We were waiting for at least a 7% price jump when we published our article Hard Drives, Floods, Monasteries and Investment Returns on Dec. 2nd, 2011 - good thing we weren't holding our breath. We thought the article looked pretty smart, not only because the Western Digital (WDC) investment had already returned 30.2% to investors by the time we were able to publish, but also because we had this really crazy $67 dollar price target on the stock (way higher than any other analyst at the time) - we thought people would listen.
Nonetheless, we felt bashful about our price target so we prefaced it with a bit of defensive language (from the article):
"We don't mean to scandalize the readers - but our $67 per share appraisal of the WDC's business on a day when shares were selling in the 25s had much less to do with Crystal balls than with deductive reasoning."
We went on to write:
"The shares of WDC have now climbed about 30% since we started buying just over two weeks ago (although, as pointed out they are likely nowhere near their real value, having already told investors on Nov. 16th what we thought the company is worth). Nonetheless we cannot relish in the glory because the article didn't actually get published before the rebound, but our time at Halki more than makes up for the ego fest we would have enjoyed otherwise."
Since we had to postpone our "ego fest" then, and the shares have now returned ~67% to investors, we thought this would be a good opportunity to make up for lost time.
|S & P 500||11/16/2011||1243.6||3/23/2012||1,388.70||11.7%||4.21||33.3%|
Getting Back to Holding our Breath
Presently we're less interested with the return achieved for investors (which annualizes to 190%), and more interested in the mystery of how analysts determine their upgrades, downgrades and future price targets - which the market then promptly respond to - we want to know how the analysts do it.
We considered a bad write-up full of misguided information, that we would then later do an "about-face" on, hoping readers would be more responsive to our writing - we entertained the idea for a while, but then had to dismiss the strategy on the grounds that it was unethical and since we are dealing with Wall Street operators, we realize we must be ethical.
In addition to overreacting to the potential long-term effects of the flooding in Thailand last year, there was doubt cast on the ability of WDC's management team to execute on its proposed acquisition of Hitachi Global Storage Technologies (HGST). We weren't buying it (the story that is). In our rough notes, published at the bottom of our Dec. 2nd, 2011 article we simply commented:
"Argument against the Hitachi unit integration: When have they ever not executed?"
Then we had an unmistakable regression and departed from the "high road" when we blurted out in the same notes:
"…we'll have to remember to send Needham and Co. a Christmas card for helping serve up such a great value."
From their November 15th report:
"Needham & Company downgraded Western Digital Corp. from Buy to Hold.
Needham analyst says, "We are downgrading Western Digital based on concerns that either 1) the company is successful in acquiring Hitachi but is forced to pay higher interest rates, issue more stock, or otherwise make concessions that limit the benefits of the deal and that the combined company's balance sheet is severely impaired; or 2) that the deal does not go through (which seems increasingly likely) and that WD is forced to expend considerable cash to recover its manufacturing capabilities, weakening its balance sheet (when combined with the losses), and loses access to the enterprise market and Hitachi's advanced head technology."
Two days after they wrote these words, we were out buying for investors accounts - it's like it was raining gold - hence our decision to add Needham and Co. to our list of Christmas card recipients.
The curious thing about point one above is that WDC had always planned to issue stock in order to acquire the assets of Hitachi (and the deal did in fact cost them more, something to the tune of 400 mln., making their March 22nd release - below - all the more curious).
Point two was completely unfounded, how was it ever "increasingly "likely" that the deal would not go through? Last time we checked Chinese regulators don't play Mahjong with U.S. based analysts. If they do, surely the SEC would like to know.
If the analyst had studied WDC post flood, how is it possible that they arrived at the conclusion that WDC would be forced to "expend considerable cash to recover it's manufacturing capabilities" as the amvona article mentions repeatedly, there is this thing called "insurance" which they may or may not have heard of, but the existence of which is plastered all over WDC's annual reports.
Even though the analysts were wrong, it appears market participants continue to eagerly await their proclamations - though they may constitute a complete reversal in opinion.
Consider the case of the Needham and Co. March 22nd upgrade:
"Needham & Company upgraded Western Digital Corp. from Buy to Strong Buy with a price target of $66.00.
Analyst, Richard Kugele, said, "We believe that the acquisition of HGST by WD, despite regulatory hurdles worldwide, will prove to be a pivotal point in HDD history and drive record profitability for both WDC and STX. As we have also argued in our Seagate (STX) thesis, with no need to price aggressively to maintain market share, we expect the group to operate rationally and within profitability metrics consistent with WD historically. Given the significant upside we project in our model to our new $66 price target, we upgrade shares."
In this case, some 4-plus months later they have said almost precisely the opposite of what was said back in November of 2011. About-face or not, the shares jumped almost 7% on the news.
When investors are lead to think a security is less desirable at the very moment it's pricing is most attractive, and told it's a "strong buy" once it has risen some 65%, it's a problem.
If Time Arbitrage Counts
Assuming the $66 PT is a good number, and in Needham's defense, their followers would still make a very healthy 57% return. However, amvona investors having acquired the shares at ~25.60 would make about 157% at $66 per share.
A Further Note to Investors:
On February 24th we wrote in a note to investors the following:
"WDC will (contrary to media/analyst reports):
- Jump on announcement of finalization of Hitachi merger (may be accretive in Q1)
- Announce return to full production before Q3
- Despite returning to 100% capacity, ASP's will remain drastically elevated for remainder of 2012
- Announce huge insurance payment in Q2 CC
- Jump on release of Windows 8 (probably Q3) - desktop resets
- STX mediation award for 650 mln. will be thrown out
The combination of these six factors should result in a much higher shares price - perhaps 60-70 per share."
Time will tell if the remaining five forecasts hold true as well.
A Hypothesis for Memory and Storage
We had intended to write a related article last week called "A hypothesis for Memory and Storage." We began by scribbling out some notes on March 13th , 2012 - the notes never became an article because of an intervening visit from an old friend - and rather than adding the notes to the growing pile of "unpublished articles," we thought we would add them here since they may be of some small benefit to readers interested in the industry. Along the way, we used these ideas as our basis for getting at least one investor out of a large stake in SanDisk (SNDK) writing in part:
"... just to clarify the advice given last Friday the 16th regarding SNDK - the (proposed) essay written last week involves the movement from device level SSD to enterprise SSD (the cloud) - this should be a steady transition over the next three years - while SNDK leads in "DLS", it is not at all clear how they intend to compete in enterprise (here) - whereas it is clear how the likes of WDC intends to compete.
It seems likely that device level storage will face similar dynamics (pricing pressure) as DRAMM has over recent years.
SNDK is a strong company financially (although nothing like WDC) with impressive operating results (not sure it can be maintained).
... have not done deep DD on SNDK, only cursory ... however, it does not represent a "value" at the current price, and the industry seems very shifty ...
It is for all of the reasons above, that the suggestion to sell SNDK was made last Friday."
Here are the notes for "A hypothesis for Memory and Storage":
HDD still low cost
- Not that durable - finite life span.
- Industry reduced warranties from five to two years.
- Increased areal density increases coincide with obsolesces (moors law) - buyers have three advantages to replacing storage - upgrade, reset, increase.
- Investors know SDD is important, even revolutionary.
- Investors know future devices will all move to SSD storage.
- The thinking is that cost is the prohibitive factor that keeps HDD relevant - this is true only in the short term - SSD prices have been falling and will continue to fall.
- What investors are missing is that devices are well established as conduit to information (the web), but what is not fully realized yet, is that they are more importantly now a conduit to storage.
- Storage at the device level will be no more important than the finite parameters of information at the device level. For e.g. who know how much information a device can store locally? The web of information is what matters. Likewise, who cares how much memory a device has locally, but matters is connectivity to enterprise storage.
- So the question is not how and when will devices all move to SSD - but rather, how and when will devices move to enterprise storage running SSD vs. HDD. This is a very important distinction when trying to understand the general direction things are going, and which companies are going along for the ride.
On the Nature of Devices
- People used to buy devices because they could "do things"
- Now they buy devices, because perhaps of a stable platform, and access to these small pieces of highly specialized software called "apps," which given their highly specific boundaries are almost the antithesis of the "web," so instead of a "www" or "world wide web" we now have a "aww" an "app wide web."
- The apps are not qualitatively better (than traditional software), but they do add tremendous diversity and choice to consumers, and often at little cost - further they seem to suit the needs and taste of a new generation of content consumers who have clear requirements.
- Therefore there are two roles for the use of modern devices - siloed information visa vie apps. And media storage - increasingly via the remote enterprise servers (the cloud).
- The battle lines on app platforms has been a very public affair, waged mostly between Android (GOOG) and Apple's (AAPL) ios. A point which is far less considered by the investing public however, is where is the battle lines in enterprise storage? Instead the rather simplified dichotomy is established between HDD and SDD - this of course is not the real question, as SDD prices continue to fall, they will be implemented in enterprise - the real question is who is best suited to profit from this increasing shift?
- For the device manufacturers such as AAPL, the key thing is not the device as "powerful data processor" (PDP) but the device as "recurring revenue stream" (RRS).
- Between May and September of 2010 the price of concerns such as STX and WDC fell precipitously for no other reason than the rise of tablet computers, and more specifically the launch of AAPL's iPad. This was a huge misstep however by the market. Of course the initial iPad launch predated the release of Apple's cloud service by about a year and a half, but it wasn't hard to see where things were going, since Amazon's (AMZN) EC2 and S3 services had been available for years already - in fact amvona and its related content search engines which we developed in 2006 were hosted on EC2.
- There was a reason that the new tablet devices came with surprisingly little on board storage (sometimes less than a cell phone), it's because the device was always meant to be connected not just to a net of information, but a net of storage. The net of information is hard to monetize, hence the rise of "apps." Likewise, too much storage on a device is a "one time" sales event, but dependence on remote storage carries with it the possibility of an ongoing revenue stream visa vie subscription fees.
- Companies like Apple look really complex at first glance, but is it possible that a big part of their future revenue may be derived from a business model no more sophisticated that selling remote (and proprietary) storage? And that their devices really represent a conduit to these subscription fees?