Seeking Alpha
View as an RSS Feed

Retired Securities Attorney  

View Retired Securities Attorney's Comments BY TICKER:
Latest  |  Highest rated
  • Micron: A Tough Quarter Before A Solid Second Half [View article]
    Thanks, 203,

    32 layer, higher node, longer wear and more reliability and cheaper too.
    Mar 26, 2015. 02:37 PM | 1 Like Like |Link to Comment
  • More Woes For Intel: Chip, Chip, Chipping Away At Chipzilla [View article]

    Here's a starting point for server TCO.

    You also need to take into account the cost of the power needed to run the server and the [additional] cost of cooling down the space in which the server is located. Servers can run very hot.

    Server farms, which are large clusters of servers, typically used by operations such as google, Amazon, etc. have different Cost of Ownership parameters. As you put more and more servers in an enclosed space the cost of the electricity needed to run them and run the cooling system to keep them cool becomes more significant comparatively.

    Then there is the cost of downtime. You build in redundancy for that but redundancy costs money.

    At higher, hotter nodes ARM chips have an inherent advantage in TCO because they run cooler. At lower nodes like 14nm that advantage dissipates to a small percentage of the TCO.
    Mar 26, 2015. 02:10 PM | 1 Like Like |Link to Comment
  • Micron: New Projections And Price Targets [View article]

    >>If we ignore currency trauma for a moment (I'm far too dumb to understand it) and assume everything besides PC DRAM remains flat wouldn't Micron need a 90% drop in PC DRAM revenues to hit .54/share?<<

    Sound logic. Let me free associate and ramble a bit about it.

    I don't know about your assumption that all else remains flat. People buy less cell phones when they have less disposable income. Less everythings, not just PCs.
    And that assumption about non-PC sales is at least partially a function of macro factors like the current currency chaos and the effect $40 oil has on disposable income. And those are functions of specific geographic factors. I am not able to even get some directionality with all those variables flying around, much less quantify them. So I am using $.50 to $.60 for the quarter rather arbitrarily to see how my long term thesis fares with that range while my head keeps spinning about it. I am also assuming that we have another quarter of world-wide currency chaos and unpredictable consumer spending due to the large effect of $40 oil on jobs, disposable income, etc.

    But I don't think that the chaotic macro environment can last more than one more full quarter.

    By then, on the demand side oil storage capacity will have completely run out if normal supply/demand factors haven't already stabilized it. The supply of oil may not react to market pricing before storage capacity is exhausted.
    The sunk costs (the already spent and gone costs of drilling an oil well and bringing its production on line) are no longer relevant to the decision as to whether to continue to produce an already dug well or to cap it. And the geography is relevant. So is the downtime refiners use mostly in the spring to make repairs, clean, maintain, etc. Those refiners are "storing" a lot of oil while refining it. This affects disposable income. I am not smart enough to figure out HOW is affects it but WHETHER it does is easier for me to see.

    On the supply side, Micron's node shrink costs in both time and money should be clearer by then. Whatever the cost in time and money is, it will pass.

    So my long term thesis still holds. Demand for memory will be going up faster than supply can come online to meet it. That means higher profits for the producers and higher profits for the producers is all I can be sure of right now. But it is enough for me. I'm patient. I'm much less concerned with when than I am with whether.

    Now let's take a quick look at that currency trauma you mention.
    On March 15, ten days ago, you would have needed 1138 South Korean Won to buy one US dollar. Today you need only 1104 South Korean Won to buy one US dollar. That's a more than 3% move in just ten days. If the price, or the profit, (of anything) is moving that fast just because of currency fluctuations no one I know can predict any numbers with much accuracy.
    But I can predict that the chaos will end in no more than another quarter or two. I can predict that the world is getting more and more technological and that means more memory demand for phones, tablets, refrigerators and cars. And I believe that the capacity does not now exist to fill that predictable demand increases; nor do I believe that there is any feasible way to build it in time to fill demand in 2015, 2016 or 2017.

    I actually do not see supply being able to meet demand in the foreseeable future without some kind of a Manhattan Project for memory or a basic technological change which is nowhere yet in sight.

    So I don't know what the numbers will be. But I don't care as much as most of you seem to.
    I often bet on "more", "bigger" or "better" without knowing how much. It has worked fine for me so far. Better than "fine" actually.
    Mar 26, 2015. 12:27 AM | 8 Likes Like |Link to Comment
  • Micron: New Projections And Price Targets [View article]

    I haven't updated any price targets.

    The dollar and the price of oil are question marks with unknowns both in magnitude and direction. Without some sense of what is going to happen in those two arenas, I haven't got a sufficient handle on the market for equities to project prices.

    I will update my targets when I get a better sense of the dollar and the price of oil.
    Mar 25, 2015. 12:46 AM | 3 Likes Like |Link to Comment
  • Micron: New Projections And Price Targets [View article]

    >>1) Strong dollar actually works in Micron's favor (for now anyways).<<

    I don't agree with that. Care to share your reasoning?
    Mar 25, 2015. 12:35 AM | Likes Like |Link to Comment
  • Micron: New Projections And Price Targets [View article]
    123, Bobcat,

    Thanks for the compliment.

    I try to keep my projections roughly continuously updated. Sometimes, it's hard - like now.

    The major problem is the currency markets.

    Costs: Micron makes its products in various places, US, Japan, Singapore, etc. And there is Inotera, all of the production of which Micron buys at a rolling price. So Micron's costs are either determined in, or, at least, heavily influence by, the value of the local currency. Those values have been moving too fast for me to quantify the effects of those movements.

    Pricing: Micron can price its product anyway it likes buy has to compete with producers (Samsung, Hynix, Sandisk) whose costs are in differing local currency and whose margins are thought of as being (YYY Remingbi minus XXX SK Won).

    I have been thinking (wrongly) that this would all slow down and stay stable enough for me to get some numbers in US dollars that I was happy with.

    I was wrong. Nothing has stabilized much yet.

    OK, currency is one problem. Is it predictable?
    Over the longer term, it is. Each currency's value will stabilize in terms of another at a level which reflects how the commodities on which the world runs impact it. Right now that comes down to oil, a big one, and it won't hold still long enough for me to get a fix. So I can't predict it. All I can do is guess.

    I am guessing that oil has not bottomed yet since the world hasn't yet run out of places to store the surplus being produced. But almost. I think the bottom comes after a sell-off down to reflect newly produced oil having no place to go wait in order to get sold. At that point producers have to cut back on producing well already in production where the marginal cost of producing them is small compared to the "sunk" costs of drilling them and bringing them to production.

    Near term there are some business issues. One, Micron is moving to a 20nm node in order to stay competitive with other producers, primarily Samsung, and that causes downtime at plants while equipment is changed over. Two, sales for use in new PCs are expected to be off since new PCs aren't selling at the projected rated.
    I consider both these issues to be shorter term in nature and not endemic to Micron. They will affect the next two quarters but then dribble into insignificance (I think).

    On the good news side, demand for memory products continues to increase at a rate above the level of supply increases due to organic (non-new construction) reasons.

    So with all that said, you should now realize how confused I am and no longer be interested in my numbers. But here is where I am right now.

    For the current fiscal year ending August 27, 2015, I am now using $3.78 in EPS. This coming quarter will "disappoint" because the Street's estimates have not reflected the downtime or the lower PC sales. I don't know how low the quarter could come in but I am using $.50 to $.60 myself.

    Further out the story is intact. Demand will continue to outstrip supply for the foreseeable future which is good for prices, profits and investors. This assumes that Samsung with its lower cost structure will keep prices at a level that maximize Samsung's profit which I believe is still the only reasonable assumption to make. Samsung's lower cost structure will squeeze Micron's margins for as long as Samsung remains technologically ahead in production.
    Mar 24, 2015. 12:29 PM | 11 Likes Like |Link to Comment
  • A Logical Reason Why Apple Will Use An Intel Modem [View article]

    >>ARM chips are now achieving core i5 levels of performance in a phone!<

    I don't think so.
    Mar 23, 2015. 05:59 PM | 3 Likes Like |Link to Comment
  • Intel: Mistakes And Opportunities [View article]
    Thanks, Bruce. Nice job.

    Now we need you to take each area of interest for Intel (like Servers) and break down its characteristics separately.
    Mar 23, 2015. 11:01 AM | 1 Like Like |Link to Comment
  • Micron: Extreme Undervaluation And Massive Growth Potential [View article]

    Goldman has the net capital not to care.

    I think market mechanics more than regulation control bear raids.
    Mar 20, 2015. 04:32 PM | 1 Like Like |Link to Comment
  • Micron: Extreme Undervaluation And Massive Growth Potential [View article]

    >>They can or could purchase the shares... or they don't<<

    True, with two qualifications.

    One, not buying in the failed shares affects the firm's net capital adversely which lowers the firm's ability to do profitable business.
    Two, if the firm knows this is the kind of bear raid you postulate, and they would know, they'd buy the shares in to stop the raid. Unless they are also short but, if so, what are they doing with a high enough bid in the market to buy the shares shorted in the first place.

    The bottom line is that bear raids do not work in today's regulated market. The used to, but not for a long time.
    Mar 20, 2015. 02:39 PM | 1 Like Like |Link to Comment
  • Micron: Extreme Undervaluation And Massive Growth Potential [View article]

    >>I was just wondering what data to use for coming up with '% of the company's total float is being shorted'.<<

    Me too. Worse, I wonder what data is being used to determine the "float".

    If by the "float", you mean the shares available to be traded in the open market, then what do you exclude? All shares held by "insiders"? Including the ones they can legally sell? What about those which are restricted because they were never registered but aren't held by insiders?

    The bottom line is that no one really knows what the float is. There are lots of ways of making more or less educated guesses about it though.
    Mar 19, 2015. 06:04 PM | 1 Like Like |Link to Comment
  • Micron: Extreme Undervaluation And Massive Growth Potential [View article]

    When someone shorts a stock a fails to deliver, the buyer (the buying broker) can purchase the shares not delivered in the open market and deliver them to itself. Thus, in theory, a "fail" generates a matching purchase to the short sale and balances the downward pressure with an equivalent amount of upward pressure.

    That is called "buying in" the failed-to-deliver shares.
    Mar 19, 2015. 05:58 PM | Likes Like |Link to Comment
  • Micron: Low Insider Sales May Buttress Second Half Turnaround Thesis [View article]
    No, but I haven't read all of them.
    Mar 17, 2015. 10:27 AM | 2 Likes Like |Link to Comment
  • Micron: Low Insider Sales May Buttress Second Half Turnaround Thesis [View article]

    This is what happens when you use secondary data and do not check your sources.

    >>Over the past three months there were no buys by insiders and 273,426 sales; sales were 100% of all transactions.<<

    Like much else in this article, this is just not true. Let's use the last in the three month period since the date of the article as an example.

    On December 15, 2014, Joel Poppen received 4,167 shares of stock free from Micron's 2007 Equity Incentive Plan.
    He did NOT sell them. He did, however, let Micron retain 2,057 of them to pay the taxes due on the issuance. He kept every share which was not used to pay taxes.

    Next is this one:
    Durcan gave 30,000 shares to charity. Not exactly a sale.

    Then this: Bailey sold 2,000 shares of Micron out of the 124,172 shares he beneficially owned.

    Next this one:
    Poppen received 6,667 shares of stock free from Micron's 2004 and 2007 Equity Incentive Plan. He did NOT sell them. He did, however, let Micron retain 3,869 out of them to pay the taxes due on the issuance. He kept every other share which was not used to pay taxes. Probably in a shoebox.

    That's the latest few. The rest are similar. You can find them here:
    Mar 16, 2015. 05:58 PM | 15 Likes Like |Link to Comment
  • Micron: Low Insider Sales May Buttress Second Half Turnaround Thesis [View article]
    >>That said, management executed stock options representing about 1.5 million shares over the past 12 months. That is more than half of all shares sold over the period.<<

    First, executing stock options is buying the shares, not selling them.
    Secondly, if those shares are sold, what exactly did they represent more than half of? Not the open market trading which is about 20 million a day.
    Mar 16, 2015. 05:27 PM | 6 Likes Like |Link to Comment