With Friday's selloff behind us, there will be a propensity to scour the markets to pick up stocks that were beaten down. Some will certainly be great values, and others will be
perceived values.
On that second list I would put
SanDisk
(SNDK)
, which sold off 15% on Friday following a huge September quarter report the night before. As I noted about 10 days ago,
memory companies are names that really need to be avoided, and if there's any rebound in SNDK today, it might be a good time to get out if you haven't already.
The logic behind investing in SanDisk is based upon the enormous demand for NAND flash in all types of consumer electronics, as well has the potential for solid-state drives (SSD) to replace today's ubiquitous mechanical disk drives in some applications. All of this is true, but it is only one-half of the supply-demand equation that rules commodities.
As I noted, SanDisk had a huge quarter, with revenue up 38% year over year, $100 million above the Street estimations. Pro forma EPS was a blowout as well at 54 cents a share vs. the Street's expectations of 32 cents. The company shipped 54% more megabytes sequentially, and pro forma product gross margin increased 740 basis points from the second quarter, so what's not to like?
The graph below shows contract pricing from DRAMeXchange for MLC NAND since April. (SanDisk makes SLC parts as well, but the price changes are about the same for both.) For the three product densities, pricing was relatively flat for the second quarter until prices begin to move up late in the quarter. That price move was exacerbated by some yield issues at
Samsung that pushed prices to a peak in mid-August. As Samsung resolved its problems, pricing began to ebb as the back-to-school build slowed, and that softness continued into early October.
The upswing in prices helped gross margin. As noted by the CFO, the majority of the 740-basis-point sequential improvement was due to the reduced level of inventory adjustments (writedowns) as a result of higher prices. While it was not specifically mentioned, I believe the sale of products previously written down or off may have contributed as well. That has been a common practice in the past, and in the first and second quarters this year, the company sold $2.4 million and $6.5 million, respectively, of previously reserved products.
While prices may have helped inventory valuations over the summer, they are not something SanDisk has control over. Despite the extraordinary demand for NAND flash, it's still a commodity (regardless of what management argues), and pricing is a function of demand and supply.
The company acknowledged on the conference call that prices are falling faster than product costs. While the company is pushing its efforts to ramp production at the next geometry node (53nm) and thereby reduce its cost per bit, such shrinks also increase supply (more dies per wafer). Obviously, SanDisk is not alone in this effort.
In what should be the seasonally strongest period of the year, the effects of excess supply are evident in the table below. From the high point of pricing in August, I compared the price changes vs. the same period last year. As is readily apparent, the price declines this year far exceed what was experienced in 2006 for any part density.
I know that management's fourth-quarter guidance (revenue down 3% to up 11% year over year) was characterized as being constrained by product availability (wafers as well as assembly and test). However, price uncertainty and the inability to push more product through if prices fall faster than expected may well create negative surprises. Add to that the fact that prices would normally soften in another five or six weeks as the holiday demand winds down and we enter the weakest period for the industry.
NAND flash manufacturers can argue about the long-term opportunity all they want. But those are the same arguments heard in the DRAM and HDD industries for decades. The bottom line is the same for every commodity. When there's too much supply, prices are under constant pressure, and suppliers cannot reduce costs at rates approaching the price declines. And it is the fringe players controlling pricing, because they are far more likely to aggressively cut prices to gain share.
There is only one long-term solution, and it involves reduced industry capacity. Unfortunately, that's not going to happen as long as there is the belief in the proverbial pot-o'-gold at the end of the rainbow for those who survive the journey.
There will be times when the wind is behind SanDisk with all of the positive leverage that entails. But right now, it is sailing into the wind, and it will be for some time to come. While there is a price at which all stocks are attractive, I really don't believe that SanDisk is even close to such a point.
RELATED STORIESSanDisk Facing a Rough Road on Hard DrivesThe Worst May Be Over for SanDiskWall Street Shuns SanDisk's Blowout Quarter
