Seagate (STX) designs, manufactures, and markets storage drives for both consumer and enterprise usage. In particular, the company is known for producing standard hard disk drives as well as hybrid (hard disk + NAND flash) drives for PC and data-center use. The commonly accepted belief that is reflected in the company's low-as-dirt valuation is that hard disk drives are on the road to obsolescence at the hands of flash storage.
In this article, while I address the quite-justified concerns of the bears, I also offer a counterargument for the long-term viability of the company (and even its core industry) going forward.
The Shift To Flash?
It is very easy to list the advantages of traditional NAND flash over today's magnetic platter hard disk technology: speed, shock resistance, physical size, and power consumption are all clear advantages for solid state drives. However, the cost per-gigabyte of NAND flash far exceeds that of the traditional hard disk drive components. To put this in perspective: a typical 1 Terabyte (1000 Gigabyte) hard disk is $80, but a 128GB solid state drive costs about $120.
While NAND flash prices will come down and cost per gigabyte will improve, the hard disk drive manufacturers also continually make improvements in platter density, which means that for the foreseeable future, users looking to get the most "bang for their buck" will undoubtedly choose to go with an HDD.
However, there are a couple of main arguments that Seagate bulls need to address:
- Smartphones and tablets, which use flash memory, are becoming much more popular. Will this negatively impact the number of hard disks sold?
- Solid state drives could become substantially cheaper on a per-gigabyte basis, especially as smaller geometries of NAND flash emerge, so won't solid state drives hit the "good enough" point in terms of local storage, negating the raw price-per-gigabyte advantage?
- As laptops move to more "thin and light" form-factors where size, battery life, and tablet-like responsiveness are a major concern, is it likely that traditional hard disk drives will no longer be required?
Case #1: Consumers Shift Away From Hard Disk Drives
Let's address the worst-case scenario: customers move almost entirely away from hard disk drives in their notebook computers and/or move to flash-based tablets. This means that users now have to contend with 16GB - 128GB of primary storage. This should be sufficient for storing the user's apps, a few documents, and a game, but what about consumers that like to enjoy things such as:
These people (which, I would guess, is most) have two options:
- The Cloud/Streaming: As fast internet connection speeds have become ubiquitous in the US, users are more likely to rely on streaming their content from service providers such as Amazon (AMZN) or Netflix (NFLX) rather than keeping the data locally on the machine.
- External/Remote Storage: People with a good amount of data that they need to keep stored locally often buy several external hard disk drives to offload the bulk storage.
In the first case, the cloud service providers have scores of hard disk drives storing all of that content, which means that Seagate wins. While these datacenters also have some sort of flash acceleration (think Fusion-io (FIO)) to give nearly the speed of an all-flash solution, the bulk of the storage is on inexpensive hard disk drives. In the second case, Seagate wins by selling external hard disk drives.
Case #2: Consumers Go Hybrid
I noted above that many datacenters utilize flash caching technologies in order to get full leverage from the investments in both HDDs and flash technology. The same principles can be applied to the consumer space with hybrid solutions. These come in two forms:
- A hybrid drive that, in one package, fuses both hard disk drives and a small amount of NAND flash
- Combination of hard disk drive and solid state drive, tied together with a software suite
As I stated above, flash is expensive. However, the next generation of mainstream PCs/Ultrabooks will need to compete with tablets on responsiveness if they want to stand a chance of surviving in the long run. This means that some sort of flash solution is going to be required, but at the same time, PCs need to be able to offer an advantage when it comes to raw storage capacity at a solid price point.
I believe that hybrid solutions, whether in a single drive or in an HDD/SSD pairing, are going to be the future of the mainstream PC. Don't just take my word for it, though. Intel's (INTC) "Ultrabook" specification mandates some sort of flash solution to be included in a computer in order for it to get the "Ultrabook" designation. A look at the list of Ultrabooks shows that the SSD + HDD combination is quite popular at the lower end (but not so much at the higher end):
Further, for more specialized uses such as:
- Video editing
- Photo editing
- Music creation
- 3D art
The amount of data that is needed/stored is enormous. Professionals will continue to need as much local storage as they can get, and solid state drives just aren't going to cut it.
The case for hard disk drives is clear: bulk storage is needed somewhere, whether it's on the client side or on the server side.
A Solid State Future
That being said, Seagate is not blind, and is doing the right things to future-proof its enterprise business. The thing with flash storage is that in the commodity parts of the business (i.e. consumer SSDs), the guys with the NAND fabs win, since NAND is a commodity. Think Micron (MU), Intel, Samsung (SSNLF.PK), and Hynix. To "win" in this space without a NAND fab, the differentiation really needs to be at the total solution level (in particular, software).
Seagate has made a few very sharp strategic moves to insure its long-term viability in enterprise flash:
- The company recently invested $40M in Virident, a maker of PCIe SSDs, in a bid to offer a line of flash-based PCIe products.
- In mid 2012, Seagate announced a strategic partnership with DensBits, a maker of solid state drive controllers, in order to develop future enterprise flash solutions.
Clearly the company is aiming at the high margin, high ASP segments of the market that do not depend necessarily on having the best cost on NAND. Seagate will be a player in enterprise flash, securing its long-term position as a complete storage provider.
The stock is cheap, and the short sellers were desperate to cover on that drop to the $33 level, quickly sending shares back to the $35 zone.
The company continues to increase its dividend (and given EPS of $5.75 projected for the year, that $1.52/share dividend has a lot of room to grow), buy back an enormous amount of shares, and is well positioned for the future. The 36M shares short will eventually be squeezed as Seagate proves its long-term viability (they already seem desperate to cover), so buy any major dip whenever you get the opportunity. There is plenty of demand for this stock whenever it gets cheaper.
I'm keeping this one on my watchlist and plan to pounce on any major pullback.
Disclosure: I am long INTC.
Additional disclosure: I may initiate a long position in STX at any time.