You are currently viewing SemiWiki as a guest which gives you limited access to the site. To view blog comments and experience other SemiWiki features you must be a registered member. Registration is fast, simple, and absolutely free so please, join our community today!
Hi all! Thanks for letting me in the forum. I would appreciate some insight on the wafer market (shinetsu, sumco, siltronic, sunedison semi).
There's been a huge oversupply since 2007 and these companies financials have struggled for ages. They all say they are in the low 90s% capacity utilization and prices are too low to add capacity. Also my understanding is substrate content of total semiconductor selling price is only about 2-3%? There's also been some consolidation (samsung exit). And yet, year after year, they've been unable to increase prices, year after year. Why is that?
Is it because as photo lithography gets finer and finer, you create so much additional capacity that you don't need a lot of sq inch (sort of like what dwdm did to fiber transport prices)? Or because pc/dram demand keeps shrinking and there's no major trend that's going to drive semiconductor demand in the future (like smartphones did)?
Is it because most of the capacity is in Japan and the yen has depreciated from 2010 (76 to 120 back to 100)?
Also if the largest player builds the first 450mm fab, would that kill everyone else because they are adding so much capacity in sq in that the others would not be able to compete.
Anyway, I am a newbie and am trying to get a handle on how to think about the industry.
You may be over-thinking this one. If there's a huge oversupply of silicon wafers then by definition the suppliers cannot raise prices, it's a basic economic tenet of "supply and demand".
If the wafer suppliers wanted to raise prices then they could always decrease their supply and create a shortage of wafers, which would cause prices to increase.
Intel built a 450mm fab in 2013, but in general Intel is having difficult filling their fab with products that consumers want to buy. The world has plenty of 300mm fabs in production, and so the transition to 450mm is quite slow because there isn't an economic incentive to do so at this time.
Appreciate the input. Definitely oversupply. Just wondering when it will start to turn. There was a lot of optimism in 2Q/3Q 2015 and then they got hit (mainly by the smartphone & pc slowdown?) So now some people are optimistic (SEMI) - but seems they are optimistic people but they are 2.5x up from the lows. Siltronics is more sober.
Am just wondering what would signal a change in the demand situation and when or if it's ever going to happen. I've seen a lot of flat/1% growth projections. But TSM seems more optimistic.
If the wafer suppliers wanted to raise prices then they could always decrease their supply and create a shortage of wafers, which would cause prices to increase.
More likely scenario: suppliers go in price war to protect market share; war ends if demand rises or one or more of the war lords is defeated, e.g. goes bankrupt or is bought.
SunEdison Semiconductor investors will get $12 a share cash, 45 percent more than Wednesday’s closing price, according to a statement from the companies. The deal has the backing of both boards and will be financed through existing cash and bank financing
So...I decided to buy Siltronic last August (waf.de). All the wafer guys are up. Waf is up 90% (15 to 29). Sumco is up about 40% (700ish to 1000). In yesterday's con call, Siltronic said that wafer demand is 2016, they now expect 3.8% vs 1.3 % previously. Demand strengthened a lot in the past few weeks. They had to draw down on inventory of 200mm and 300mm and had to turn away orders. They also said that they are in discussions with customers regarding increasing prices in 2017
Would you guys know if it's realistic to increase prices for wafers by 5-20%? Shinetsu/Sumco/Waf all said last year that they would not be able to increase capacity at current prices and implied that a 20% increase (from last year's higher prices) would be necessary.
Waf trades at about half of greenfield replacement cost and Sumco at replacement cost.
I think what they are saying is a brownfield investment would require a 40% price increase to breakeven (5 yr equipment depreciation + 31 years for building) and a greenfied expansion would require 50% to breakeven.
In the meantime there's a lot of fab capacity from 2016-2018 (also in presentation) so I am betting we are just at the start of the price upswing
but they are 2.5x up from the lows. Siltronics is more sober.