in our opinion). The ridiculous margin Apple currently captures by charging $100 for an incremental 16GB of flash storage (for capacity increase from 16GB to 32GB) or for 32GB (for a capacity increase from 32GB to 64GB) will compress. We recognize that Apple has negotiated certain preferred supplier agreements, but we understand that these are at a discount to market - so if market is rising, Apple's costs will go along with it. And given Apple's numerous competitors who will all be vying for limited supply, Apple can't slap around commodity suppliers the way it did Cirrus. Based on this, coupled with the introduction of low-cost iPhones, we are not optimistic regarding meaningful margin improvement for Apple in coming quarters.
Further bolstering this strategic shift from Samsung is that the memory industry is now essentially an oligopoly, with just Samsung, Hynix (OTC:HXSCL) and Micron playing. With 3 players, the motivation to get aggressive simply isn't what it had been for decades. The difference in a couple of points of market share, simply isn't worth it, especially if it stands to harm Samsung's fastest growing business, smartphones, which were up 60% y/y in 1Q according to Gartner, by lowering costs for competitors.
Finally, we believe this could be a multi-year phenomena. It simply isn't as easy technologically or as low cost to add capacity as it once was. Each next generation shrink is more complex, and yields less cost savings. At the same time, fab tools, including technologies such as EUV are becoming more expensive. The eventual move to 18" wafers will only further heighten the challenge of adding capacity - and could place suppliers in a position where they request deposits or financial contributions from customers ahead of capacity adds.
Micron is the big winner
Based on our expectation of tight, perhaps extremely tight (reminiscent of late '99 tight) component supplies into the smartphone/tablet supply chains (and some overlapping segments of PC and TV - i.e. driver ICs, DRAM, displays) we believe component names are the right place to be entering 2H. Given the significant run many stocks in this segment have enjoyed - for example, we like Corning (GLW) - we believe much is baked in. Still, valuations do not appear to be extended, and in the event that lead times extend, ASPs increase, and shortages are reported, we could see shares rising dramatically from current levels.
While Apple laid the smack-down on Cirrus this week, we'd expect that to be an exception, as it faces rising ASPs from many commodity component suppliers.
We particularly like Micron, because it appears to be perfectly situated in the midst of the smartphone/tablet boom, in addition to PC/notebook improvement. Further, we have little doubt that as positive as Micron management has been regarding the benefits of the Elpida acquisition, it is dramatically understating its actual benefits, lest it give credence to the claims of objecting creditors. However, with the deal looking weeks away from closure, the implications to future profitability will soon be reflected in street estimates. While we believe many buyside models are positive on Elpida, we think the upside is far greater than all but the most optimistic views.
Finally, as an idea closely related to Micron and the themes of this article, we continue to recommend shares of ChipMOS (IMOS) which is up about 50% since our March 25th article, and $1.20 away from a 5-year high. Micron is its second biggest customer, at 14% of revenue, and we believe that could double within 12 months, assuming the successful close of the Elpida transaction. In addition, DRAM, specialty/mobile DRAM and driver-ICs, all beneficiaries of the smartphone/tablet trend, make up the vast majority of IMOS's revenue. Finally, despite the 50% move in shares, IMOS still trades at exceedingly low multiples. Specifically, based on our expectation of $220 million of EBITDA (1Q was $45.5 million), $90 million of free cash flow (1Q $17.3 million), share count of 29 million, and net cash of $33 million, shares trade at 2.2x EV/EBITDA and sport a free cash flow yield approaching 20%. In addition, there are multiple near-term catalysts, but that's a topic for a future article.
Additional disclosure: We conduct thorough research on our ideas, but our views are our own. Please do your own research. We are short AMZN.