Tokyo Electron (8035): A good bet

Positive impression. Q3 orders of Y95bn for SPE/FPD were up 27% QoQ, above the firm’s original guidance of “slightly up”, owing to strong Taiwan foundry and US (MPU and Korean maker with US factory) orders. Memory capex is sluggish from Samsung, Toshiba, etc., but could pick-up for NAND flash in the 2H of 2013. TEL projects -10% YoY capex for CY2013 as both memory and logic-related capex remains sluggish. This appears slightly conservative with more upside potential than downside risk. Other upside catalysts for FY3/14 are NEXX systems, a recent high growth acquisition into back-end equipment (packaging), and the improvement in price competitiveness as the yen weakens. TEL is arguably the overall winner versus foreign and domestic rivals given its recent FOREX advantage against AMAT and Lam Research, large size and deep pockets against DNS, Hitachi, etc., which will be tested on the 450 mm wafer platform.

SPE orders: About 2/3 of Q3 orders were from Taiwan and USA, mainly from TSMC, Intel, and Samsung Texas plant in the USA (memory and logic). Korea and Japan are very sluggish owing to weak memory capex. Samsung is building a China NAND flash plant for 2014 production. The China project will cost about $7 billion, and Samsung will invest $2.3bn initially in the Xi’an plant, Suwon. The Chinese plant, where 10-nano class NAND flash memory chips are planned to be manufactured, would be able to produce 100,000 wafers of 12-inch size per month. 

450 mm outlook/impact on TEL:

TEL believes it could gain market share to Intel in various tools including thermal processing, etch, and clean tools. This excludes the coater/developer, which they already have 100% market share. Intel is driving the 450 mm research along with Samsung and TSMC. Prices could rise 30-50% for 450 mm tools, as they did when wafer size shifted to 300 mm from 200 mm. In 1997-2003, during the last shift to 300 mm, TEL’s profits improved significantly, although there were various positive external factors absent today. However, profitability was actually better in 2005-2008 versus 1998-2001 despite the lack of wafer size change in 2005-2008. We believe it will be at an advantage given its large size versus rivals DNS, Hitachi Kokusai, Tokyo Seimitsu, ASMI (atomic layer deposition), etc. AMAT and Lam are also large though. All tools on a 450 mm line must be redesigned for 450 mm wafer size.

TEL’s OP and OP margin history

1993	1994	1995	1996	1997	1998	1999	2000	2001	2002	2003	2004	2005	2006	2007	2008	2009	2010	2011	2012
2.8	9.8	21.6	60.2	54.1	56.5	6.4	35.8	121.1	(18.3)	1.1	22.3	64.0	75.7	144.0	168.5	14.7	(2.2)	97.9	60.4
1.8%	5.2%	8.6%	15.0%	12.5%	12.4%	2.0%	8.1%	16.7%	-4.4%	0.2%	4.2%	10.1%	11.2%	16.9%	18.6%	2.8%	-0.5%	14.6%	9.5%