GuruVision: The Semi-nal Debate, Reprise
SAN FRANCISCO -- Today provided little guidance on the overriding question of whether
last week marked just a "bear market rally" or something more meaningful. Tomorrow could be a far different story following the warning after the bell by
. But "today, I could take a nap it's so slow," said one market participant.
Volume was punk as major averages posted relatively modest moves: The
Dow Jones Industrial Average
rose 0.3%; the
shed 0.3%, while the
slid 2.7% after rising 14% last week.
-riding debate regarding the semiconductor industry was far from sleepy. Recall, the
Philadelphia Stock Exchange Semiconductor Index
soared 22.6% last week amid the overall rally and, more specifically, following
positive comments from
Salomon Smith Barney's
Today, the SOX fell 3.6% after
Morgan Stanley Dean Witter's
Mark Edelstone and
Dan Niles separately issued negative comments on the sector.
"The bottom line is that the macro environment continues to be difficult, with all PC and related semiconductor companies adjusting business models to adapt," Niles wrote. "As a result, we think it is too early to call the bottom regarding the fundamentals of
Advanced Micro Devices
or the PC industry as a whole."
Intel fell 6.5% while AMD shed 5.4%. Lehman Brothers has not done underwriting for either.
Edlestone, meanwhile, cut his earnings outlook on Intel and downgraded
to outperform from strong buy. Xilinx shed 2.2%, Lattice lost 3.6%, and Broadcom fell 12.1%. (Morgan Stanley Dean Witter has done underwriting for Broadcom and Lattice.)
"Most of the near-term bad news
was discounted" at the SOX's 52-week low of 455 on April 9, Edelstone commented. "However, given current valuation parameters and our belief that fundamentals will not improve much until the fourth quarter, a sustainable rally is unlikely to occur until" the end of the third-quarter.
Edelstone and Niles joined other sell-side analysts who
disparaged Joseph's call last week, as did
in Boston, as
Chip-sector skeptics also might recall that in
early March, Bill Fleckenstein of
in Seattle declared: "You can't get a bottom when everyday people try to buy the SOX."
Not surprisingly, nothing that's transpired since has changed the hedge fund manager's outlook.
"At the end of the day, these
semiconductor companies are wildly overpriced," said Fleckenstein, who remains short "most of the big chip
and equipment companies," including Intel,
. "It's a joke to think this kind of valuation is the lowest we're going to see given" the industry's fundamentals.
To say -- as Joseph did last week -- that because a company such as
had no orders in March and so that even one order will be an improvement is "sophomoric analysis," Fleckenstein said.
Joseph's comment about Cypress brought derision from a number of sources, who noted the Salomon analyst conceded the group's fundamentals have not improved.
Lehman's Niles observed that Cypress was able to generate revenues in the first quarter by scaling down its inventories. If the company does start getting orders this quarter they "won't have as much backlog to work off," he said.
Additionally, "we don't know how bad
pricing is going to be," he continued, noting only DRAM pricing has suffered to date, while flash memory and SRAM chip prices are "just starting to get hit."
Citing pricing, valuation and the macroeconomic picture both here and aboard, there are "too many variables for anyone to claim they have any idea how earnings play out," Niles said. "Just because it's really bad
now doesn't mean it can't get worse. Unless you're willing to say valuation doesn't matter,
you're buying on faith
because there's nothing to hang your hat on."
Niles and Fleckenstein both noted the developments in the
bond market, which continued its recent slide today; the 10-year Treasury bond fell 25/32, its yield rising to 5.281%. Neither said the action has reached a critical stage, but conceded rising bond yields make equity valuations relatively less attractive. There's also the question of whether the bond market is starting to anticipate a weakening dollar, and what impact that would have on the U.S. economy.
Joseph Vs. the World
If this dynamic of Joseph saying one thing about semiconductors and seemingly everyone else saying the opposite seems familiar, it may be because that's largely what occurred almost one year ago.
In July 2000, Joseph issued a
controversial call in which he essentially predicted the peak in the semiconductor cycle. As with last week's optimistic call, that forecast was treated with derision, including from this columnist. As the chip sector continued to deteriorate through the latter half of 2000 Joseph was widely credited for forecasting the downturn (and I conceded being
Past performance is no guarantee of future prescience, but fans of history (and chip stocks) should take heart that Joseph has stood alone previously and come out ahead.
In hindsight, Niles and Edelstone were (among many others) too optimistic about the chip cycle last summer.
Niles conceded as much but noted he downgraded AMD last May and Texas Instruments last June. Additionally, he forecast in December that a "headfake" rally in the SOX was likely to occur in January, which certainly proved the case. Early this year, he recommended
, suggesting the PC maker was more attractive on a valuation basis than communication chipmakers
Applied Micro Circuits
So far in 2001, Compaq is up 19.5% while AMCC is down 73.4% and PMC is off 65.8%.
Today, Niles disagreed with Joseph's contention that the chips stocks can sustain a rally before the fundamentals improve, arguing the stocks have historically followed the fundamentals. "Unless you're willing to make the case it's different this time,
the stocks are going to go down. I'm not willing to say 'things are different this time.' "
Other critics say Joseph's call last July was mainly about the wireless chipmakers and that he extrapolated their woes for the entire industry. "He got lucky on that one," said one market player, who noted Joseph also gave a wide six-to-nine month time frame last summer for when the chip cycle would peak.
Similarly, Joseph wrote last week that the SOX could trade as high as 700 but as low as 400 in the coming months, noted the source, who requested anonymity. "It's a psychology call with no facts."
Neither Joseph nor Edelstone returned phone calls seeking comment.
Aaron L. Task writes daily for TheStreet.com. In keeping with TSC's editorial policy, he doesn't own or short individual stocks, although he owns stock in TheStreet.com. He also doesn't invest in hedge funds or other private investment partnerships. He invites you to send your feedback to
Aaron L. Task.