MU - Micron Technology Inc.

[Ross812]

Moved from General Discussion: (http://cornerofberkshireandfairfax.ca/forum/index.php?topic=3077.0)

 

On beerbaron's request under the investment ideas topic for start price; the price when I posted this thread initially on October 6, 2010 $6.76 - $6.96 (Low-High)

 

Hi All,

 

One of my hobbies is building custom computers for family and friends; in all, I have built close to 40 computers and newegg.com is making out with 20k in revenue!  In the last year I have discovered the benefits of the new Solid State Drive (SSD) technology.   Traditional (using traditional in reference to technology just seems wrong) Hard Disk Drives (HDD) have seek times of roughly 9 milliseconds versus seek times of .01 ms for SSDs.  The low latency advantages really add up when booting or loading programs.  Boot times are reduced from 45 seconds to under 15 seconds and large programs load almost instantaneously.  The performance gains from SSDs are such that I foresee all storage eventually moving to solid state in the coming years.

 

SSDs are created by placing NAND Flash modules in an array so they work in tandem to store information. Think 8 flash drives connected to a splitter so they work together.  SSD technology has come a long way.  Drive instability has been corrected by high quality NAND flash memory and improved memory controllers (processor on the SSD).  I started researching the manufacturing process involved when creating SSDs and discovered cost effective high quality NAND Flash is the commodity holding back the many SSD manufacturing companies.  Instead of buying expensive flash, the smaller companies have been trying to improve their memory controllers and utilize cheap flash memory.  This can only go on for so long because cheap flash memory (created with a  50 nm process) has a low memory density and limits the size of the drives and prevents costs from coming down due to raw material cost per gigabyte.

   

I discovered Intel and Crucial (Micron) are the only companies to use high quality 34nm NAND flash memory in their SSDs.  I did some digging and found Intel has access to their NAND flash memory from a company called IM Flash Technologies which stands for Intel-Micron Flash Technologies.  The IMFT was a partnership created between the two companies in 2006 to create the best flash memory available on the market.  The partnership is 51% Micron 49% Intel.  IMFT sells flash memory to Intel and Micron at cost while receiving intellectual property concerning the latest technology and manufacturing processes from both companies for their mutual benefit.  

 

I looked into Intel (INTC) but I think Micron (Mu) is a much better investment for SSD because they are not focused on the CPU market.   Mu stock appeared to actually be selling at a discount so I dug deeper to discover the reason for the depressed stock price.  The price per share appears to be depressed for two reasons.  An impending trial scheduled for January 2011 and poor forecasts for DRAM sales (their main revenue stream) next year.  

The court case concerns anti-trust (AT) issues and price fixing conspiracy between Mu, samsung, and Hynex.  The suit was brought by Rambus.  Rambus contends Mu, Samsung and Hynex fixed prices and pushed their DDRX ram to become the new standard over their own RD RAM.  Samsung and Hynex both settled out of court for 500 million and 280 million respectively with Rambus but Micron is still holding out.  They do not have a projected max loss in their annual report.  I have heard the maximum they could be liable for is 4B (1.3 billion subject to trebling) but the more likely scenario is a settlement for somewhere in the 300-500 million dollar range.  Rambus is by no means a victim here as they were conspiring with Intel to corner the market with their own RD ram but were foiled when DDR ram introduced on AMD based motherboards hit the market first and turned out to have better benchmarks than Rambus's product.  Mu may have been anti competitive but Rambus could be unable to prove financial loss.  Mu already settled a criminal case with the DOJ concerning the AT trial.  Mu agreed to pay 200 m in 67 mil installments over 2 years for the case.  

 

Links to legal documents and evidence:

http://rambus.org/legal

 

The DRAM market is projected to soften next year but Mu has revealed their plans to convert many of their DRAM facilities to produce higher margin NAND flash.  The flash market should explode as SSDs gain market share.  IMFT is the only company cable of producing their new 25nm flash memory (most SSDs are produced with inferior 50nm flash).

 

The best article I could find regarding the 25nm process is by legit reviews:

http://www.legitreviews.com/article/1179/1/

 

25nm technology will significantly decrease their costs to create flash memory as smaller wafers can be used.  

Another article from Daily Tech:

http://www.dailytech.com/IMFTs+25nm+NAND+Flash+Will+Cut+Production+Costs+in+Half+Spur+New+SSDs/article17549.htm

 

IMFT says 1TB 2.5" SSDs will be possible in 2010.  IMFT had problems when shrinking the 50nm process down to 34nm because they skipped the 43nm node.  This time 34nm to 25nm should scale easily without process problems.

Samsung released their 20nm class nand flash this April and reports they will start shipping the flash later this year.  Samsung says their flash is 20nm class but does not specify what nm process they are using.  Reviews have suggested that Samsung is still behind IMFT because if they were producing sub 25nm flash they would be sure to let the world know there is a new NAND flash leader in town.  

 

SO WHERE IS THE MOAT?

 

IMFT, 51% owned by Micron, has shown to be the leader in NAND flash technology.  At the very least their manufacturing process is 4 months ahead of their next competitor Samsung.  In actuality IMFT is probably at least 8-12 months ahead of Samsung's manufacturing process because IMFT started shipping 25nm flash in mid to late February 2010 and Samsung claims they will start shipping by the end of 2010 Oct-Dec.  This is assuming Samsung's technology matches IMFTs technology.  In true Intel fashion IMFT continues to shrink their NAND wafers on a 9-12 month time frame meaning IMFT should stay comfortably in the lead.  

 

Micron and Intel have the best nand flash out there but what is their competitive advantage?  If you read into IMFT’s articles you will notice the company was formed in order to combine "Micron’s leadership in process and product technology with Intel’s multi-level cell technology and history of innovation in Flash memory to successfully compete in the NAND Flash memory business over the past several years."  Micron has the brawn and Intel provides the brains.  Micron can acquire raw materials and equipment at reduced cost from their broad network of foreign and domestic suppliers.  Intel provides R&D to keep IMFT a step ahead of the competition.  The largest moat MU and INTC receive from their partnership is SUPERIOR nand flash memory AT COST.  Samsung can produce all the nand flash memory they want but they are going to mark it up before they sell it to OCZ, G-skill, A-data, Supertalent, Western Digital ect.  Thus far, the state of the art technology has been too expensive for all companies but Intel and Crucial (Mu) evidenced by other companies have opting for cheap and inferior 50nm nand flash memory for their SSDs in order to keep acceptable margins.

 

What I do not like about micron:

-DRAM and NAND Flash are commodities.

-EPS have been very inconsistent over the years as DRAM prices have fluctuated all over the place.

-The number of shares has escalated almost 300 million in ten years. (30%)

-Debt in 10 years has gone up about 200 percent.

-Book value half of what it was ten years ago.

A poster on another board brought up:

-“I don’t like it when management issues more shares, more debt, and Holds on to cash. Those three things make me very leery.”

 

Earnings over the last 10 years are inconsistent but if you know the landscape of the DRAM market they are predictable.  Major changes in RAM type increase margins and make the business quite profitable for 2-3 years.  This is followed by a time when margins decrease as the technology is being phased out.  Right now (as of December 2010) we are in the 2-3 year cyclical upswing in the RAM market (though computer sales are predicted to be down).  Add to this the introduction of the SSD.  SSDs will become a commodity product in 5-10 years but for the next few years Mu's moat is their joint partnership with Intel to produce the best flash memory in the world and use it in their SSD technology at cost which allows MU to produce superior drives cheaper than any of their competitors.  Mu's stock price is low right now because of uncertainty about the AT trial (which I think is overblown), the semiconductor industry being out of favor, and shock from a rough 2008 and 2009. If Mu, selling at $7 today, makes 1.9-2.3 a share 2010, 2011, and 2012 and holds $2 in cash a share with a tangible book value of $7.30.  Baring a catastrophic court ruling in the AT case Mu is worth at least 8x earnings.  Assuming they must pay 2B in damages (I believe absolute worst case) over 3 years; they are still making $1.34 a year. This yields a worst case stock price of 10.72 in 1 year.  I do not trust management whole heartedly but I believe the prospectus may warrant an investment.  I would be very interested to hear everyone’s opinions and I am very excited to join the forum!

[Ross812]

I'm sorry the replies are going to be hard to follow.  There was some good conversation about this company and I am going to copy relevant posts.

 

onyx1 said:

 

Quote from: Ross812 on October 06, 2010, 08:27:28 AM

  The performance gains from SSDs are such that I foresee all storage eventually moving to solid state in the coming years.

 

Welcome to the board Ross, and thanks for an insightful post! 

 

What about costs?  SSD clearly have performance benefits.  SSDs have taken over the handheld device market but that is because HDDs cannot compete with the compact nature of a SSDs.  Currently, desktop's and laptop's (where imporvements in access speed would be a big plus) OEM's are still using 2.5/3.5-inch HDDs because SSDs cannot compete with HDDs on cost.  For other applications such as storage area networks, video surveillance, DVRs & DVD recorders and long term storage of personal media content, it is unclear to me how the demand for compact size and/or access speed evolve if there is a significant tradeoff in cost. 

 

What do you see as the current economic tradeoff for the improvement of speed, and how will that change over the next five years?  Both WDC and STX believe that demand for SSDs will outpace that for HDDs in the coming years but will remain a small portion of the of the total storage market on a unit basis over the next four years.

 

FYI,  I have a position in WDC based on the thesis that overall storage demand will continue to grow, and that if SSD do take over the storage market, it will take many years for it to happen. 

 

ross812 (me) said:

 

Onyx1,

Right now SSD technology is quite expensive for consumers at $2.20/GB vs. $0.08/GB for HDD.  On a manufacturing level, I am not sure how far cost can be cut for SSDs.  There is still an expense for the raw materials and even with huge advances in manufacturing tech I’m not sure SSDs will ever get to 8 cents a gigabyte.  I may have overstepped my bounds by stating all data will be stored on SSD.  I think data for devices like DVR’s, file servers, etc. will always reside on the least expensive platform.  Will SSD ever be cheaper than HDD?  I’ll have to check into material costs but the answer is probably no.  An investment in HDD is probably not terrible but HDD companies will not warrant their traditional P/E ratio in the future.

 

The performance gains are worth $2.20/GB to me but I am probably the exception.  Our culture has many people who have no knowledge of hardware and others who buy a computer and all they know about hardware is: fast cpu=good, more RAM and HD space = good. Low latency arguments are lost on the vast majority.  I have watched SSD prices go from $4/gb to 2.2/gb in a year and the performance vs. price may become compelling to OEMs soon.  If two OEMs have computers sitting side by side at Best Buy, one with a HDD and one with a SSD, the consumer WILL be able to tell a BIG difference.

 

Once SSDs get into the laptop market, they are going to explode.  Desktops will follow once people come to expect the speed they get from a SSD.  The desktop market is already gearing for the SSD as the SATA 3.0 interface has been released on many new motherboards.  SATA 3.0 transports data at 600 megabytes per second and was necessary to keep motherboards up to SSD specs.  SATA 3.0 has been released solely for SSDs as HDDs don’t even saturate the SATA 2.0 benchmark. 

 

I think the ultimate use of the SSD will come when/if personal computing migrates to cloud computing.  In a cloud situation, time from startup to access to the cloud is the paramount.  How can you make a mobile device boot extremely quickly? A solid state disk. 

 

Myth465 said:

 

Interesting idea. I own WDC due to it being cheap and growth in storage. I think things will move to SSDs but it will be overtime, and people will still need huge HDDs for excess storage. I see some sort of Hybrid where SSD is used for the OS / programs, and HDD is used to store massive amounts of pirated material I mean emm content.

 

I was looking for a SSD producer to hedge my bets / pair my trade with and looked at Sandisk but moved on. How does this compare to them? It appears to be cheap on a cash flow basis, and has no net debt. They have had some huge capex though over the years. I am guessing its related to that factory.

 

Beerbaron said:

 

Very great post. I like the analisys.

 

On the financial side, my favorite question is: Does the company have a history of generating 1$ of market value for every 1$ invested?

That answer is definitely no: Over the last 129 months (10Y+9M) the company invested about 13B$ in CapEx and generated about 13B$ in operating profits.

 

On the subjective side my other favorite question is: Does the company have durable a moat?

Technologically, yes. But that moat comes at a very high price in CapEx.

 

One thing nobody seems to be aware is that there is physical limits for transistor size. Around 5nm tunneling starts to occur in transistors. So even using the best theorical technologies we can only increase the capacity 7 times from those 34nm wafers. SSD will probably never be as cost efficient as a standard hard drive unless the high grade silicon wafers go down drastically in price. My guess is we will have an hybrid as stated earlier.

 

So Ross, my question to you is:

Does this deal with Intel a complete game changer to increase Micron's return on investment?

What is the market going to look like in 10 years?

 

BeerBaron

 

 

 

[Ross812]

continued replies:

 

nodnub said:

 

ross, thanks for taking the time to post this and share your thoughts.

 

myth and beerbaron, you are right, and in fact there is already "Hybrid" hard drives in the market.  (eg. 500GB standard drive with 4GB NAND memory (SSD) included on the hard drive. http://www2.ncix.com/products/index.php?sku=53491 )

 

The big advantage of SSD is in mobile uses because there are no moving parts and are much less fragile than normal hard drives. So you will see them in a lot of netbook computers, laptops, tablet computers like iPad, some phones, etc.  I think this segment is tied pretty heavily to consumer spending.

 

Desktop computers at home and at work will be much slower to migrate to SSDs, as these are commodity products with low price as the major concern (top-level performance and shock damage protection are not major concerns here).

 

I think it will depend on how well the retail consumer spends.

 

TariqAli said:

 

Quote from: nodnub on October 06, 2010, 04:07:43 PM

The big advantage of SSD is in mobile uses because there are no moving parts and are much less fragile than normal hard drives. So you will see them in a lot of netbook computers, laptops, tablet computers like iPad, some phones, etc.  I think this segment is tied pretty heavily to consumer spending.

 

Desktop computers at home and at work will be much slower to migrate to SSDs, as these are commodity products with low price as the major concern (top-level performance and shock damage protection are not major concerns here).

 

I think it will depend on how well the retail consumer spends.

 

I think there are reasons beyond fragility to adopt SSDs. They make computers boot up and access programs/files on the hard drive at much faster rates. Friends with SSD drives can turn on the computer and be surfing the internet in just 10 seconds.

 

Myth465 said:

 

Quote from: nodnub on October 06, 2010, 03:58:47 PM

ross, thanks for taking the time to post this and share your thoughts.

 

myth and beerbaron, you are right, and in fact there is already "Hybrid" hard drives in the market.  (eg. 500GB standard drive with 4GB NAND memory (SSD) included on the hard drive. http://www2.ncix.com/products/index.php?sku=53491 )

 

Damn, I never thought they would put them on the same platform, but it makes sense. I figured you would have a 5 gig os hdd with a separate 1 tb hdd. This makes more sense and is the best of both worlds. Its also better for the HDD manufacturers.

 

My thesis is that overtime the storage companies can buy / R&D there way into the market and retain the desktop and corporate markets. They however will never have access to flash, mobile, phone, and cameras due to not having first mover / tech advantage.  It could be wrong though.

 

nodnub said:

Quote from: TariqAli on October 06, 2010, 04:53:54 PM

Quote from: nodnub on October 06, 2010, 04:07:43 PM

The big advantage of SSD is in mobile uses because there are no moving parts and are much less fragile than normal hard drives. So you will see them in a lot of netbook computers, laptops, tablet computers like iPad, some phones, etc.  I think this segment is tied pretty heavily to consumer spending.

 

Desktop computers at home and at work will be much slower to migrate to SSDs, as these are commodity products with low price as the major concern (top-level performance and shock damage protection are not major concerns here).

 

I think it will depend on how well the retail consumer spends.

 

I think there are reasons beyond fragility to adopt SSDs. They make computers boot up and access programs/files on the hard drive at much faster rates. Friends with SSD drives can turn on the computer and be surfing the internet in just 10 seconds.

 

Tariq, of course, you are correct.  but those advantages were already described by Ross in the first post. I should have said that I think the other big advantage is less fragile technology and that this means people will choose first and foremost to use these on laptops and mobile devices.

 

I have an old fashioned 5400rpm hard drive in my laptop and I can resume from a hibernate state in 15 seconds. (Hibernate is a power state that does not require any power at all and preserves all open windows on your desktop).

 

I agree that most people will find SSD performance a lot better, but I don't think it is currently worth it for most consumer or business desktops.  The current price versus capacity means that it is still something that only appeals to power users and early adopters.

 

Capacity is also an issue. Many of these SSD drives at a reasonable price point are only 120GB capacity, while most of my friends are already filling up much larger capacity drives in their current laptops and desktops. A lot of folks I know have 50GB of music, 60GB of downloaded movies, lot of hi res digital photos and movies. I shoot at lower resolution and never take video and I have about 10GB of personal photos over the last couple years.

 

So right now, at the current price, I think it appeals to niche users that are willing to pay a premium for performance and are willing to sacrifice drive space. I think this would be executive level employees and some computer afficionados (early adopters).

 

Also remember that NAND and flash memory has a finite number of write cycles before it fails (something between 30,000 to  100,000 write cycles per block). Peer to peer software like bittorrent that is constantly reading and writing to the hard drive will chew through this. The jury is still out on how reliable these will be for a base OS install over a period of several years.

 

I think hybrid drives are more likely to take off.  Or maybe some variation like small capacity SSD drives that are built onto the motherboard to quickly load key operating system files and main applications (maybe 30 GB), and then you would still have a normal internal hard drive for data as well.

[Ross812]

continued replies:

 

TariqAli said:

 

I think the ultimate winner is going to be SSD but who knows how long that transition will take. Most of my friends switching to SSD are power users (software developers, employees at tech startups). They tell me that even though you end up with less space, it is not a big inconvenience. They tend to use cloud services for listening to music (rdio, mog, pandora, hypemachine) and services like netflix/hulu for video. This decreases the need for a big hard drive.

 

NodNub I agree. Hybrids will probably do well for an extended period of time since people will still want big drives for backing up and archiving data. Seagate is a nice bet in that space given its current valuation.

 

alertmeipp said:

 

transformational technology often doesn't translate to transformational investment.

 

ericd1 said:

Great post Ross...Thanks for sharing...

 

I've been following a SSD company, STEC, Inc (STEC) -- They have a competing SSD technology and a "first mover" advantage. Their stock made a great run ($5 to over $40) last year, but has come back to reality ($13 w/P/E of 14x). Their SSD is designed to replace large enterprise (mainframe) HDD and they are working with the military on secure SSD products.

 

They also have an "embedded" small form SlimSATA and SATA-CF, up to 64GB and 32GB..."Increasing demand for high-performance and high-reliability SSDs within the embedded markets is a trend we are beginning to see and on which we have put ourselves in a position to capitalize," said Manouch Moshayedi, Chairman and Chief Executive Officer of STEC. "The introduction of our new lines of small form factor SSDs for embedded applications is in line with our strategy to help proliferate the use of SSDs across a broad range of applications. Just as we have helped to usher in the use of SSDs in the high-end Enterprise market, we intend to bring the latest SSD advancements to our customers in the embedded community, many of whom have been long-time STEC partners."

 

I don't think they are looking at consumer product applications, but rather the higher-end commercial SSD market.

 

I have been watching the company on the sidelines thinking they could be a buy-out candidate, but I don't have the industry insight to know if my thinking is right.

 

Appreciate your comments...

 

Ross812 (me) said:

Alright, I’m going to try to answer everybody! 

 

Myth,

Sandisk is very similar to Micron but they only offer flash based technologies and have no exposure to the DRAM market.  Sandisk developed their 32 nm flash in February 2010; so they are at least 12 months behind Micron in the flash race.  Sandisk trades at a p/e of 8.5 in a soft semi market and 15 in a strong semi market.  I think Micron will get the same treatment when all their legal battles are behind them.  I have not looked in to Sandisk at length but I think it is interesting Intel chose micron for their joint partnership.  Intel, the behemoth, wanted a company to supply raw materials at the lowest cost and chose Micron.  Many of Sandisk’s product offerings (mp3 players, flash drives, sd cards ect…) do not require good flash.  In order for Sandisk to really compete in the SSD market they are going to have to develop their manufacturing technology.

 

A comparison of product offerings:

Micron:

http://www.newegg.com/Store/Brand.aspx?Brand=1455&name=Crucial-Technology

Sandisk:

http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&Manufactory=1404&N=50001404&IsNodeId=1&SpeTabStoreType=0

 

Beerbarron,

Micron lowers costs by joining with Asian distributors and manufacturers.  In many cases, Micron buys controlling interests in the subject companies.  I believe the large capex charges you are seeing are due to acquisitions and factory startups.  Capex for micron may be elevated in the future because the company has plans to convert some of their DRAM facilities to NAND flash facilities.  As far as return on capital, you’re right, it stinks.  The DRAM market has not been a good place to be in the last decade which is why I think Micron is branching into the flash market.  Flash memory pricing can also be very variable but I hope with increased demand because of SSDs demand will outpace supply.  Flash has traditionally had higher margins because it is more of a consumer product than DRAM.  Most individuals do not go out to the store and buy a stick of RAM; it comes from the OEM.

 

I have no idea what is going to happen in 10 years, which is why Buffett stays away from technology.  10 years ago I was using a 1.8 GHz Pentium 4 with 256 megs of Ram.  The new droid smartphones with the Intel Atom processor can run circles around that computer.  I do know SSD will be big and I know micron/intel are the lead horses right now.  Intel has a history of keeping their lead.  I am hoping Micron can tag along and I think I can buy them today at a good price.

 

Nodnub,

This is the first time I have seen a hybrid drive.  It is an interesting idea.  A few problems though.  It is more complex than either an SSD or HDD so the chance of failure is greater.  The durability advantage of a SSD is negated by the inclusion of an HDD.  The storage problem is solved but the speed is not going to be up to SSD specs.  It sounds like the page file may be kept on the SSD and perhaps the ram will save to the SSD portion during hibernation mode.  This will help the speed on resume and searching but will not help boot up or large program loading.  The solution for desktops is to actually use two drives.  One SSD and one HDD.  Programs are stored on the SSD for quick access and media is stored on the HDD for economy.

 

I am not too concerned with durability.  Intel ran a test that simulated completely filling and deleting the data on their X25-M for 5 years without data loss.  I have not noticed a problem with bit-torrents (when downloading only legal files  ) and I don’t think there are more writes by copying a file from a bit-torrent or from another source.  I was under the impression the bit-torrent borrowed small portions of the complete file from each location and assembled the complete file on your computer on the fly.

       

Ericd1,

I’ll have to look at STEC when I have some more time.  I too looked at it but that was more than a year ago…

 

opihiman2 said:

Quote from: ericd1 on October 06, 2010, 06:54:17 PM

Great post Ross...Thanks for sharing...

 

I've been following a SSD company, STEC, Inc (STEC) -- They have a competing SSD technology and a "first mover" advantage. Their stock made a great run ($5 to over $40) last year, but has come back to reality ($13 w/P/E of 14x). Their SSD is designed to replace large enterprise (mainframe) HDD and they are working with the military on secure SSD products.

 

They also have an "embedded" small form SlimSATA and SATA-CF, up to 64GB and 32GB..."Increasing demand for high-performance and high-reliability SSDs within the embedded markets is a trend we are beginning to see and on which we have put ourselves in a position to capitalize," said Manouch Moshayedi, Chairman and Chief Executive Officer of STEC. "The introduction of our new lines of small form factor SSDs for embedded applications is in line with our strategy to help proliferate the use of SSDs across a broad range of applications. Just as we have helped to usher in the use of SSDs in the high-end Enterprise market, we intend to bring the latest SSD advancements to our customers in the embedded community, many of whom have been long-time STEC partners."

 

I don't think they are looking at consumer product applications, but rather the higher-end commercial SSD market.

 

I have been watching the company on the sidelines thinking they could be a buy-out candidate, but I don't have the industry insight to know if my thinking is right.

 

Appreciate your comments...

 

 

I have been following STEC too.  They are not big in the consumer space.  I believe Intel will lead that area, along with the traditional drive manufacturers and SSD drives.  What they are known for is enterprise RAID SSD storage.  They have proprietary technology that allows SSD to be more reliable across RAIDed storage.  And the system is fast.  The fiber adapters utilize proprietary nonvolatile caching to complement the SSD array.  I think STEC is going to be a buyout candidate by either DELL or IBM.  I'm looking to buy the stock.

 

 

 

[Ross812]

Continued replies (Almost done!):

 

Josh4580 said:

 

ericd...I see that you own NVEC...Whats your thinking on that stock and does it relate to this convo?

 

Ross812 (me) said:

 

Micron released earnings yesterday for Q4 2010.  Income was 342 million from net sales of 2.5 billion.  This compares to a net income of 939 million Q3 10 but 488 million were from the one time sale of a subsidiary Numonyx.  DRAM shrank 14% and Flash 9% from Q3.  Gross margins shrank slightly to 37% from 40%. 

Cash and investments now equal 2.9 billion. For FY10 1.9 billion in earnings (1.85/share) off 8.5 billion in revenue.  950 million in capital was invested and debt was paid down by 640 million. 

 

As of Q4 2010:

EPS Q4- 0.32

Cash- $2.93/share

Tangible Book Value- $8.17/Share

 

Samsung and Micron signed a patent agreement and Samsung agreed to pay Micron $275 million by March 31, 2010.  This should give a nice .27/share boost in Q2 11.  The Micron rep in the conference call was a little evasive when answering questions about what Samsung wanted.  The rep replied the agreement was beneficial to both companies.  The .32/share result missed consensus by .06/share but shares jumped 6.75% today.  I was actually hoping I could pick up more in the $6 dollar range but I guess irrational exuberance is today’s market for you.

 

I was struck by how commodity based Mu seemed in the conference call.  The company expressed it would like to increase its foot print in the consumer market to increase margins.

 

I was neutral on the outlook for Q1 11.  DRAM production bit growth- Up mid single digits.  NAND production bit growth- Up mid to high teens.  Total DRAM (think actual memory sticks not GB size) down high teen (should reflect conversion from DRAM to 25 nm NAND).  Trade NAND (excluding INTC) down mid to high single digits (this may mean they are using more of their own NAND….).  DRAM cost/bit- Down mid to high single digits. NAND cost/bit down mid to high teens.  The company expressed in the conference call that production cost would likely fall in the mid single digits.  This looks like more of the same to me… I thought they predicted neutral to perhaps a little poorer outlook from Q4 10 which is alright because the company is so undervalued.    Capex for FY11 is expected to be 2.4-2.9 billion; there was a question about this on the conference call and I remember thinking the answer was fine and made sense but I can’t remember it right now for the life of me!

 

I think it is going to come down to the Rambus court case resolution to really get this thing moving.  This company is worth 14-17 billion if all goes well.   

 

*Note:

Nodnub,

I thought you would like this article.  It seems Micron is providing flash to Hitachi for a Hybrid SSD/HDD.  “Micron's 25nm NAND process technology delivers 8-gigabytes (GB) of storage in a single device, enabling new applications that require high capacity storage in space constrained designs. The initial portfolio of HLDS Hybrid Drives will be offered in a range of embedded flash memory capacities including 16GB, 32GB and 64GB. Higher NAND capacities of the Hybrid Drive will be available in future versions.”

http://investors.micron.com/releasedetail.cfm?ReleaseID=513642

 

 

All, I apologize for the confusing repost in advance!

[jasonw1]

How much pricing power does MU have? Let's say they are 1-2 step ahead of their competition, will they be able to charge a premium for the same storage? Are consumers willing to pay higher price for that?

 

My suspicion is that most people don't even know the difference and probably don't care that much. For an average consumer, what comes to mind when I shop for a flash storage card? storage, price, reliability, performance comes after those.

[ericd1]

Josh - Belated response to your question about  NVEC - No it doesn't relate to Micron, but here's why I hold it...

 

I started looking around the nano-tech industry because I believe future developments could provide game-changing products. I wanted to invest in a pure tech company rather than a tiny division of a major like GE...NVE actually produces commercial nano-tech products and makes a profit. They also have no debt and are expanding their business. Growing and profitable made sense to me if you want to make a long-term investment in a high-tech industry--similar to BYD. They also hold a large number of patents and a lot of their on-going R&D is paid for by prospective clients (interesting business model)...Sales and profits growth has been solid and > P/E (under-valued). I read the quarterly transcripts and listened to recent calls. I don't know if it will be a big winner, but it feels good to have my toe in the water...BTW, it's a small position! 

 

 

 

 

[Ross812]

Jason,

 

I would have to say that Micron really has no pricing power with their ssd drives.  The reason I like this company is they make the best NAND flash available and produce it at the lowest cost.  SSDs will simply be the catalyst that causes huge demand for NAND going forward.  Will Micron sell more SSD drive than their competitors?  Its likely that some enterprise users will demand the best SSD technology which is provided by Micron and Intel so I believe yes.  Can any other company produce 25 nm flash?  No, and this allows micron to produce flash at a lower price per gig than their competitors.  Which companies are able to get raw materials (ie. silicon wafers) for the lowest cost?  Micron and Samsung, I believe Intel chose Micron to create IMFT because they are less diversified and can supply raw materials at a lower cost than any of their competitors.  I look at this play like Chesapeake for Natural Gas.  CHK is the low cost supplier as well, the only difference is there is a huge catalyst for Micron's product in the near future.  The icing on the cake is Mu is trading at a P/E of 2.5x if you back out their $3 pile of cash. 

[dcollon]

Do they really have $3 of cash?  It appears to me, net of debt, they don't have any cash.  I'm probably missing something though.

 

I look forward to re-reading this long thread on MU.  I have always viewed this company as a terrible allocator of capital, but it sounds like some things might have changed for the better.

 

Thanks for the posting the idea.

[Ross812]

http://biz.yahoo.com/e/101029/mu8-k.html

 

Seems like Micron is paying off some debt and refinancing at 1.85%.  This is good new for Mu and there should be a nice up tick today.  I am now watching the short interest; with 10% short, a move north should bring a lot of covering.

[Josh4580]

Thanks ericd

[Ross812]

The street thinks DRAM prices have stabilized and will start to increase from here.  Mu was upgraded by several firms after it was up 8% yesterday on 3x normal volume.  I'm looking for a large beat on the first quarter's earnings due to DRAM stabilizing early and the popularity of Mu's superior NAND Flash memory.  Mu expects to get a 50/50 mix with NAND and DRAM this year which should smooth out their earnings stream. 

[Myth465]

Ram is on sale. I have seen about 15 deals on Laptop Ram. Just bought some for my new laptop.

[Ross812]

Good point Myth, I should add- If anyone wants to build/buy a computer, now is the time.  Components are really cheap right now.  Feel free to message me if you want recommendations or need help building your own computer. 

[Myth465]

Ross do you game. Im thinking of going all laptop and buying a high spec one. Is there any reason why you still use PCs. I have always built my own PC and it will be tough not to do it again, but I am beginning to like laptop with docking station.

[Ross812]

Myth, I don't play very graphically demanding games, though I may get the new Call of Duty on my computer instead of on my x-box.  The advantage of a DIY desktop over a high spec laptop is the ability to upgrade the computer in the future.  I find the price point for desktop video cards is at the 1.5-2 year mark.  When the new state of the art video card comes out it will cost $400 but drop to $100-150 after the next volley of new cards come out.  With a laptop you are going to have a video card that performs well for a couple years, but to upgrade the card you need to buy another expensive laptop. 

 

My setup at home is a data/media server with several TB's of space.  A desktop in my office and a netbook.  Any work I do of the desktop or the netbook is synced with the server so I don't lose anything.  I like having a full desktop setup with 2 screens and a nice mouse and keyboard when I am really trying to get work done, and I like my 2.5 pound netbook instead of an 8 pound laptop when I have to travel.

 

It is all personal preference though.  If you want to game on the go, then a nice laptop is your best bet.  Whatever you decide, I hope you put a Micron SSD and Ram in it! :-)   

[Myth465]

Thanks for the insight. I too like the dual screen, Logitech mouse and keyboard setup. Its tough when I have to use my laptop.

 

Lol just put some Micron ram in my Laptop. I am too cheap for SSD though. I like to buy off cycle.

[Ross812]

Update: I sold yesterday at $11.03.  I am a little leery of the market right now and I wanted to raise some cash.  I felt that Mu had a nice margin of safety at $7.00, but I don’t want to hold it during a market correction at $11.00.  Micron grew to represent 25% of my portfolio and I can sleep much better at night now that I am 40% cash. 

 

Blasphemy Alert!!!

 

I would like to thank Jim Cramer for his Micron pumping last month!   

[Ross812]

Micron is down below $6 again. I haven't had time to read any of their reports, but from reading a few articles it seems they are moving from DRAM to NAND ram. Micron now supplies Intel, Crucial (Micron), and OCZ with their industry leading 25nm NAND ram for use in solid state disks. Yahoo is showing cash at $2.39/share and BV at $8.55.

 

I still like Mu from a fundamental business prospective. Tablets are using SSD technology and will drive demand for NAND ram. Tablet technology is advancing year by year like computer technology in the early 2000's and creates a market where consumers will upgrade their tablet every year or two. From an enterprise standpoint, I think we are 3-5 years out from seeing workstations utilize SSD. IT departments are slow to change and want to make sure the technology is proven before making a big change. What do you all think?

[BargainValueHunter]

Micron shares halted after CEO dies in plane crash...

 

http://abcnews.go.com/US/wireStory/micron-ceo-steve-appleton-died-boise-plane-crash-15508116#.Tyw4QOU_xOQ

 

Steve Appleton, the chief executive and chairman of Micron, died Friday morning in a small, experimental fixed-wing plane crash, the company said. He was 51.

 

Micron spokesman Dan Francisco confirmed Appleton's death in a release, and trading in Micron stocks has been halted.

 

Appleton, a professional stunt plane pilot and former motocross racer, was the only one in the plane when it crashed at the Boise airport.

 

Micron Technology Inc. is one of the world's leading providers of advanced semiconductor solutions. Through its worldwide operations, Micron manufactures and markets a full range of DRAM, NAND and NOR flash memory.

[cemadh]

I do not see any posts on Micron Technology (MU) since Feb 2012. Any reason why?

MU is supposed to close the Elpida acquisition at the end of July 2013.

 

There are some articles in the press that say that MU (post-Elpida) could earn between $2.25 to $3 per share in GAAP earnings next year depending on how DRAM prices and DRAM and NAND demand shape up in the coming months/quarters.

 

At $12.71 per share that equates to a PE ratio 5.65 to 4.24. Cheap?

There are other metrics that also seem to indicate that the Elpida acquisition and projected NAND and DRAM demand will be prove to be very good for MU.

 

Is it a good buy at this level?

By the way - thanks to Ross for starting this thread with an excellent write-up on MU way back in 2010. 

[ItsAValueTrap]

If you look into the history of Intel... it started off as a memory company.  Then they starting losing a lot of money in the business.  Andy Grove took a long time to get Intel out of the memory business... but fortunately for Intel he did.  Intel subsequently morphed into a very successful microprocessor company.

 

The memory business is scary and ugly.  The main competitive advantage is scale.  Unfortunately, it means that many industry players will overbuild capacity because they are all trying to get bigger.  Then everybody loses a lot of money.

 

The market leaders in memory are no longer the market leaders today.  Many of these companies have gone out of business over the years.  Historically it has been a crazy and ugly business.

[moody202]

Just checking what you guys think of MU given investment by David Einhorn and the recently Quaterly results!

[racemize]

It is also the largest equity holding of Baupost, probably worth looking into.

[dbuch]

I think Micron is definitely worth looking into. You have a consolidated industry with just three players in DRAM: Samsung, Hynix and Micron. The DRAM players will most likely not be focused on market share but ROIC going forward. Gross margins have already risen to 34% and could reach mid 40% very soon. The DRAM industry is expecting 27% CAGR through 2017 and supply is now matching demand which should mean very little price degradation. If demand rises 27% and prices remain stable (with oligopoly no reason they should compete over price) you'll experience 27% revenue growth. Micron is 80% DRAM 20% NAND so this should drive the stock. Micron should have $16B of revenue this year with near 40% gross margin which would imply near $3 EPS (near $4 EPS with current low tax rate). As demand increases at a healthy CAGR you should get very healthy incremental profits on each dollar of revenue growth. Assuming a 20% revenue growth over next couple of years should easily get you $5+ of EPS and FCF. Once the market believes the story I don't see any reason it couldn't trade at similar multiples as other consolidated industries such as the rails or hard drive manufacturers. I think this is what Klarman and Einhorn see and the reason they believe it is a low risk, high reward investment. I just wanted to check and see if anyone else has done any further due diligence on this.