Marvell (MRVL) reported its earnings results on 11/15. The company has had a rough year, with the last several quarters involving year-over-year deterioration in revenue, free-cash-flow, and gross margins. In Thursday's earnings report, the situation was no different.
After issuing a revenue warning about a month ago that the initial guidance of $800M - $850M would be chopped to $765M - $785M, Marvell ended up reporting $781M in sales which was at the high end of the revised guidance, but a nearly 18% drop in sales year-over-year from $950M.
Earnings per share declined sharply from the year-ago period from $0.32 to $0.20, a 37.5% decline. Note that Marvell has been buying back its shares quite aggressively, so to get a read on the firm's operational strength, it is more important to look at net income. In the most recent quarter, GAAP net-income was $69M against last year's $256M, a staggering 73% decline. This was due to materially lower gross margins (59.3% versus 52%) as well as lower total revenues.
What's going on here?
Hard Disk Drives/PCs Are Weak, But Share Gains Strong
The PC sector is weak, and almost all of the component vendors in the space saw year-over-year declines in both gross margins as well as total sales (a notable exception was Nvidia (NVDA), which actually saw across-the-board increases). Marvell, which produces controllers for hard disks as well as solid state drives that primarily go into PCs, was obviously not immune.
47% of the company's sales are in storage-related semiconductor products, with the majority of the business coming from consumer hard disk drives (and to a lesser extent, solid state drives). The firm has - and continues to - talk up share gains against competitor LSI (LSI) and cites continued gains as the primary growth driver in the segment while hard disk TAM remains low.
While I am confident that Marvell's execution in the storage segment has been excellent despite the TAM, it is important to examine the industry as a whole to see if there is more permanent structural weakness in the hard-disk drive space. In particular, how big of a threat (or opportunity?) will solid state and hybrid drives be?
Solid State & Hybrid Drives - Boon Or Bane?
The most important thing to keep an eye out for is how well Marvell will be transitioning to take advantage of the eventual rise of "hybrid" drives - or hard disks that also come packaged with NAND flash as a "cache". Essentially, the user's data is kept stored on the hard disk portion of the drive and when data is accessed, it is moved to the smaller included NAND flash. For frequently accessed data, this scheme can essentially approximate the speed of a pure SSD at minimal additional overhead cost.
This type of solution is very useful for size/cost constrained machines such as notebooks and ultrabooks, since it very nicely optimizes performance and cost. I expect that is why Seagate (STX) and Western Digital (WDC) have been so bullish on this technology rather than pure-play solutions (in addition to the obvious desire to push their HDD expertise in the new era of compute). Marvell and LSI dominate the hybrid HDD/SSD space given their expertise in both required areas. In particular, take note of the following question/answer pair from the earnings call:
But when you talk about shipping a hybrid drive, are you shipping both a hard disk drive and a SSD controller? Or is there -- I mean is there a scaled down controller there? Roughly speaking, what are we talking about in terms of extra content in the hybrid drive for you guys versus a hard disk drive?
Yes. For the early -- as 'usually what will happen in the early phase of the hybrid adoptions, okay, you will have 2 chip solutions, one chip addressing the HDD, then another chip addressing SSD. So there are customers, they want full-blown SSD. Their customer want to have what you call the scaled down versions of the SSD. Now, eventually as this matures, this solution will be fully integrated. So what you'll see, there will be single-chip handling HDD and SSD, okay, and this will be -- ultimately, will be the lowest cost solutions for hybrid.
While the NAND flash players have been acquiring their own flash controller companies over the last year for pure-play SSDs, LSI and Marvell will be able to stay one step ahead of the pure NAND flash/SSD players by being able to sell a single-chip hybrid drive controller solution to the HDD makers. Should hybrid drives become the "dominant" solution and pure SSDs become a higher end luxury item, Marvell and LSI both benefit. Given Marvell's historical ability to steal market share from LSI in the hard disk space as well as the SSD space, it is likely that Marvell will be the dominant player in the hybrid drive space.
Networking - Steady, Solid, And Boring
Marvell's networking division generally performs solidly. In the most recent quarter, this made up 23% of revenues and came in at a 1% sequential decline from the previous quarter. The company is guiding for a low single-digit sequential revenue increase in this area, driven by growth in enterprise switching and a pick-up in infrastructure product sales as inventory corrections at its key customers play out.
Wireless And Mobile - The Missing Puzzle Piece
The reason that Marvell rallied from its 2008 lows of $5.80 to its 2010 peak of a hair over $22 was because of Marvell's presence in the "hot" Research In Motion (RIMM) "BlackBerry" phone. Of course, as RIM became an increasingly irrelevant player in the Android/iOS smartphone world, as did Marvell in the US smartphone space. To add further salt to the wound, as RIM readies its next generation BlackBerry 10 phones, it has deserted Marvell for Qualcomm (QCOM). This is due to the latter's lead in both baseband as well as applications processors.
Marvell has been fairly successfully competing in the TD-SCDMA smartphone market in China. The TAM here is quite substantial: 60 million devices in 2012 with an expectation of growth to 100 million units in 2013. While Marvell's chief competition here is Mediatek with Qualcomm ready to enter the market in 2013, the CEO describes Marvell's competitive positioning as the following:
Okay. So a simple, maybe like one sentence, will be Qualcomm performance for Mediatek price.
Marvell is taking a particularly bullish stance on its position in the smartphone market during 2013 by asseting that an internal goal for the firm is to reach 10% of the WCDMA market. It pays to be skeptical as to how close to this goal Marvell will be, especially as competition strengthens, but it is encouraging to see that management understands and is trying to capitalize on the growth opportunity here.
However that's the future. As of today, Marvell's mobile and wireless division made up 25% of revenues in the current quarter, and was off 10% sequentially. Further, the guidance was particularly bad with a forecast of down 30% sequentially (although a decline is in line with typical seasonality). It is encouraging, though, that the firm's new mobile platform has apparently signed on "about a dozen" new customers.
Conclusion - Wait And See, But Price Is Very Attractive Here
Marvell's balance sheet is strong with ~50% of its market capitalization made up by the $2.1B that it has in the bank (and it is debt free). The company's shares yield just over 3% at these levels and management is aggressively buying back shares at these prices. In fact, at the call, when the firm was asked about the viability of its buyback program in light of the weakened profitability and sales situation, the interim CFO gave the following answer:
We are absolutely committed to continuing the buyback at the current levels and potentially higher, right? We think the stock is at a very low point right now, and we will be very aggressive in that respect.
It also doesn't hurt that David Einhorn, of Greenlight Capital Fame, did not sell his stake in the company in light of the recent decline. In fact, according to Greenlight's 13-F, the legendary hedge fund manager yet again upped his stake in the firm after initially purchasing at the $14 level and adding in the $11 range last quarter. David Einhorn is a particularly good short seller, so if there were any fraud/accounting issues (which the Street seemed to fear upon the CFO's departure), he would be the man to have sniffed such a thing out before sinking well over $300M into the company's shares.