Micron Technology (MU) is scheduled to report their fiscal q4 '13 earnings after the closing bell on October 10, 2013. Analyst consensus is looking for a $0.24 in EPS (versus a loss of $0.24 in last year's q4 '12) on $2.7 billion in revenues for expected revenue growth of 38%. (The recent Elpida acquisition by MU which closed during the quarter, is only expected to have one month's financials in MU's Q4 13 results.)
I've always been a big believer that to understand where a company is going, from a structurally competitive standpoint, you have to know where it has been, and Micron and the commodity DRAM and memory makers have been plagued with structural overcapacity and supply issues for 30 years.
In the 1980's it was Japan that was accused of dumping memory on the market and in the 1990's while PC growth and the ever-growing demand for memory in PC's and technology drive secular demand for years, however eventually supply caught up and supply exceeded demand. (From July of 2000's high, to December, 2000's close, MU fell from $97.50 to $30.
Micron's moniker or nom de guerre (for lack of a better term) is that it was "an airline with a fab attached" which if you read our June '13 earnings preview around capex and cash-flow you would understand why MU earned this title: MU has regularly and consistently "destroyed capital" over the years as the capex required to sustain the economics of the of the memory business, routinely exceeded the company's cash-flow and profitability (net income), which makes you wonder about the economics of the business.
Which brings us to today's Micron: we are hearing from a number of analysts and producers that the "memory and capacity cycle" has really changed, and for the better, as with this upturn in memory prices and constrained or limited capacity, the memory manufacturers are acting rationally and keeping capacity mothballed.
Here is a quote from a Credit Suisse research report, dated October 3, 2013 and written by John Pitzer and his team, previewing MU's earnings, which talks about the capacity and supply cycle:
Supply/Demand Undergoing a Paradigm Shift:
We believe that memory supply/demand is undergoing a paradigm shift as improved supply/demand dynamics, industry consolidation, and slowing Moore's law will result in an improved and sustainable margins for the memory suppliers - recall the DRAM industry had seen excess capacity built out in 2007-2008, and industry capacity has largely been able to meet new demand through node shrinks. We expect 2014 to be the first year since 2008 where the DRAM industry will need to add new capacity to meet demand, as bit growth from node shrinks is unable to support growing bit demand. Assuming memory manufacturers warrant a 12% ROI, we estimate that new capacity additions at 25nm will be economical at ASPs of $0.71/Gb vs. current DRAM contract ASP of $0.86 and our CY14 ASP assumption of $0.66. We also believe that a more consolidated supplier base (only three DRAM suppliers) and slowing Moore's law - faster shrinks does not mean market dominance anymore - will discourage aggressive capacity additions. On the demand side, we continue to see strong growth driven by mobility (DRAM bits growing at 70% +CAGR) and Server (adoption of in-memory servers dramatically changes capacity requirements). We would also highlight the inelasticity of demand and the high margins downstream which can absorb higher prices as a reason for high optionality value in case of a significant supply shortage.
(We get access to Credit Suisse research as a Schwab advisor, and through ThomsonReuters. We consider the Credit Suisse research team as excellent and of high quality, and the above assessment of memory supply / demand is an agreement with others DRAM and memory analysts on the Street.)
Our own barometer for whether this is true or not is to watch the consensus change in analyst estimates for revenue and earnings per share (EPS) for MU and see which way the revisions are trending:
Here is the EPS estimates for the last 21 months:
* Source: ThomsonReuters estimate detail
Since the Elpida acquisition by MU will only contribute about one month's revenue in fiscal '13 (ended 8/13), note how the EPS estimate jumps for 2014 to $2.10, as of 9/30/13 and as of this morning, $2.14.
For the right reasons, the Elpida acquisition (a mobile memory manufacturer) is expected to be accretive for MU and given the secular growth in the MobileDram market, drive profitable growth and cash-flow at MU over the next few years.
Here is what consensus analyst estimates for the next few years:
|'14||$2.14||a lot||$14.4 bl||62%|
* Source: ThomsonReuters
Valuation: at $18 per share, MU is trading at 8.5(x) and 7.5(x) the consensus '14 and '15 EPES estimates for decent growth, although it is acquired growth from Elpida, and we don't yet know "organic" growth.
In addition, pre-merger, MU was trading at 7(x) price to cash-flow with very little free-cash-flow.
Technically, MU needs to trade above the Sept. '06 high of $18.65 and needs to do so on volume, to attract more interest. The SOX itself (the semiconductor ETF) is nearing its mid-2006 and mid-2007 highs near 550, which was the last decade's highs before falling to below 200 during the Financial Crisis of '08 and '09.
The chart below shows MU pushing against its 200-month moving average at $16.98 - $17. Trading above $17 and then $18.65 post earnings release, and on volume, would likely mean a longer-term breakout on the stock.
The Elpida acquisition has been well received by Street analysts and as the reader can quickly see from the numbers, Elpida will just about double MU's existing revenues, and add materially to profitability.
The questions remain however:
- DRAM, SSD, and NAND memory manufacturing remain a volatile business from both the supply and demand perspectives;
- Has there really been a "paradigm shift" in supply and capacity i.e. doesn't every cyclical business say that when times are good ?
- MU has lost money 4 of the last 6 years, according to our historical data. While Elpida will likely contribute nicely over the next few years in terms of Mobile DRAM, is this company still the same old MU ?
We have been long MU for clients since March, '13. It is unlikely we would increase our stake pre-earnings release on October 10.