GS US SEMI EQUIP WEEKLY: CAPITAL
EFFICIENCY, SHORT INTEREST AND ENTG
AS A HEDGE
Summary: (1) We examine the impact of the foundry model and 300mm on the semi
equipment cycle. Along with the lack of new entrants in the semi industry, we conclude
that foundries and partnerships driven by 300mm are leading to more efficient capital
spending in this cycle. We believe that investors who argue that 2005 must be a good year
in semi capex because 2004 capex as a percent of revenue is at low levels should
reexamine this view in light of our analysis. (2) Short interest levels in most of the semi
equipment stocks are still meaningfully below peak cycle short levels. We therefore
believe that shorts could continue to pressure the stocks unless/until fundamental
reacceleration becomes apparent in H2'04. (3) We believe that ENTG (along with FORM
and ATMI) is an interesting long idea for small cap investors who must maintain some
exposure to the group or as a hedge for investors with a more cautious stock stance and
(4) News, events and price performance.
EXAMINING THE IMPACT OF THE FOUNDRY MODEL AND 300MM ON THE
SEMI EQUIPMENT CYCLE. As we first highlighted in our weekly several weeks ago,
many clients continue to argue that the capex as a percentage of sales ratio is lower in
2004 than it was in the 2000 peak and therefore 2005 must be a good year for
industry-wide capital spending. We estimate that capex as a percentage of sales reached
approximately 26%-27% at the peak of 2000 cycle while we estimate that the 2004 capex
as a percentage of sales ratio is about 24%. While it is evident that the ratio is lower in
2004 than it was in 2000, we do not believe that this implies that 2005 must be a strong
year in capex. When we first addressed this issue, we noted that fewer new entrants into
the semiconductor market this cycle was a driver of the lower capex as a percentage of
sales ratio as new entrants into the market tend to be the most irrational spenders of
capital. Below we address what we believe to be the two other primary drivers of lower
capex as a percentage of sales, the foundry model and the 300mm technology transition.
We believe that the increased prevalence of the foundry model this cycle is driving more
efficient capital spending which in turn is leading to a lower capex as a percentage of
sales ratio this cycle (see the table below). During the 2000 cycle, we estimate that
Motorola, Texas Instruments, Philips, and STMicro spent about $10 billion in capex in
aggregate. These four chipmakers are outsourcing at least some of their manufacturing
needs (Motorola more so than the others) this cycle and as a consequence we are
estimating that they are spending in aggregate about $4.5 billion in capex in 2004, about
56% less than they spent in the 2000 cycle. While one would expect that the foundries
would be making up some of this difference by spending more on capital this cycle, we
are estimating that TSMC will spend about $2.5 billion in capex in 2004 (note that we are
modeling capex above the company's $2.0 billion guided budget) vs. $3.2 billion in 2000
and UMC will spend $2.2 billion in 2004 vs. $2.8 billion in 2000. We believe that these
numbers demonstrate that the economies of scale recognized by the foundries allow them
to spend capital more efficiently than any of their individual customers would alone.
Capex, US$ millions
Company 2000 2004E
Motorola 2,400 400
Texas Instruments 2,800 1,300
Philips Electronics 1,700 600
Total 10,200 4,500
Difference (%) -56%
Source: Company data, Goldman Sachs Research estimates.
In addition to the foundry model driving more efficient capital spending, we believe that the
300mm technology transition is driving a consolidation in the customer base that is also leading to
lower capex as a percentage of sales this cycle. With the cost of building a 300mm fab approaching
$3.0 billion, few semiconductor companies can afford the expense of providing all of their own
300mm manufacturing capacity. Moreover, few semiconductor companies can afford to remain
only at 200mm, because they are very likely to be driven out of business by other semiconductor
companies that will be able to offer lower cost products because they have achieved lower
manufacturing expenses by transitioning to 300mm. This phenomenon is leading to a consolidation
of the customer base as many chipmakers have either partnered or formed joint-ventures in order to
fund 300mm fabs. One need only review a partial list of partnerships that have been formed,
including a) Sony, Toshiba, and IBM, b) Nanya and Infineon, c) STMicroelectronics, Motorola,
and Philips, d) Hitachi and Mitsubishi, e) Toshiba and Sandisk, and f) NEC and Hitachi to
understand that many semiconductor companies have turned to sharing facilities to resolve the
issue of the expense of 300mm.
In 1991, Intel, the foundries and DRAM only accounted for 29% of industry- wide capital spending
while all other IDMs accounted for the remainder. In 2003, that same group of customers
accounted for about 52% of industry-wide capital spending while all other IDMs accounted for
about 48%. Note that the standard deviation between the four primary groups of customers (Intel,
the foundries, DRAM, and all other IDMs) has tightened substantially over time. This trend clearly
demonstrates that the customer base has consolidated significantly. Another manner in which to
note the significant consolidation in the customer base is to examine the percentage of capex that
is coming from customers that are spending over $1 billion this cycle versus in previous cycles. In
1996, spenders of over $1 billion accounted for 60% of total capex and in 2004 spenders of over
$1.0 billion will represent about 75% of total capex according to our estimates.
This increased customer concentration is a negative for the Semi Equipment companies for a few
reasons. First, when a larger number of customers are competing to add capacity in order to gain
market share at the top of a cycle there is excess capital spending. We believe that it is precisely
this lack of excess capital spending at the peak of this cycle that is leading to a lower capex as a
percentage of sales ratio. Second, increased customer concentration affords the chipmakers greater
pricing leverage with the equipment vendors. While we do not believe that this customer
concentration allows the chipmakers to enforce price cuts from the semi equipment suppliers, we
do believe that it limits the magnitude of potential price increases the semi equipment companies
are able to enact. We believe that this phenomenon is apparent this cycle, as there have yet to be
significant price increases enacted across the semi equipment industry whereas we believe that
significant price increases would have been enacted even earlier in the fundamental cycle in
previous cycles
While these trends allow one to make what we believe to be a reasonable argument that the next
downturn may not be as severe as the last, we believe that those investors who own the semi
equipment stocks based on the expectation of a re-acceleration in fundamental momentum should
at least take note. First, lower capex as a percentage of sales in 2004 than in previous cycles does
not imply that 2005 capex has to increase. Second, increased customer concentration this cycle
makes it less likely that capital spending will reach as frenzied of a pitch as it did in previous
cycles. Finally, for those investors who would argue that this lack of frenzied spending is leading to
a shortage in capacity that will drive higher spending in 2005, we would point toward the all
important spread between semi revenue growth and semi capex growth that we touched upon in
our weekly out last week, which indicates that the industry is currently adding excess capacity
which will become apparent in 2005. Given all of these factors, we continue to recommend that
investors avoid overweight positions in the semi equipment stocks unless/until we get greater
visibility into whether orders to the semi equipment suppliers will re- accelerate in H2'04.
SHORT INTEREST IN APRIL STILL AT LOW LEVELS. The short interest data for April
highlights that short interest for most of the semi equipment stocks is still at low levels relative to
the cycle peak in short interest. Recall that the monthly short interest data is reported as of the 15th
of the month so we do not have data that reflects what we believe is an increase in short interest as
earnings season progressed. That said, as the data below highlights, short interest as of April 15th
was so significantly below levels previously seen this cycle that is it likely that short selling could
still put significant pressure on stocks unless fundamentals reaccelerate in H2'04. Short interest for
all of the stocks in our coverage universe increased about 7% month-over-month in April but for
the majority of stocks short interest is at low absolute levels compared to historical levels. For
example, short interest in AMAT is at about 32.3 million shares in April vs. 37.3 million shares 6
months ago and 48 million shares 12 months ago. Short interest in KLAC is at about 10.0 million
shares in April vs. 13.7 million shares 6 months ago and 25.8 million shares 12 months ago. The
same trend holds true for LRCX with 5.8 million shares short in April vs. 9.0 million shares short 6
months ago and 7.7 million shares short 12 months ago, TER with 8.9 million shares short in April
vs. 9.7 million shares short 6 months ago and 14.0 million shares short 12 months ago, and NVLS
with 8.9 million shares short in April vs. 11.0 million shares short 6 months ago and 14.0 million
shares short 12 months ago.
ENTEGRIS IS AN INTERESTING LONG IDEA FOR SMALL CAP INVESTORS WHO MUST
MAINTAIN SOME EXPOSURE TO THE GROUP OR AS A HEDGE FOR INVESTORS WITH
A MORE CAUTIOUS STOCK STANCE. For small cap investors we continue to prefer more
wafer-start or technology driven names like ATMI, ENTG, and FORM, which we would expect to
hold up better as cyclical capital spending momentum continues to slow due to their more
defensive revenue drivers. Recall that ATMI and ENTG are both primarily driven by wafer-starts
(which tend to be significantly less volatile than fluctuations in capital spending) while FORM is
primarily driven by technology transitions in the DRAM industry. While ATMI and FORM have
been relatively good performers YTD with ATMI down 5% and FORM down 11% YTD vs. the
group mean of -23% YTD, ENTG has not performed as well as the stock is down 21% YTD. In
addition to expecting ENTG to close the valuation gap with its peers, our checks indicate that
business is tracking in-line with expectations for the May quarter and we note that short interest in
the stock is near an all time high (with about 3 million shares short in April). We therefore believe
that ENTG represents an interesting long idea for investors who must have at least some exposure
to the group and/or the stock may be a good hedge on the long side for investors with a more
cautious stock stance.
News, Events and Price Performance
Last week
Monday 26 April (1) Veeco Instruments received an order in excess of $10 million for multiple
MOCVD (metal organic chemical vapor deposition) production systems from Fujian Quanzhou
Sanan Group, a manufacturer of high brightness light-emitting diodes based in Fujian Province,
China. (2) Veeco Instruments announced that it received over $5 mil
lion in orders from thin film magnetic head manufacturers for its next generation NEXUS Ion
Beam Etch System. (3) Veeco Instruments (VECO-$23; NC) reported $0.11; Street $0.09. (4)
MKS Instruments announced that its Board of Directors elected Leo Berlinghieri to the position of
President and Chief Operating Officer. Berlinghieri previously served as Vice President and Chief
Operating Officer. (5) Rudolph Technologies (RTEC-$16; NC) reported $0.06; Street $0.05.
Tuesday 27 April (1) Axcelis Technologies (ACLS-$10; IL/N) reported $0.13; GS $0.13; Street
$0.10. Please see our 4/27 note for additional details. (2) Semitool (SMTL-$11; NC) reported
$0.14; Street $0.02. (3) Semitool named Larry Murphy executive vice president and chief operating
officer. The company also announced the appointment of Steve Stahlberg to the board of directors
and audit committee. (4) Teradyne received Microchip Technology's 2003 Supplier of the Year
Award. Teradyne supplies Microchip's operations in Arizona and Thailand with J750 and FLEX
test systems for microcontroller and RF device test. (5) Mykrolis (MYK-$14; NC) reported $0.16;
Street $0.13.
Wednesday 28 April (1) Robotic Vision Systems received orders for a total of four of its WS-series
bumped wafer inspection systems from two manufacturers of bumped-wafer products. The
WS-3000 systems will be installed between June and August at facilities in Asia and North
America. The value of the orders is approximately $3 million. (2) NPTest (NPTT-$15; NC)
reported $0.03; Street $0.01. (3) Varian Semiconductor Equipment Associates won an order for its
single wafer VIISta 80HP high current ion implanter from Chartered Semiconductor
Manufacturing. The VIISta 80HP will be installed at Chartered's 300mm manufacturing facility
(fab 7) in Singapore. (4) FEI Company (FEIC-$20; NC) reported $0.11; Street $0.09. (5) MEMC
Electronic Materials (WFR-$8; NC) reported $0.16; Street $0.14.
Thursday 29 April (1) Helix Technology Corporation (HELX-$17; NC) reported $0.18; Street
$0.16. (2) Entegris announced the Entegris Automated Front Opening Shipping Box (FOSB)
Partner Program. Through this partnership, providers of equipment that interact with automated
FIMS FOSBs will work with Entegris to ensure optimal compatibility and reliability between
300mm loadport, wafer sorter and stocker equipment and Entegris' FIMS FOSB. (3) Varian
Semiconductor Equipment Associates (VSEA-$33; NC) reported $0.38; Street $0.37. (4) Mykrolis
Corporation announced that Employment Inducement Stock Options covering an aggregate of
4,600 shares had been granted on April 28, 2004 to 5 newly hired employees of Mykrolis.
Friday 30 April: (1) KLA-Tencor announced that Gary Dickerson has resigned his position as
president and chief operating officer to take a personal leave of absence. Effective immediately,
Ken Schroeder, chief executive officer, will assume Mr. Dickerson's responsibilities as president.
This week's calendar:
Wednesday 5 May: (1) Asyst Technologies (ASYT-$7; NC) reporting earnings. Street -$0.18.
GS Universe Price Performance 4/30/04 Price performance
Ticker Company Name Rtg Close Week MTD QTD YTD Y-Y
Semiconductor Capital Equipment
AEIS Advanced Energy IL/N 13 -15% -35% -35% -49% 37%
AMAT Applied Materials IL/N 18 -9% -14% -14% -19% 25%
ATMI ATMI Inc. IL/N 22 -10% -16% -16% -5% 5%
ACLS Axcelis Technologies IL/N 11 -2% -5% -5% 2% 85%
BRKS Brooks Automation IL/N 17 -14% -21% -21% -30% 97%
CMOS Credence Systems U/N 11 -14% -7% -7% -16% 58%
ENTG Entegris IL/N 10 -8% -20% -20% -21% -11%
FORM FormFactor OP/N 18 -14% -16% -16% -11% N.A.
KLAC KLA-Tencor OP/N 42 -10% -17% -17% -29% 2%
LRCX Lam Research IL/N 22 -12% -12% -12% -32% 50%
MKSI MKS Instruments IL/N 19 -13% -20% -20% -34% 38%
NVLS Novellus Systems IL/N 29 -6% -9% -9% -31% 3%
TER Teradyne Inc. U/N 20 -13% -14% -14% -20% 76%
Mean -- -- -11% -16% -16% -23% 39%
Median -- -- -12% -16% -16% -21% 38%
Source: Factset.
I, Jim Covello, hereby certify.. |
