Where's the semiconductor equipment market headed in 2010?

From Dr. Robert N. Castellano, The Information Network, USA

NEW TRIPOLI, USA: There have been very bullish forecasts about the semiconductor equipment market for 2010 in recent weeks. SEMI (Semiconductor Equipment and Materials International) issued a press release on December 1 forecasting a 53 percent increase in 2010 following a 46 percent decrease in 2009.

Gartner issued a release on December 11 forecasting a 56.6 percent increase in 2010 following a 48.1 percent decrease in 2009.

Both forecasts are based on technology purchases for 2010. SEMI in a different press release suggested that there will be no new fabs built in 2010 and pointed out that there were also no new fabs built in 2009.

These comments beg the question -- how will the equipment market grow 50+ percent in 2010 with no new fabs and no new capacity additions when it dropped nearly 50 percent in 2009 with no new fabs and no new capacity additions? Which leads to a follow-up question - will technology purchases alone sustain 50 percent growth?

Reaching 50+ percent growth will result in nearly $20 billion in wafer fab equipment purchases for technology improvements, $7 billion more than in 2009 for a net zero growth in new fabs built.

We need to look to the lithography sector a bit more closely, as it historically makes up about 25 percent of equipment purchases - in 2008 the lithography sector recorded revenues of $5.4 billion according to our analysis while the total wafer processing market was $22 billion.

As I stated in my column of November 2 on TheStreet.com, our analysis suggests the semiconductor industry is coming down to a critical inflection point for lithography.

Nikon recently shipped its first model S620 tool and ASML (ASML) recently shipped its first NXT tool recently. The NXT and S620 will be running off against each other at multiple key customer sites around the world over the next six to eight months. After that time, the market share for the lithography market will be set for the next several years as semiconductor manufacturers design-in DUV immersion processes in their future chips.

Once customers make a decision they will use that immersion tool for the next three to four years before EUVL comes along. Customers don’t want to change after they optimize their immersion process.

By mid 2010, the market will change – either ASML will be set as the dominant immersion company probably through the end of the immersion era (since no new immersion tools are being developed after these) or Nikon will take a big jump forward and gain position with the new S620.

Nikon’s S620 has a throughput increase of >70 percent vs. its older S610. The S620 immersion tool is already shipping to companies like Intel (INTC). ASML recently reported DRAM manufacturers, in particular Samsung and Hynix, are placing new orders with ASML for 193nm ArF immersion tools.

The S620 is Nikon’s most competitive product in the last 10 years. It is capable of achieving 200 WPH, 2 nm overlay and a resolution equivalent to the NXT. ASML has commanded a price premium over the years because they had a throughput advantage. That advantage will be gone when the NXT and S620 run off. Customers will make decisions by mid-2010 based on these run offs.

ASML already has a large market share so best case they keep their dominant position. On the other side, if Nikon wins 1-2 major accounts with the new S620, it will be a major win for Nikon and shift in market share. In 2008, our analysis shows that there were a total of 69 immersion lithography tools sold. ASML sold 54 and Nikon sold 15.

But, will the lithography market grow 50+ percent in 2010 to maintain its historic 25 percent of the wafer processing equipment market?

I stated in an article in Seeking Alpha on September 15 that “We nevertheless are cautious about 2010, because all we see going forward are technology purchases, rather than capacity purchases. Utilization is increasing, but with 31 fab closures in 2009, this is analogous to increased productivity due to layoffs. The lithography sector will drag the market as the 200 $5 million i-line tools typically sold in a year as capacity purchases will not be sold. Excess equipment inventory from the 31 fab closures will also weigh down the market.”

That means that the lithography market must be buoyed by immersion tools as described above selling for about $45 million each. In 2009, we estimate that 59 immersion steppers were sold, primarily by ASML. In 2010, more than 90 will need to be sold to generate 50+ percent growth. Interestingly, Cymer (CYMI), a major supplier of lasers for immersion lithography tools was downgraded on December 23 by GC Research from Overweight to Neutral. Also, on October 15, ASML was downgraded by Citigroup from Hold to Sell.

One would think that with a projected growth of 50+ percent, we would have seen the reverse. Our analysis points to a growth in the semiconductor equipment market of less than 20 percent in 2010, but with 50+ percent growth forecast in 2011. Time will tell.