Tech Roundup – January 8, 2009

NAND flash spot prices retreated slightly with 8gb and 16gb MLC trading off ~4%. DRAM contract prices for 1H January were posted with zero change from the previous 2 weeks. As Nanya and Samsung had publicly discussed price hikes, this is somewhat surprising. It's tough to push price hikes through with demand not there to back it up.
Powerchip, a struggling Taiwanese DRAM manufacturer, submitted an aid request to the government. They remain opposed to consolidation which is in opposition to the government's preference. As Powerchip is on the brink of failing, its unlikely they will ultimately have much of a choice in the matter – expect them to be shotgun wedded to another player.

ThinkEquity expects Intel's Q1 guidance to represent the bottom in downward revisions. He has a 54 cent estimate for 2009. Consensus is 71 cents. Jeffries believes Q1 guidance will be down 15-20% q/q and lowers his revenue estimate to $6.84 bil. His revenue estimate for the full year is 10% below the street and his 09 earnings estimate goes to 63 cents. Intel will shrink by 15-20% this year and the stock sports a P/E far in excess of 20* on even the most optimistic estimates. The stock belongs $10-12.

Lenovo issued a profit warning. Shares plunged. Global PC demand is poor.

HONG KONG, Jan 8 (Reuters) - Lenovo Group <0992.HK>, the world's no. 4 personal computer vendor, said on Thursday it was likely to incur a material loss for the three months ended in December 2008 and would cut staff by about 11 percent to reduce costs. The potential loss was due to unprecedented global economic challenges reducing demand for personal computers, the company said in a filing to Hong Kong bourse. Lenovo also announced a plan to cut 2,500 employees worldwide, or 11 percent of its total workforce, as part of an effort to realise annual savings of $300 million for the year ending in March 2010. The company said it would incur a pre-tax restructuring charge of about $150 million for financial year 2008/09, which would be largely reflected in the fourth quarter of the year ending in March 2009.

Taiwan's UMC Corp, a major foundry for worldwide semiconductor production, announced sales dropped 45.5% y/y in December. This is somewhat worse than expectations but reflects the well-known contraction in semiconductor orders. Foundries are an important bogey to watch for an upturn as they'll see business improve when chip inventories get too low. No sign of that so far!

EMC Corp preannounced revenues slightly above the street consensus. Non-GAAP EPS looks in-line. EMC is likely to outperform near-term relative to tech in general. Generally, online storage is becoming more pervasive as bandwidth speeds have grown. Pushes to so-called cloud computing by most major technology vendors (MSFT, GOOG, AAPL, IBM) will continue to bolster ongoing storage needs. EMC has been showing good cost controls and operating leverage under its present management, remains a potential acquisition target and probably ought to be a core holding.

Microsoft (MSFT) has been aggressively cutting search deals. Yesterday it came to light that they had beaten Google (GOOG) for the Verizon search deal – beaten is always debatable with deals like this as they come with minimum payment guarantees – this one seems to be $550 mil over 5 years. Microsoft also won Dell's search business – several years ago, Google overpaid for the rights to default search settings on Dell computers. More to the point, Microsoft is making it more difficult for Google, regardless of the P&L implications to Microsoft. Google's market share in search significantly dwarfs Microsoft.

FBR pulls the plug on some semicaps. His comments follow:

We are downgrading the semiconductor manufacturing industry as our recent channel checks in Asia, as well as our semiconductor inventory analysis (illustrated inside this report), indicate the recovery in semiconductor manufacturing fundamentals will be slower than currently dialed into Street expectations. Although we were late in 2H08 in downgrading, and actually did not find it prudent to downgrade when stocks hit 52-week lows in November '08, we believe the recent enthusiasm in semiconductor manufacturing stocks and the associated 25% runup in SOX are premature and, thus, are using this occasion as an opportunity to finally downgrade. Our analysis indicates that the semi inventory refresh that typically occurs in the 2Q time frame (every year) will be muted this year since our analysis indicates that forward DOI (throughout the channel) will continue to increase from 3Q08 into 4Q08 as the rate of sell-through continues to track below the rate of sell-in. This is expected to drive forward DOI to above 40 days in 4Q08 and, thus, set a new five-year high. Consequently, the work-down in semi inventory will be longer than what has been experienced in the past few years and, thus, 2Q's inventory refresh will be a muted one. This is expected to push semi manufacturing utilization rates to historical lows in 1Q09. And, although the utilization rates could bottom in 1Q09 since they are at historical lows, it will take more than four quarters for any meaningful improvement to materialize, thus pushing out the rebound in equipment bookings to sometime in CY10 vs. 2H09. Other reasons behind a more muted inventory refresh in 2Q are lack of killer applications, fear of continued excess inventories, and short manufacturing lead times. The downside risk to stocks is considered limited to 20% to 30%; but we do not see any meaningful earnings power in CY10 that would otherwise help sustain the current rally. To that end, we expect semiconductor manufacturing stocks to give back the recent gains and remain in a trading range. We are downgrading AMAT, ASML, FORM, KLAC, TER, and WFR from Outperform to Market Perform and reducing TER's price target from $8 to $6; we reduce WFR's price target from $25 to $21.

Inventory is still growing. Investors seem to be betting on a snapback V shaped recovery. It's not a friendly set-up for stock returns near-term. We'll start to get a sense of inventory levels over the next couple of weeks as companies report. Risk is being priced in, though, and that's good longer-term. Though I've been advocating playing some memory stocks long (SNDK, MU), I don't much care for semicaps at these levels. Memory is still an enormous portion of the semicap revenue base despite all the capacity addition cuts of late.

Gamestop (GME) saw much better holiday sales on the back of better hardware availability (Wii was actually in inventory for a while this quarter, something that pretty much has never happened.) Nintendo shipped more product into the channel this year – almost double what they've had out there. It helped. Gamestop said margins will be somewhat lower as hardware skewed the revenue growth. The stock is a sale into the strength. Though the holidays held up, post-holidays there are no significant catalysts to drive the stock and there may be an air pocket of demand without holiday spending drivers. Prefer Nintendo (NTDOY) in the space -- they continue to take share.