SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Semi Equipment Analysis
SOXX 297.50-2.6%Nov 6 4:00 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
From: FJB5/3/2011 3:57:32 PM
2 Recommendations   of 95383
 
Top Picks in Chip-Equipment Stocks
Credit Suisse likes Teradyne, Lam Research and Novellus.

Credit Suisse

Following earnings from the major chip companies, we are updating our bottom-up capital-expenditure model. Our new calendar 2011 bottom-up capex estimate is $57.3 billion (up 15% year-over-year), up from pre-earnings estimate of $55.1 billion (up 11% year-over-year).

We are seeing a mild mid-cycle pull back driven by seasonality, slightly weaker end-demand and Japan quake-related issues. We think this pullback is healthy to realign supply and demand, but we cyclical trends are intact and valuations remain supportive. We, therefore, expect stocks to work ahead of a seasonal rebound in third quarter. Post-earnings semiconductor-capital-equipment (SCE) stocks are trading at 9.7 times calendar-2011 earnings per share (and 7.9 times excluding-cash basis).

We think Teradyne (ticker: TER), Lam Research (LRCX) and Novellus Systems (NVLS) screen most attractive given recent sharper pull back; though ASML Holding (ASML), KLA-Tencor (KLAC) should be core positions as well.

Teradyne should benefit as subcon/foundry utilizations improve in the third quarter, and Lam Research as memory spending should improve in the second half.

SCE companies saw a 5% and 12% negative-swing factor to first- and second-quarter orders/shipments respectively, compared to pre-earnings expectations. We think most of the weakness occurred in late March and April. The pushouts are due to customers pausing to reassess demand and supply trends in due to weaker-than-expected end-demand trends in wireless handsets, as well as post-quake effects. In addition there may be some uncertainty on foundry market-share shifts ( Apple (AAPL) potentially may move some volumes from Samsung to Taiwan Semiconductor Manufacturing (TSM) in 2012).

The main pushouts are from Samsung and Taiwan Semi, but smaller pushouts were seen from China foundries and Elpida [of Tokyo] (DRAM). There could be some timing issues with Toshiba Y5 orders for some companies, but NAND end-demand is strong, and SanDisk (SNDK) reaffirmed capex was on track. On the other hand, we think Intel (INTC) (PC), Globalfoundries (PCs to Advanced Micro Devices (AMD) biggest exposure), other major Korean company (memory) and Micron Technology (MU) (memory) order and shipment trends remain fairly steady.

Credit Suisse foundry analyst Randy Abrams had noted "uneven original equipment manufacturs (OEM) sell-through" ahead of his pre-earnings notes on Taiwan Semi. Some OEM datapoints from earnings: Apple iPhone 4 shipments were very strong, but if that 1.7 million was channel-fill in the first quarter); Nokia (product wind-down caused some Texas Instruments (TXN) weakness, which then affected Amkor Technology (AMKR) by $20 million in June), Research in Motion (RIMM) (negative preorders for May quarter due to weaker smartphone units), Samsung (first-quarter smartphone units 12.7 million units about one million units shy), LG (feature phone weakness).

A sampling of datapoints – related wafer starts downticked (e.g., Texas Instruments, Broadcom (BRCM) and Qualcomm (QCOM) all were below precall Credit Suisse expectations for mobile divisions.

The issues directly contributed to part of Amkor's miss in the second quarter, and we suspect also resulted in Teradyne not being able to guide above consensus in the second quarter.

Although SCE companies sounded cautious, and although there was mixed wireless end-demand datapoints, for the full year, all chip companies reiterated their capex guidance, and one notable increase came from Intel. We think at least in Intel's case it was evidence of better-than-expected emerging market and enterprise demand -- and also increasing capital intensity. We could get some insights on the capital intensity theme -- Intel has talked about unveiling a "revolutionary" technology at its analyst day in May. Our suspicion at this point is it may have something to do with 3-D transistor technology, which we think is more capital intensive to make than 2-D transistors used through the 32-nanometer (nm) nodes; just like double-patterning immersion made 32-nm mode capital intensive than 45-nm. Our new bottom-up capex estimate has moved higher-post earnings, and our analysis suggests that only 26% of the full-year capex has been spent in the first quarter.

Structural demand drivers for handsets remain intact and our hardware analysts model smartphones will be more than one billion units by 2015, versus less than 300 million units in 2010. We think this mild midcycle-order correction is healthy to realign supply/demand for fabrication plants (fabs), and inventory for chips, and we do not think there will be a sharp fall off in the second half for semicap shipments.

We think visibility on wafer starts in third quarter at the foundries will be a key catalyst, which will drive confidence in order/shipment trends in second-half 2011.

Off our post-earnings estimates, SCE stocks (ASML, Lam Research, KLA-Tencor, Novellus, Teradyne, Varian Semiconductor Equipment Associates (VSEA) only; excluding Applied Materials (AMAT), which has not reported; and Cymer (CYMI) as it is an outlier on multiples) are trading at 9.7 times EPS and 7.9 times excluding cash calendar 2011 EPS. We continue to believe structurally capital intensity will increase (no new wafer-size transitions, faster rate of mask-level increases, faster end-demand unit growth next five years versus last 10 years, die size trends due to multi-core in mobile) and group is not valued for secular changes.

-- Satya Kumar
-- Brandon Heiken
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext