Read the article,,
Part II of ,,
November 9, 1998
Tough Techs
Our experts find winning stocks in the year's wreckage
By Eric J. Savitz
November 9, 1998
Tough Techs -- Part 2
Cover Story, Part 1
Q: Kevin?
Landis: I've had my heart broken several times by AMD. So I'll stay clear of
that one. On the other hand, I have a hard time imagining a future in which
being the leading microprocessor company isn't a sweet business. But it will
be a different business. Microprocessors are going to be cheaper. Intel is
going to show the world just how well they can adjust all the knobs in their
fab and get costs down, get their die sizes down, get their yields up, so they
can make money at lower and lower price points. But if the PC market is
growing at low double digits, or even single digits, and the prices are coming
down, they'll be helpless to grow revenues.
McNamee: Intel's great core competence is taking large, complex chips and
getting them into high-volume production quickly. That runs contrary to the
most important current factor in the semiconductor industry, which is that
we're drowning the state-of-the-art capacity. This is a perfect time to be
fabless. That really undercuts Intel's core competence.
Wick: Intel, as it happens, has about one fab too many right now. That's one
heavy brick. But they're also starting to figure out that the world is not just
about PCs anymore. You have to look where they're headed. They want to
fill that extra fab with networking silicon.
Landis: So far, they've been a huge loser in networking. They have been in
the hub business, and the low-end Ethernet switch business, but they're going
nowhere fast.
Wick: But what are their alternatives? They know that's the place to go.
McNamee: Intel has annualized revenues of $27 billion. They can own
100% of every semiconductor that gets made for data-networking today, and
it would not be a meaningful portion of their income statement. The law of
large numbers suggests that for Intel to reproduce its business model around
networking stuff is --
Wick: Impossible.
McNamee: Let's put it this way. It is certainly impossible with respect to a
time frame associated with today's stock price.
Q: What about the notion of Intel buying 3Com or one of its rivals?
Landis: Intel-3Com discussions have been talked about ad nauseam. I don't
like the idea of them buying all of 3Com's current headaches.
Wick: It would be awkward. The Justice Department probably wouldn't like
the idea of the No.1 and No. 2 players in the Ethernet-adapter-card market
merging together. So scratch that part of the business. The modem business
long-term is something that is probably going to experience even further price
pressure, and might move on to being a few chips on a motherboard. And I
don't think Intel wants to be in the Palm Pilot business.
Q: What do you guys think
about the outsourcing
phenomenon? Is it a passing
fad?
McNamee: There's been a
definite shift toward greater
specialization. We've seen the
emergence of industries targeted
at tasks previously accomplished
in-house. The number of companies that can be truly successful through
vertical integration is shrinking. It's going to make more and more sense to
focus on the things you do well. We're even seeing that happen on the
Internet, with the emergence of companies like Exodus Communications,
which provides Web hosting services. If you have a big site, you can focus on
making it look great and do the right things for customers, rather than handling
communications infrastructure. If you're in the portal business, you don't need
to design your own search engines; you can outsource that to Inktomi.
Q: What about contract manufacturing?
McNamee: We've got a big position in Flextronics, which makes gear for
networking companies like Cisco and Ericsson. They also do lots of
Microsoft's contract manufacturing. In the services area, we've bought a
position in Cambridge Technology Partners, which does systems integration
and application deployment. Cambridge has had some operating issues, and
the stock got creamed.
Q: What happened to them?
McNamee: They'd been growing rapidly, lost a couple of key managers, got
a little bit out of touch with their business and paid the price. The result is,
investors have an opportunity to own this stock at a very good valuation.
Wick: We've been playing the outsourcing theme for the last two or three
years. We own Jabil, SCI Systems, Saumina and Smart Modular, all in the
contract-manufacturing area.
Q: Kevin, have you been investing in this group?
Landis: I haven't. To me, it's the kind of business where somebody like
Cisco can play one of them off against another, and grind them down to a
minimal level of profitability.
McNamee: But they don't actually choose to do that. Cisco has been a great
partner for contract manufacturers.
Q: Let's turn to the ERP business, which has been a nightmare the last
few months. PeopleSoft has been disappointing. Baan has been a
disaster. Even SAP has lost ground.
McNamee: Enterprise software is here to stay. The days of custom
applications developed from the ground up are behind us. We're really
excited about software that does one of two things: increase revenues by
directly touching customers, or optimize business processes to make you
more productive. The idea is to squeeze expenses or raise product quality or
increase customer satisfaction. Traditional ERP software is about tracking the
business. It tells you how many units you're making, what they cost, and how
much revenue you're getting. Not very exciting.
Q: Which companies make
sense to you here?
McNamee: We especially like
I2, the leader in the
supply-chain-management
category. It's a business with
quick payback. At the end of the
day if you use I2 software and
your competitor doesn't, you
win. People will spend $20 million or $30 million on an SAP installation, and
all they get is prettier reports about accounting information they already knew.
You buy I2 software, and your whole manufacturing line works better.
Q: Are there other enterprise software stocks worth buying?
McNamee: We're high on Seibel Systems and Genesys
Telecommunications. Seibel is the leader in sales-force automation. Genesys
automates call centers. They both have big market opportunities, and
exceptional management. Again, those stocks have come way down.
Q: Any others?
McNamee: We like two companies that benefit from the shift to Windows
NT. One is Platinum Software, which is the No. 1 vendor in ERP software
for NT. They're focused on middle-market customers, which historically
haven't been touched by SAP and the rest of the big players.
Q: Paul, have you been playing in enterprise software?
Wick: We missed the huge runup in the ERP stocks, and then we missed the
express elevator down. We own a couple of mainframe software and
open-systems companies, Compuware and Computer Associates. Computer
Associates, in particular, looks cheap, and it has very high margins.
Q: Gentlemen, let's get serious about stockpicking.
Landis: My top choice should come as no surprise: it's PMC-Sierra. It's
been my favorite stock for two years now, and continues to be. It's a bet on
raw traffic. PMC is a preferred supplier to the Ciscos and Lucents of the
world.
Q: How are they doing?
Landis: In the latest quarter, they earned about 29 cents a share. They are
just now phasing out some of their older products, so the top-line growth rate
will converge with their networking products growth. That's a business
growing double digits every 90 days.
Q: What else?
Landis: If you're impressed by what Dell has done with redefining the right
way to run a business, then you can do a lot worse than to quit kicking
yourself over not owning Dell and buy i2.
Q: Another i2 devotee, eh? If the company is so great, and the stock so
cheap, why doesn't Oracle or PeopleSoft or SAP just go buy them?
Landis: Sometimes, you know you're invested in the right stock because you
don't want the company to get acquired. You don't buy a stock like i2
because you can make 40% in a week when somebody announces a buyout.
You get into it because you can want it to double, and then double again, and
again.
Q: But strategically, wouldn't it make sense for SAP to buy I2?
Landis: Absolutely. It would make even more sense for Oracle. SAP seems
more inclined to create their own products and muscle their way in. Oracle,
on the other hand, might try to leapfrog in.
Another long-term trend I'm
playing is photonics. And the
way I've been playing it is
Uniphase. They make the basic
components you need to build a
fiber-optic network. The
company has gone about
methodically collecting all the
best minds in the underlying
technology of fiber optics. As the
markets move forward, they're always going to be the company that has the
better product sooner, that has the product that doesn't have the gotchas,
doesn't have the bugs. And the communications world, unlike the computing
world, is incredibly unforgiving. It is not enough to be 99% reliable. You need
five nines of reliability. You need to know your stuff. They do. And they make
money, to boot.
Q: Paul, give us some stock tips.
Wick: My first choice is the Learning Co It plays into our thinking that 1999
will be a better year for consumer computing than corporate computing. As
PC prices fall, they've become more common in Middle America. The
company benefits from being the leading education- and reference-software
company in the U.S. They've got aggressive management, which has turned
around a company that a few years ago was saddled with lots of debt from
some acquisitions made at admittedly high prices.
Q: What do they sell?
Wick: The company is an agglomeration of some of the best retail software
brands. Reader Rabbit, Sesame Street, Princeton Review. Over the summer,
they bought Broderbund incredibly cheaply -- pulling out balance-sheet cash,
they paid just 0.8 times sales. That's cheap for a company with some good
franchises of its own, like Dr. Seuss, Carmen Sandiego and the Berenstain
Bears. On the reference side, they've got National Geographic and
Compton's Interactive. They also have a substantial presence in productivity
applications Print Master, Calendar Creator and Family Tree Creator. The
company has a strong distribution network. They do direct mail. They sell
directly to schools. And they're the largest player in terms of shelf space at
retail.
Q: Sounds impressive.
Wick: And they're going to wring a lot of cost savings out of Broderbund.
They plan to fire about 80% of Broderbund's employees. Learning Co. has
operating profit margins in the neighborhood of 30%, which is among the best
in the software industry. The valuation is pretty attractive for a company
growing about 20% a year. The stock is around $25. They should earn about
$1.50 this year and $1.80-plus next year. That's 15 or 16 times earnings for a
great franchise with good cash flow.
Q: What else do you like, Paul?
Wick: We like a stock called Amkor, which has been a less-than-stellar
performer since it went public five or six months ago.
Q: That's for sure.
Wick: It came out at 11. Now it's 4 1/2 .
Q: Yuck.
Wick: Amkor is the world's largest semiconductor-packaging company.
There's been controversy over the linkage between Amkor and an affiliated
Korean company called Anam. The CEO of Amkor and his family, which
owns 60% of Amkor, also owns 40% of Anam. Amkor resells virtually all of
the semiconductor packaging services of Anam to people like Intel. Anam is
heavily leveraged, with lots of debt on their balance sheet. Interest coverage
and cash flow are fairly thin. It bit off more than it could chew. To expand
fast, Anam leveraged itself up to the hilt. There's been a lot of concern that
Anam could fail, which would be bad for Amkor, since it gets two-thirds of its
sales through Anam. Events of the last few months suggest the fear is
overstated. Anam is going through a "workout" process with its Korean bank
lenders. The company will renegotiate the maturities of its bank debt, and
likely get a lower interest rate. The important thing is, it doesn't look like
Anam is going to go under.
The stock has gone from 11 to 4 1/2 . But this is still a relatively stable
business. The company is not losing business. Their operating margins are
about 9%. Estimates for 1999, which started the year at $1.05, are now
running about 75 cents. The top-line growth looks flat on a year-overyear
basis for the near term. And their capacity utilization is running 65%-68%,
down from a more typical 75%-80%. If you assume the chip industry is going
to have even a modest recovery, with less capacity having been added to the
system, pricing pressure is likely to abate. Capacity utilization should improve.
The estimates for next year could be low. Moreover, their business is
cash-flow positive, and their balance sheet is healthy. It's a $4.50 stock with
75 cents a share in earnings. A year or two out, you should have north of $1 a
share in earnings power. It's trading at less than 0.4 times revenues, which is
incredibly cheap.
Q: Another pick?
Wick: A company we've owned forever, in a space where we don't have
much exposure otherwise, is ECI Telecom. It's an Israeli company, which
specializes in a product called digital-circuit-multiplication equipment. In some
sense, what Ciena does for fiber ECI does for copper.
Copper-bandwidth-multiplication technology is predominately used to expand
voice capacity for long distance. That's the bulk of their business, about 50%
of sales. People thought this was going to become a mature business. Instead,
it's been quite robust, growing about 25% a year, with no signs of slowing
down.
Q: What else do they do?
Wick: They've got some other telecommunications business, including a
video-over- ATM business called Hi-TV. They've got a Sonnet-like business
called SDH, a nonU.S. transmission standard that's accepted in Europe. They
also have Digiloop, repeaterless T1 technology similar to PairGain's. Their
stronghold is in Europe and to some degree the developing world -- they're
not as big a player in the U.S.
Q: Why do you like them?
Wick: What's been driving them has been the deregulation of the European
market, with new carriers offering voice services. ECI has been on a roll for a
couple of years. They've consistently met or exceeded estimates. They've got
operating profit margins north of 20%. And because they're an Israeli
company, they pay almost no taxes. It's trading at about 13 times earnings --
they should report $2.50 next year. If it were an American or European
company, it would trade at 25-30 times.
Q: Roger, give us some picks.
McNamee: I'm going to give you two lists. First, the momentum picks. Cisco
remains one of the pre-eminent business franchises. Dell remains an
extraordinary business, and one of the really exciting things going on in the
economy right now. In the business-services sector, we like a company called
International Network Services, which helps deploy Cisco networks. It's not
a cheap stock, but it's extraordinarily well positioned. Literally half of all the
Cisco certified-deployment engineers in the world work for that company.
The stock is about 41; they should earn just under a buck in 1999. Earnings
doubled this year, and next year should grow about 50%.
Q: And your value picks?
McNamee: You want to be in enterprise software. I agree with Kevin that i2
is the single most attractive business franchise in that space. That's the first
stock I'd buy in any sector. I mentioned Platinum Software, Genesys
Telecommunications, Flextronics and Cambridge Technology Partners. And
I'd echo Kevin's recommendation of PMC-Sierra.
Q: Finally, gentlemen, any stocks people should avoid -- or short?
Wick: One that strikes me as particularly absurd is Geocities, which has
about a $1 billion market value. I just don't think there's much there. In the
long run, they'll have trouble competing with people like Yahoo, Excite,
America Online and Netscape. Geocities is unprofitable, and will be for as far
as any Wall Street analyst can see. To me, this looks like a company that
could go to zero.
Q: Roger?
McNamee: I'm pretty negative on the European enterprise-software
companies. SAP is still overpriced. And I don't understand the valuation of
LHS Group. They do something important, selling billing services to the
telecommunications industry. But their customers are consolidating rapidly.
There are plenty of players. And they sell at 13 times revenues.
Q: Any others?
McNamee: I don't get Lernout & Hauspie, the speech-recognition software
company. It sells for more than 10 times revenues. It might be a fine business,
but it's valued in a way I just don't understand. I'm also cautious on people
who sell to the RBOCs, which makes me cautious on Lucent.
Q: Guys, it's been an eye-opener. Thanks.
Return to top of page | Format for printing
Copyright © 1998 Dow Jones & Company, Inc. All Rights Reserved.
|
