08:46am EST 7-Jan-98 Lehman Brothers (Michael A. Gumport, CFA 1(212)526-) ALSC
Semiconductors: 1998 Semiconductor Outlook (Part 1 of 3)
Ticker :
Today's Date : 01/07/98
------------------------------------------------------------------------------
* 1998 SEMICONDUCTOR OUTLOOK - PART I OF III.
----------------------------------------------------------------------------
PART I OF III
1998: EXPECT THE BROAD THEME FOR CHIP STOCKS TO SHIFT TOWARDS RENEWED HIGH,
LONG TERM GROWTH AND AWAY FROM TODAY'S FOCUS ON "DECELERATION AND ASIA/PACIFIC".
Despite current chip industry turmoil, we remain impressed by the new uses/new
users for chips. The industry has averaged roughly 17% growth in every 10 year
period since the early 1980's. We see no sign of saturation. To the contrary,
an increasingly rich product mix (more high growth digital CMOS ICs, less low
growth bipolar and discrete) suggests growth over the next 5 years could
accelerate to 19%. That ongoing strong growth pattern should be visible by late
1998. We expect about 15% growth in 1998 and to end the year with particularly
strong comparisons. US and international chip sales patterns have never
radically diverged over any significant period of time, and the strong US
economy should end up as the dominant factor for 1998 (about half of all chip
sales go directly to the US market). OVERALL, EXPECT STRONG PERFORMANCE BY YEAR
END.
HERE ARE THE UPS AND DOWNS WE EXPECT TO DELIVER 30% CHIP STOCK APPRECIATION IN
1998: 1) In early January, tax selling/window dressing ends; group goes up 0-
20% (now complete), 2) Mid-January 4Q97 EPS reported at low end of consensus;
group drops back 10% (any concern on U.S. growth will exacerbate setback); 3)
February seasonal rebound in orders (plus cyclical strength from US expansion)
resumes; group rises 30-50%; 4) April-June seasonal strength peaks; group begins
25% selloff; 5) September-December strength of DVD drives, internet appliances,
digital imaging emerges, pushes industry sales up 15% for the year (particularly
strong 4Q comparisons) and stocks rebound 25%. Overall 1998 gain: 30%
(assuming modest growth in the broad averages).
GIVEN THE UNCERTAIN WORLD ECONOMY, INVESTORS SEEM LIKELY TO FAVOR COMPANIES WITH
THE HIGHEST SUSTAINABLE MARGINS (AND, IF POSSIBLE, BEST GROWTH). OUR REVISED
RECOMMENDATIONS REEMPHASIZE OUR FOCUS ON THOSE PROPRIETARY PRODUCT LEADERS.
TOP CHOICE: VITESSE (VTSS - 38 7/8; rated 1). Most significantly, VTSS is the
only name in our group where we are comfortable they will print the numbers.
VTSS continues to ride the need for bandwidth. No other company competes in its
segment - high speed digital, high integration GaAs market.
NEXT CHOICES: TOP QUALITY NOW, INCREASED MOMENTUM LATER. XILINX (XLNX -
37 3/8; rated 1) AND LATTICE (LSCC - 49 1/8; rated 1), LIKE VTSS, ENJOY GROSS
MARGINS IN THE 60% RANGE. Both these companies are leaders in the single most
attractive segment of the digital chip market, user programmable logic devices.
We expect XLNX to retake the leadership from Altera (ALTR - 36 1/4; not rated).
LSCC's older products are dying, but its new product line is the fastest growing
within the sector and will represent 80% of sales within two years.
OUR 2 BIG COMPANY FOCUS NAMES: SYSTEM LEVEL, MIXED SIGNAL (ANALOG/DIGITAL)
INTEGRATION LEADERS TEXAS INSTRUMENTS (TXN - 47 1/4; rated 1) AND SGS-THOMSON
(STM - 61 7/16; rated 1). Both these companies lack near term momentum, but
similarly, both are among the industry's most improved companies, are leaders in
the trend towards system level analog/digital integration, and both stocks have
trended sharply higher in recent years as multiples have expanded. We think the
trends are strong enough at these companies to merit ongoing commitments.
LONGER TERM APPRECIATION AND RELATIVE SAFETY -- AVNET (AVT - 64 7/16; rated 2).
Distributors have moved up from P/Es of 4-5x in the early 1980s to 15x today.
But the good long term prospects of distributors deserves still higher
multiples, and we expect Avnet to continue to weather its way gradually upwards.
Because cConsensus estimates appear high, however, we rate Avnet a "2" and see
some risk of a short term setback.
CAUTIOUS ON X86. On the one hand, the X86 code is clearly still the key
beneficiary of each PC sale (aside from software). But the increased level of
competition in this arena makes us relatively cautious. Leader Intel (INTC -
73 1/8; rated 2) has been giving up share and margins have eroded. Challengers
Advanced Micro Devices (AMD - 20; rated 3) and National/Cyrix (NSM - 25; rated
2) have gained share, but results have been extraordinarily volatile and profits
elusive. AMD and NSM each seem more likely to move up 20-40% before Intel, but,
given the risks and volatility and to underline more predictable performance
elsewhere, we rate AMD a "3" and NSM a "2" along with Intel. WE ARE CONCERNED
AT THE POTENTIAL FOR A NEAR TERM SETBACK AT AMD ON POOR UPCOMING EPS
ANNOUNCEMENTS AND THAT AMD WILL SEE ANOTHER DOWN LEG IN ITS STOCK BEFORE IT
MOVES UP AGAIN.
CAUTIOUS ON MEMORY: LOOK FOR DIVERSIFICATION AND VALUE; RATE ALLIANCE (ALSC -
5 5/8; rated 2) AND INTEGRATED DEVICE TECH. (IDTI - 10 1/4; rated 2) "2", MICRON
(MU-$26 7/8-4) "4". Prospects for late 1998 look much better, but prospects for
early 1998 just keep getting worse. The commodity memory sector remains
extremely difficult. We do expect the Korean crisis ultimately to be a big
positive for memory makers, but the benefits of the Korean problems do not look
likely to help the short term at all. The DRAM sector is by far the worst, but
SRAM's are not good. The program here is to 1) Ride out the industry
consolidation and reemerge as a low cost producer, 2) Diversify. We think IDTI
is furthest along in the diversification effort with X86's due out in volume by
March, user programmable logic devices imminent (1/19 announcement a major long
term positive), ATM switching products, and graphics accelerators. ALSC is next
with a major effort to bring to market embedded memory products (plus hidden
value in joint ventures). MU remains most DRAM dependent despite success with
smart cards and some early signs of success in flat panels. Reflecting the
level of diversification and quantifiable value, we rate ALSC and IDTI a "2" and
MU a "4".
CAUTIOUS ON LOWER MARGIN COMPANIES. We are betting more heavily on companies
with the best demonstrated levels of profitability. In that light, although
International Rectifier (IRF) is the biggest player in the high growth (25%)
power MOSFET market, gross margins have been under 40%. We believe news of
IRF's role in the mobile Pentium 2 (Deschutes) will be a positive, but, until
momentum becomes more evident and the profitability story better elaborated, we
have trouble making it among our top choices. The good news, however, is
consensus seems too conservative. Still, consistent with our focus on fewer
names with clearly enunciated value and momentum, we rate IRF "2". Similarly,
lack of momentum at LSI and margins in the mid-40's make us relatively cautious
on near term appreciation potential despite the fact that LSI is a leader in
system level integration, a key industry trend.
NEAR VENTURE AND VENTURE RATINGS. Near venture and venture stocks inherently
display above average risk. Among stocks we follow in this category, our
favorite near-venture pick is Ramtron (RMTR - 5 3/4; rated 1), which could pick
up momentum later this year from a possible JV with SGS-THOMSON and the
licensing of other new chip makers. However, there is little positive news in
the next few months to move the stock upward. Venture stock, PixTech, (PIXT -
2 3/4; rated V2) while currently priced for disaster, will likely gain momentum
later this year from a ramp up of FED (field emission display) production at
foundry Unipac. |
