LSI Corporation (LSI) is a fabless semiconductor company that is the direct competitor of very recently wounded Marvell (MRVL). While SA's Mark Hartner provides a very detailed description of the finer points of a patent trial that could very well bankrupt the current leader in hard-disk controller technology, I view this as an opportunity to go long shares of LSI for the following reasons:
- Marvell will likely be severely financially damaged, if not completely crippled, as a result of this trial, which could allow LSI to take significant market share
- Marvell, in addition to damages, will now be forced pay a $0.50/controller license to Carnegie Mellon University, which could give LSI a significant cost advantage, further fueling a potential market share coup.
The Duopoly Nature Of The Market
In the storage controller market (and in particular hard disk drive controllers), LSI and Marvell are the two major players, with LSI trailing Marvell by a significant margin (58 to 30).
The company grew its market share position from 18% of the market to 30% from 2007 - 2011, while "Competitor A" (Marvell) also grew share to take 58% of the market. With very few remaining smaller competitors to take share from, LSI likely faced a significant roadblock in share gains here, as its competitor also showed significant prowess here. Further, the flat-to-negative growth nature of the hard disk drive business as PC sales have hit a slump has not been kind to the TAM [total available market] for such controllers. Thus, any growth -- at least in the near term -- comes from share gains at the expense of competition.
Size Of The Opportunity
According to LSI, the firm derives 37% of its revenues from the hard disk drive/flash end markets. The patents that Marvell violated are within the domain of hard-disk drive system-on-chip solutions (and not flash, PCIe, etc.), so we must do some legwork to try to determine the size of this opportunity to within a reasonable degree of accuracy.
From the $751M total HDD/Flash, we note from the most recently filed 10-K, 25% of LSI's total revenue of $2.044B came from sales to Seagate (STX), which sells primarily hard-disk drives and as such likely buys mostly hard disk SoCs and pre-amps. LSI does not break down the costs of these individually, so we will assume that pre-amps are sold bundled with the controllers, and then the opportunity comes out to roughly $500M. This is slighly conservative as LSI does act as a second source to Western Digital (WDC) [which is primarily serviced by Marvell]. However, working with the $500M number, and with the 30% market share number in the first slide above, we can work out the yearly TAM to be somewhere in the vicinity of $1.6B.
With some sense of the opportunity, it is time to try to measure it.
The opportunity for share gains is predicated on the following assumptions:
- Marvell's cost structure becomes unfavorable in light of the $0.50/unit royalty that LSI does not need to worry about
- Customers become less eager to use Marvell's products over a competitor's in fear that product support may not be there in light of a potential bankruptcy
Each percentage point of market share gain represents an additional $16M to the top line with this model. Assuming that LSI is able to capitalize on Marvell's misfortune and capture 50% of the HDD controller market, then this could add $320M to the top line over a 12-month period, representing 15% revenue growth!
There are a number of risks to this thesis:
- Marvell could fight the verdict and this could be in court for years before any of these consequences take effect
- The sentence could be amended to remove the rather high royalty (the biggest risk to the thesis)
- The share gains will likely be realized slowly over a protracted period of time rather than in one explosive "burst," limiting investor enthusiasm
- LSI may also be guilty of infringement (although no case appears to have been filed at this time)
Should Marvell walk away from this in a much better position than it is in today, then the opportunity for share gains here may not be as pronounced, but the worst case is that there is no incremental upside here, rather than anything particularly catastrophic for LSI (which has been, as demonstrated, gaining share).
In short, there is a very real opportunity for LSI to expand its market share in one of its core businesses at the expense of Marvell due to the latter's patent trial loss. In a duopoly with a flat-to-negative near-term TAM, market share shifts are critical, and if LSI can offer fundamentally the same product at a better price, it stands to gain significant market share.