Why I'm Buying Micron- Part II

This post is a follow up on my earlier promise to provide a thesis for adding a small Micron (MU) position to the Secular Trends Portfolio. With Fiscal Q3 earnings out Friday as well, I’ll also prep for that and give a modest report after hearing the conference call.

Many readers have been wondering, quite bluntly, “Why the hell to you like this POS company that hasn’t made a dime in 11 straight quarters?!”

Well, I can’t argue with those comments; Micron has been a corporate failure in almost any investing metric for the past two years. As to why, take your pick:

1) Massive oversupply of DRAM thanks to commoditzation, and what essentially amounts to subsidies from certain Asian governments (yeah, I’m talking to you Taiwan!)

2) Weakening global PC demand, The Great Recession, etc…

3) dovetailing from #1, the entry into the market from every Tom, Dick, and Harry (or whatever those name equivalents are in Asia) because memory chips are the easiest to copy and spit out (Intel (INTC) learned this lesson over a decade ago).

But yes, I do have an investing thesis here. I’m not supremely confident that Micron is the prime beneficiary of it, as I don’t have a PhD in Techie and don’t care to begin training for it now. (Read my earlier post on NVDA (NVDA) to hear why I don’t think this really makes for a smart tech investor anyway, or view the performance of my Technology allocation versus industry benchmarks here).

There are several promising developments in the wastelands that have been commodity tech hardware:

1) Micron is a partner with Intel, not a competitor (this is so important for a weakened tech company like Micron that if there was a #0 on this list, it would go there)

2) Several smaller players (Qimonda AG (QI) & Spansion Inc. (SPSN)) have already declared bankruptcy, helping to thin out some of the herd. In addition, major S. Korean competitors Samsung (SSNLF.PK) and Hynix (have announced major production cuts, as has Micron. Is it enough to meet with the demand level of the much-hyped “new normal”? I don’t know yet, because we’re not there yet.

3) Micron recently completed a $450 equity & convertible notes offering - a major accomplishment a this time, in this particular market. MU now sits on over $950 million in cash with minimal debt coming due in the next 12 months (about $350m). Granted, MU may need all this cash just to wait out the storm, but they at least have some breathing room. Samsung and Hynix, meanwhile are adding to their debt and may need to conduct offerings of their own in the near future (this is coming from their creditors, not the companies - that’s never a good sign).

4) Micron was able to unload the worst of their 3 divisions, the CMOS images sensors company (Aptima Imaging Corp.), to a private equity group earlier this month. MU retains a 35% share and production credits, but they dumped a real dog here. I don’t think the division ever earned a taxable dime. It’s someone else’s capital issue now; Micron is just running the plants. A good move, so long as you don’t care about turning a profit, which investors shouldn’t at this point.

5) DRAM and NAND flash pricing has improved measurably in the past few months. Now here is where I get into tricky territory, because I frankly can’t keep up with the technology cycles here. And more importantly, I don’t want to get into prolonged aruments with others over how this product performs better on this custom system or that, etc. There are dozens of sites and blogs devoted to just that - I know because I read several of them. Not a one has ever provided me with an actionable investing idea. That said, I’m always willing to learn more, so feel free to comment on my overall thesis but please, please, keep it all in context.

6) Industry catalysts - From all that I hear, the release of Windows 7 may finally do what Vista could have done if it didn’t happen to be a complete abomination. A successful launch of “7″ would force PC users around the world to upgrade their RAM en masse. I don’t know the exact specs, but I’m sure the new Windows will be a RAM hog. In addition, a new release of Windows Server should help Micron in the same way on the server end (servers are 20% of Micron’s business and rising).

This should spur OEMs to invest in some inventory building as we head into the all-important Fall buying period.

Also, there is increasing interest in 3G smartphones, digital movie cameras, and portable computing/storage/game/music devices of all sorts. The secular trends here are strong and growing. That’s what we like at Epiphany Investing.

Solid-state devices seem to be the future of memory, and Micron should be able to carve out a space for itself and possibly establish a new pricing trend. DRAM has been an utter disaster, but there’s only two ways the future can go. Either all flash memory products become a government-subsidized version of packing peanuts, or they become an innovative tech product again.

I’m making a small bet on the latter.

Earnings Report was after Thursday’s close; I’ll report after reading through the CCall…

Read 'Why I'm Buying Micron- Part I'

Comments

[Ames Tiedeman:]

LSI, ONNN, RFMD, CRUS, and NVDA make much more sense than MU.
All you are getting is an old commodity DRAM maker with MU.

Jun 28 10:02 AM

[amdman:]

DRAM technology is reaching the limits of how small the capacitors that hold the charges can be scaled down. Other technology has caught up so the embedded folk have moved the memory (EDRAM, etc) onto there own chips instead of buying separate DRAM chips.
In another ~2 years or so all the chip makers will switch to RRAM, which can be done as a top layer on existing chips.
DRAM is as dead as NVDA, a zombie on the chopping block once Intel moves graphics onto the CPU.

Jun 28 12:59 PM

[CPST1:]

A deep article on the state of the DRAM/CMOx/NAND/

Of note: Samsung's Greatness/Samsung has led and dominated every memory market they participated in, for as long as they wanted

www.denali.com/wordpre...

Jun 28 05:04 PM

[CPST1:]

Memory business model is broken

eetimes.com/news/lates...

Jun 28 05:08 PM

[PA:]

You definitely want to run and read Raymond James writeup on MU Friday.......
U.S. Research
Please read domestic and foreign disclosure/risk information beginning on page 6 and Analyst Certification on page 9.
© 2009 Raymond James & Associates, Inc., member New York Stock Exchange/SIPC. All rights reserved.
International Headquarters:
The Raymond James Financial Center | 880 Carillon Parkway | St. Petersburg, Florida 33716 | 800-248-8863
Micron Technology, Inc. June 26, 2009
(MU:NYSE) Company Comment
Hans Mosesmann, [phone], Hans.Mosesmann@Raymond...
Semiconductors ______________________...
MU: DDR3 Transition Starting - Advantage Micron
♦ Thursday after the close Micron reported May quarter EPS of $(0.36), above our $(0.53)
and the Street’s $(0.43) estimate primarily driven by a $242 million benefit from sales of
products written down in prior periods and significant reductions in manufacturing costs.
May quarter sales were $1.1 billion, up 11% sequentially and in-line with our estimates.
♦ This was a solid recovery quarter for Micron as demand trends in DRAM are good and
suggest seasonality entering the second half of the year.
♦ Core DRAM ASPs continued to improve, however specialty DRAMs (non-PC) continued to
languish keeping overall ASP’s in check. Micron gained DRAM market share in the quarter.
♦ NAND memories were a mixed bag with trade ASPs up significantly offset by lower ASPs
overall given sales to JV partner Intel (which benefited from lower NAND wafer expenses as
Micron sells to Intel at cost). Unit demand is seasonally weak entering the summer.
♦ The CMOS sensor business is being sold and will become less of a drag. This benefits our
operating expense assumptions going forward.
♦ Micron expects PC demand to be fairly seasonal going forward and has not seen any
evidence of channel accumulation of inventories as all players are cash flow conscious.
♦ Interestingly, Micron believes that ~30% of global DRAM capacity that was “turned-off”
earlier in the year and only 5-10% may be coming back. Current DRAM prices are still
below the cost of the industry laggards.
♦ More intriguing is that the bulk of that 30% industry capacity (and some that was not taken
off) is not viable in producing DDR3 DRAM. Hence, the decision to bring this existing
capacity back to the market is one that is complicated by the fact that DDR2 has begun a
decline into obsolescence in PCs.
♦ The shares may trade off somewhat today due to recent upward estimate revision, but the
structural changes in the DRAM industry that are driving the current consolidation will, in
our view, lead to memory undersupply in 2H09 and 2010. Micron's commentary confirms
this worldview, in our opinion.
♦ We are raising our price target to $15 per share from $14 based on a mid-teens P/E
multiple to our new CY10 EPS estimate of $1.05, which we view as conservative given that
the consolidation dynamic set to play out over the next one to two years is one that has
never occurred in the history of commodity memories. We continue to recommend the
shares as a Strong Buy for more aggressive investors.

Jun 28 10:19 PM

[Ryan Barnes:]

To CPST1 - I appreciate your posting of those articles; I am always on the lookout for brain food.

But you'll forgive me if I'm not convinced that the $18b market in flash is "broken" because of an article that chiefly cites CEO Darrell Rinerson of Unity Semiconductor, a "next-generation memory hopeful". Can anyone say "bias"?

Hey, this realm of technology is all about change; what works today is bound to be upended tomorrow. We know this. And as soon as Unity Semiconductor gets a $100m OEM deal for their "passive rewritable crosspoint memory array" technology, I hereby promise to figure out just what the heck that means.

To amdman - appreciate your comments as well, and will try to read up on EDRAM and RRAM at some point in the future. I happen to really like NVDA, and pointedly disagree that they are a "dead stick". Intel is obviously one helluva scary competitor, but they don't win in every market they enter; NVDA is a large holding in my model and some of my thoughts on GPU vs CPU architecture can be found here :
epiphanyinvesting.com/.../

Intel would love to prove NVDA CEO Huang utterly wrong on GPU, but I am a believer. So far the tests show that a CPU carrying all the heavy lifting of processing and graphics simply runs too hot for smaller devices.

Healthy competition is always a good thing. As with all, time will tell; best of luck in your investing efforts.

PA; thanks for the posting; $1.05 in Fiscal 2010 would be quite the feat, but hey, I'm all for the occasional bout of optimism.

EpiphanyOne

Jun 29 12:35 AM

[PeterB:]

Puzzled:- "Thursday after the close Micron reported May quarter EPS of $(0.36), above our $(0.53) and the Street’s $(0.43) estimate "

I will trade you $0.36 for your choice of $0.53 or $0.43 as many times as you like.

Jun 29 05:09 AM

[Roger Knights:]

Micron is the target of a huge anti-trust lawsuit by Rambus. Trial date is September. Check the Rambus discussion board on Investor's Village for more.

Jun 30 08:48 AM

[spqrusa:]

Aptina was their only profitable division for 6 consecutive quarters... Aptina will continue to be profitable - that is why they are keeping a 35% stake in the spinoff. Duh.

Jul 07 12:37 PM