Predicting Micron Technology's (NASDAQ:MU) quarterly earnings with a month remaining in the quarter is a fool's errand. EP is a fool, and this is his errand. <\/p>\n
A lot has happened since the 2Q earnings release on March 23. One of the best things is the appointment of Sanjay Mehrotra, the former CEO and co-founder of Sandisk as the company's new CEO. The mystery of who would replace Mark Durcan (current CEO retiring at the May board meeting), and when his replacement would be announced, was a significant dark cloud. Mehrotra was a model of clarity at Sandisk (despite his sometimes very thick accent!). He'd spend all day in analyst presentations with audio (and video!) that worked, enough seats for all would be attendees, piles and piles and piles of excellent slides, and excellent presenters from several layers of management. Then there was his wonderful, incomparable CFO, Judy Bruner, who would explain complex financials that would make you slap your forehead with the simplicity of some difficult concepts she conveyed with a minimum of pain. May Mehrotra bring that much better interface to the investment community! May we have well organized and presented and detailed analyst days such as Sandisk used to have. Ernie (current CFO) is far better than Ron (blessedly replaced CFO). But could there be a Judy in our future? Under his leadership, the Mehrotra-led Sandisk sported a multiple, using almost any conceivable metric, well in excess of peers. Mehrotra's Sandisk doubled from a level where I already thought it was fairly priced and was sold to Western Digital (NYSE:WDC) for a nice premium to that higher price. And we Micron shareholders won't talk about what's happened to WDC since this takeover. <\/p>\n
Lots of other things have been going on. Bernstein's Mark Newman believes contract length on chip pricing has lengthened out to a quarter. The eyepopping bids for Toshiba's NAND division have shown that lots of large users feel assured NAND availability is a \"must have\". Could disappointed bidders in that auction want a piece or all of Micron? I think so. And then there is spot and contract pricing of chips. DRAM spot and contract pricing have taken a breather after a multi month impressive run; DRAM spot remains above contract which is a bullish sign. And NAND prices, both spot and contract, continue to post impressive nightly upward moves. Check my instablog posts updating chip price action, which I won't repeat here. <\/p>\n
And what about market share? For the first time its pretty apparent that Micron is gaining DRAM share from both SK Hynix (OTC:HXSCF) and Samsung(OTC:SSNLF). Lets not go too crazy, but perhaps they are also currently gaining NAND share from those two and from Toshiba (OTCPK:TOSBF) and Western Digital. <\/p>\n
At the winter analyst day, on February 2, management presented their view that industry DRAM demand was running at a 20% CAGR (compound annual growth rate) and NAND demand is running at a 45% CAGR. I've taken the high end of the ranges in each case. <\/p>\n
And how are the competitors doing? For its 2Q ended February, Micron reported that its DRAM bits were up 1%. And here's Goldman Sachs on April 27th on the other two competitors, Hynix and Samsung who have just reported: <\/p>\n
Samsung reported that its 1Q17 DRAM bits were down low teens qoq (in line with guidance); Hynix's DRAM bit shipments for 1Q17 were down 5% qoq, in line to slightly below guidance. Samsung maintained 2017 guidance for industry bit growth to be in the high teens range yoy, and said that it will be in line with the market. Hynix fractionally (and this may be semantics) adjusted its DRAM view for 2017 demand growth to be \"a little over 20%\" from \"around 20%\", and for demand to be ahead of supply. <\/p>\n
So Micron is picking up DRAM share from both of its competitors. <\/p>\n
On the NAND side, Goldman reports that Hynix bits were down 3% QoQ and Samsung's were down \"mid teens\". So here too it seems that Micron is picking up share. Western Digital is expected to report as this is being written, so their numbers aren't known yet. Micron's NAND bits for the 2Q were up 18%. <\/p>\n
And yes, mathematicians and purists among you will point out that there is a quarter end mis-match here on the comparisons where Micron's quarter ended in February and the competitors quarters ended in March. But I think its safe to say something VERY good is going on with Micron and that they are picking up market share in both NAND and DRAM. When has that ever happened? Given that the pricing environment is probably very similar across competitors, I would venture that the market is giving gold stars to Micron's current offerings in DRAM, and in 3DNAND where it also appears to be the cost leader. Time will tell. <\/p>\n
And what about demand? Today's average sellside report would have one think the world is ending in chip pricing. Certainly Micron's and Hynix's low single digit PE's are suffering from this analytical gloom. <\/p>\n
But we've recently heard from HPE (NYSE:HPE), Cisco (NASDAQ:CSCO), and Seagate (NASDAQ:STX), to name a few, that their earnings have been hurt by the unavailability and high price of chips. Here's Seagate on the subject, in their earnings call of April 26th: <\/p>\n
Stephen James Luczo - Seagate Technology Plc <\/p>\n
Well, I think the NAND shortages are interesting because, at the end of the day, I think it's more challenging for the technology industry to deal with the shortages than it is maybe beneficial to HDDs because of some comparison on a 500 gig. I mean, the reality is even a 500-gig NAND drive, at today's prices or even at six months ago prices, aren't remotely competitive to an HDD price. <\/p>\n
I do think that the lack of availability of NAND in certain market segments results in people then shifting their strategies around do they use HDDs or not. So I think, for example, the NAND companies are constantly optimizing where do they shift their NAND. Does to go into phones? Does it go into the data centers? Does it go into the servers or does it go into the PCs? And depending on the grade of flash you're building, the capacity plans you put in six months ago and then what customers are asking for, there's this constant re-optimization of where the NAND is flowing. <\/p>\n
I think in the short term probably, and I think HP indicated this on their call two quarters ago, that they felt that the PC industry was being constrained a bit on NAND. I think that probably has shifted some longer-term strategy around product portfolios that breathes some more life into the HDD space, in that people don't want to be caught short with storage technology of any type. And, of course, there again we're talking about 128 or 256. <\/p>\n
A hedge fund friend reports that he heard on a conference call that SK Hynix is \"sold out on DRAM for 2017\". <\/p>\n
When are the sellside analysts going to get the message that the memory cycle hasn't turned (yet)? <\/p>\n
What about a 3Q earnings forecast? Enough with the yammering about boring stuff like company changes and market conditions! Where's that earnings forecast? <\/p>\n
With apologies to Bernstein's Mark Newman, whose lucid model format I've used, here are some basic model input and ouput comparisons: <\/p>\n
And here's the model, simplified and broken into three chunks of revenue, COGS, and consolidated P&L. First the revenue assumptions: <\/p>\n
Next the COGs assumptions: <\/p>\n
And finally the simplified consolidate P&L: <\/p>\n
Conclusions. For my 2Q model, most of the individual variables were off but directionally correct. One of the most glaring errors was NAND pricing which the company explained by a heavy mix of dense TLC parts. NAND prices have continued to zoom, so I'm once again leading with a solid price increase and this may prove the weakest part of this model. <\/p>\n
Hopefully I will have the time to tweak this before the end of the quarter. Hopefully any feedback will be specific, with reasons behind a grievance on any specific variable. <\/p>\n
So that's the fool's errand. Good luck to all. <\/p>\n
Disclosure: I am\/we are long MU, INTC. <\/p>\n
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article. <\/p>\n
<\/p>\n
The rest is here: <\/p>\n
Micron Technology Should Exceed Its Bullish Q3 Forecast - Seeking Alpha<\/a><\/p>\n","protected":false},"excerpt":{"rendered":" Predicting Micron Technology's (NASDAQ:MU) quarterly earnings with a month remaining in the quarter is a fool's errand. EP is a fool, and this is his errand. Continue reading