GS US SEMI EQIP: KEY TAKEAWAYS FROM
BOSTON ANALYST MEETINGS AND THE
BOOK-TO-BILL
Summary: We attended two days of analyst meetings with MYK, LTXX, BRKS, and
ACLS on Weds and Thurs. Below we provide our key takeaways from the meetings and
note that overall managements remain very bullish while highlighting a possible decline in
Q3 orders. SEMI also released the April US book-to-bill (b-t-b) ratio of 1.14 (GS 1.09 &
the Street 1.10) on Thurs, with orders and shipments both coming in significantly above
our and the Street's expectations. Orders were up 16% m-m (16% above our estimate) and
shipments were up 11% m-m (11% above our estimate). We are estimating a flat b-t-b
ratio in May on order and shipment growth of 5% m-m. We continue to emphasize that
the b-t-b should not be a trading event for the stocks as it is backward looking and
unaudited (b-t-b April-qtr orders were up 32% q- q, exactly in-line with AMAT's April-qtr
order growth reported on Tues). No change to our belief that it is too late in the
fundamental cycle to be overweight the semi equipment stocks.
KEY TAKEAWAYS FROM BOSTON SEMI EQUIPMENT ANALYST MEETINGS.
We attended two days of analyst meetings with Mykrolis, LTX, Brooks, and Axcelis in
Boston on Wednesday and Thursday. The meetings were extremely well attended and
management tone continues to be very bullish. We would almost characterize
management tone as defiantly bullish in response to the recent sell-off in the stocks.
Management tone not withstanding, the issue that is increasingly apparent in the semi
equipment industry is that third quarter orders look to be flat to down slightly sequentially
across the industry. While no management provided formal guidance for the third quarter,
we believe that body language and offline commentary supported this notion. Below we
provide our key takeaways.
(1) AXCELIS ANNOUNCES NEW SINGLE WAFER TOOL. During its presentation,
Axcelis announced the introduction of the Maximum platform of single wafer low energy
ion implanters. Management continued to underscore its long held notion that the
majority of applications in a fab require multi-wafer processing. That said, a significant
portion of the meeting was dedicated to highlighting that at the 65-nanometer node and
below, some additional applications will require the high-tilt functionality only available
in a single wafer platform. This is because of the need for uniform doping and to
compensate for lower diffusion rates at smaller transistor geometries. Axcelis therefore
expects to have a beta single wafer tool available in the fourth quarter of 2004. The
company intends for the tool to be available for volume production in 2005, when it
believes the mainstream adoption of 65-nanometer will begin with true volume
production of 65-nanometer beginning in 2006 and beyond.
(2) AXCELIS MANAGEMENT ALSO PROVIDED A DETAILED UPDATE ON THE
TECHNICAL ISSUES AND COMPETITIVE ENVIRONMENT IN THE DRY STRIP
SEGMENT. While the presentation centered on the technical challenges associated with
stripping true low-k materials at smaller geometries, we also gleaned some interesting
competitive insights around the dry strip market during the meeting. Most significantly,
we believe that Novellus continues to focus on the front-end of line (FEOL) dry strip
market (i.e. transistor formation) while Axcelis and Mattson continue to focus more on
the faster growing back- end of line (BEOL) dry strip market (i.e. the dielectric level), with
Axcelis seemingly well-positioned to gain market share in the strip over low-k segment. Axcelis
management believes the company will overtake Mattson as the number two provider of dry strip
tools in 2005, as Axcelis believes that Mattson's technology does not extend beyond the
90-nanometer node. Recall that in 2003, market share in the dry strip segment brokedown as
follows according to Dataquest:
2003 market share
Novellus 30%
Mattson 24%
Axcelis 18%
Source: Gartner Dataquest, Axcelis.
(3) AXCELIS SEES BUSINESS STABLE AT HIGH LEVELS. Regarding the cycle, Axcelis
management was extremely bullish, highlighting that they believe that the semi equipment market
will remain strong through the end of 2004. That said, while management did not provide formal
third quarter order guidance it did raise the possibility that third quarter orders could be down
sequentially, consistent with our view. We continue to believe that one of the disconnects between
the bullish tone notable from several management teams and the recent semi equipment stock
underperformance is that management is understandably focused on business remaining at high
levels for the next several quarters while the Street is more focused on the slowing/flattening in
sequential order growth patterns, which we believe is likely over the next several quarters. Axcelis
is the second management team in the last several weeks to highlight potentially down third
quarter orders (ASML has given similar indications).
Regarding Axcelis' stock, it has been a strong relative performer in 2004 as it is up 6% YTD versus
the group mean of down 19% YTD, as investors have focused on relative valuation based on 2004
earnings (ACLS is trading at 10x 2004 earnings versus the group mean of 20x). We maintain our
In-Line rating on the stock within our Neutral coverage view, as we see little potential upside
estimate revisions over the next several quarters.
(4) BROOKS' MANAGEMENT REITERATED ITS UPBEAT TONE AND ADDRESSED THE
ISSUE OF A GROSS MARGIN SHORTFALL IN Q2 RELATIVE TO LONG-TERM GROSS
MARGIN TARGETS. Brooks maintained its upbeat tone during its analyst meeting, highlighting
that it continues to believe that there are incremental fab opportunities to drive growth throughout
the remainder of this year. We would note that while there have been a significant number of fab
announcements and there are likely to be even more to come, we would expect many of these fabs
to be equipped during the next cycle as semi unit growth is likely to slow exiting 2004 and into
2005, thus decreasing the need for significant additional fab capacity for the current cycle.
Management also spent a significant portion of the meeting addressing the gross margin shortfall
reflected in its Q2 guidance relative to its target operating model. Recall that Brooks had
previously indicated that at $150 million in quarterly revenues, gross margins would be
approximately 42%. However, guidance for the current quarter includes a high end revenue
expectation of $150 million with high end gross margin guidance of 39%. Management explained
that product mix is driving the gross margin shortfall relative to earlier expectations. The company
indicated that it expects gross margins to move higher over the coming quarters as higher margin
software sales become a bigger part of the revenue mix given the improving software book-to-bill
last quarter.
(5) BROOKS HIGHLIGHTED ITS ONEFAB AMHS PRODUCT OFFERING. Brooks continued
to highlight that it is gaining traction with its OneFab AMHS platform. Management commented
that the Micron fab, which is the first facility to use the OneFab AMHS system, is progressing
ahead of schedule and Brooks is currently bidding on four additional AMHS opportunities. The
company further commented that it will reconsider its participation in the AMHS business in 6-9
months. Whether the company will remain in the business will depend on whether the irrational
pricing behavior of its Japanese competitors (Daifuku and Murata) stabilizes over the coming
quarters.
3-MONTH ROLLING AVERAGE BOOK-TO-BILL OF 1.14 WAS ABOVE OUR 1.09
ESTIMATE AND THE STREET 1.10 ESTIMATE. Semiconductor Equipment and Materials
International (SEMI), a semiconductor equipment industry trade association, reported its
three-month rolling average April bookings and shipments statistics on Thursday at about 6pm
eastern. The U.S.-based semi equipment suppliers' book-to-bill ratio was 1.14, 0.05 above the GS
1.09 estimate and 0.04 above the Street 1.10 estimate on significantly higher than expected orders
and shipments. Orders were $1,594 million, 16% above our estimate of $1,375 million and 15%
above the Street estimate of $1,386 million (+16% month-over-month and +111% year-over-year).
Shipments were $1,401 million, 11% above our and the Street estimate of $1,260 million (+11%
month-over- month and +67% year-over-year). The front-end book-to-bill was 1.12 on orders
+16% month-over-month and +117% year-over-year and shipments +10% month-over-month and
+59% year-over-year. The back-end book-to-bill ratio was 1.19 on orders +16% month-over-month
and +90% year-over-year and shipments +14% month-over-month and +99% year-over-year.
March's overall book-to-bill was revised downward to 1.09 from 1.10 but on 4% higher orders and
6% higher shipments (greater increase in the denominator than in the numerator drives a lower
overall ratio).
BOOK-TO-BILL BETTER THAN OUR AND THE STREET EXPECTATIONS ON HIGHER
ORDERS AND SHIPMENTS, BUT DOESN'T TELL US ANYTHING APPLIED DIDN'T
ALREADY TELL US. While the April book-to-bill was above our expectation on better than
expected orders and shipments off of an upwardly revised base for the month of March, the data do
not tell us anything that we haven't already heard from Applied Materials. The book-to-bill data
shows total bookings for the months of April, March, and February of $3,309 million, an increase
of 32% from total bookings for the months of January, December, and November of $2,514
million. Recall that on Tuesday evening Applied Materials reported earnings for the April-quarter,
noting that bookings were up 32% sequentially, exactly in-line with the improvement in
quarter-over-quarter bookings reported in the book-to-bill data.
MAY BOOK-TO-BILL EXPECTED TO REMAIN FLAT AT 1.14 ON 5% M-M ORDER
GROWTH AND 5% M-M SHIPMENT GROWTH. We are estimating a flat book-to-bill ratio in
May, on 5% m-m growth in orders and 5% m-m growth in shipments. We are modeling
three-month rolling average overall orders of $1,680 million (+5% month- over-month) and overall
shipments of $1,470 million (+5% month-over-month). We estimate front-end shipments of $1,150
million (+5% month-over-month) and front-end orders of $1,300 million (+5% month-over-month),
yielding an estimated front-end book-to-bill ratio of 1.13. We estimate back-end shipments of
$320 million (+5% month-over-month) and back-end orders of $380 million (+5%
month-over-month), yielding a back-end book-to-bill ratio of 1.19.
WE CONTINUE TO BELIEVE THAT IT IS TOO LATE IN THE FUNDAMENTAL CYCLE TO
BE OVERWEIGHT THE STOCKS. We continue to highlight that the book-to-bill should not be a
significant trading event for the stocks given that it is a backward looking and unaudited metric. As
we have noted on numerous occasions, given that the semi equipment industry passed its
normalized levels of cash flow when Applied reported EBITDA margins that were greater than
18% in February, we believe that it is too late in the fundamental cycle for investors to be
overweight the semi equipment stocks.
I, Jim Covello, hereby certify ... |
