Semiconductors: Analyst Says Don't Hock Chip Stocks, SOX Still Rocks
Miami, FL, August 9 /SHfn/ -- The dogs of yesteryear are making an exhilarating upsurge. In Chicago, the White Sox hold a
tenuous American League lead. Meanwhile, around the globe, stomachs are churning over a three-week-long slump for another type
of SOX: the Philadelphia Semiconductor Index. On Monday, the SOX gained 3.6% as the chip sector staged an applaudable rally.
But white-knuckled on-lookers are not convinced that the chips can regain the momentum of the early innings of this semiconductor
boom cycle. USB Piper Jaffray analyst Ashok Kumar threw a curve at the entire chip sector with bearish comments, declaring that
the semiconductor cycle had peaked. An all-star line up of analysts has since come to the plate with a more bullish outlook. The
bulls include Dan Morgan, a technology specialist at Noble Financial Group. Morgan, a widely quoted analyst and portfolio manager,
suggests that they'll be cueing up the highlight reel for semiconductors in the fourth quarter. In an exclusive StockHouse interview,
he names the star players.
It has been a month rife with uncertainty. Although fears that the Fed will raise rates at the August policy meeting seem to be
dissipating, the prospect of a slowing economy is stoking concerns in the high-valued tech sector. Add to this a question of slowing
growth in some high-tech markets that are the impetus behind semiconductors and telecom sectors, and you have the recipe for a
truly impassioned debate.
On Monday, Ashok Kumar questioned whether Advanced Micro Devices [AMD] could sustain
profitability. Kumar downgraded AMD to "neutral" from "buy" ahead of its earnings report this week. In
a report titled "Is The Party Over For SOX," he cited a moderation in the growth of desktop PC sales
and cellular handset sales in the third quarter of 2000. Kumar also noted that the flash chips used in
cellular phones and other electronic devices will be in oversupply by 2001. Under the scope of
technical analysis, he believes that the SOX has also peaked technically and could retest the 800
level before finding a bottom. Kumar stole headlines on Monday as his comments augur poorly for the
broader chip sector.
Daniel Morgan candidly suggests that many analysts are trying to make a name for themselves by guessing the SOX's peak. The
chips are not down, he argues, pointing to strong worldwide chip sales reported by the Semiconductor Industry Association (SIA)
last week. In June, global semiconductor sales surged 48.1 % over the same month in 1999, and improved 5.2% over the May
billings, to $16.6 billion. That marks the fourth consecutive record-setting month. However, the bears assert that strength in some
chip markets indicates that a loss of momentum is nigh. Flash memory sales, for example, saw a 167% upsurge from the June
1999 numbers. That's terrific growth, they say, but such lower-technology chips are more easily commoditized, and supply may be
close to meeting excess demand. In addition, one of the benchmarks for the semiconductor business--the capital equipment
book-to-bill ratio--is falling. The number of orders for semiconductor equipment as compared to shipments dropped from 1.38 to 1.26
in the last couple months.
Morgan, though, believes that the semiconductor cycle will continue upward for at least three years, because emerging industries
comprise a larger chunk of sales than ever. Growth in these new industries, including communications and the Internet, means that
semiconductor sales are decreasingly dependent on PC sales. PC sales now account for only about half of total chip sales. Just five
years ago, they were 85% of the total.
Communications--especially cellular phones--has been a huge driver for the current chip boom. Critics argue that growth
expectations for the industry were set far too high, and that major downward revisions are in order. Kumar's comments compound
the quandary that weighs on handset makers and their components suppliers. Suspicions of slowing worldwide cell phone sales
have hammered down shares of Nokia [NOK] by about 30%. The company reported strong second-quarter earnings, but guided for a
flat third quarter. Morgan says the resultant uproar that the "mobile phone business is going down the tubes" was laughable. He
cites Nokia's $2 billion surge in handset sales--from $2.9 billion in Q2 1999 to $4.9 billion in Q2 2000--as evidence of tremendous
growth. Says Morgan of Nokia's guidance for a slow third quarter, "It wasn't a broad-based industry problem." More accurately, he
attributes the slowdown to a product transition. Morgan likes Ericsson [ERICY] and remains very bullish on Nokia, particularly at its
discounted trading range.
The trickle-down effect from this apprehension has created a fantastic buying opportunity in Texas
Instruments [TXN], says Morgan. Trading in TI is closely tied to action in the handset business
because eight out of 10 cellular phones use the company's digital signal processor (DSP) and analog
chips. The perceived slow down in the handset business has been carried too far, says Morgan. And
he adds with some amusement, "The DSP market is not on the verge of collapse." Nonetheless,
Nokia's Q3 guidance caused TI to tumble to the $50 level, marking a 50% discount from its March
high of $100.
Also battered but not beaten is Intel [INTC], which Morgan is a long-time proponent of. He underscores Intel's 85% market share
position in the microprocessor market. Morgan is excited about the Itanium chip, which will facilitate the demands of Internet
e-commerce and high-end computing. Though the Itanium release has been postponed until the fourth quarter, Morgan is looking for
positive numbers in the company's third quarter report. Noting the deep discount in share price, Morgan wonders whether everybody
has forgotten that Intel guided for a very strong third quarter. Since a high last month of $73-3/4, Intel's share price has been socked
down to the $60s.
Morgan avoids investing in the memory chip makers. While flash memory has been hot, he prefers not to be burned by a flash in the
pan. Memory components such as DRAMs and flash memory chips are not as technologically complex as other semiconductors,
and thus are priced like a commodity. Capacity will eventually meet the demand curve, and flash memory will be a flash in the pan.
For investors seeking exposure to the scorching flash market, Morgan suggests taking a non pure-play approach by investing in a
company such as AMD, which has multiple product lines.
Price weakness in shares of customizable chip makers also gets Morgan's attention, because their
programmable logic devices (PLDs) are firmly entrenched in the evolution of technology. PLDs can be
programmed by electronics manufacturers to accommodate their particular designs. The greater
design flexibility and time to market advantages it offers is critical in this highly competitive
marketplace. According to the SIA, PLD sales doubled over the June 1999 period. Morgan likes PLD
leaders Altera [ALTR] and Xilinx [XLNX], seeing price weakness there. Wall Street was particularly
hard on Xilinx when the company called for 13 to 15% sequential growth, versus Altera's projection of
15%. Xilinx has since reissued their forecast at 15%. "So now, they are both saying the same thing: 15 percent," says Morgan. "But
everybody had [forgotten] that and it has already killed Xilinx." During the worst of the blood-letting, investors could have picked Xilinx
up at a 30% discount to its $98-5/16 high. Shares are trending higher in choppy trading, now at a 20% discount in the high $70s.
In the specialty-chip area, Morgan suggests that aggressive buyers examine Broadcom [BRCM]. Broadcom makes the chips that
go into the digital set-top boxes used by cable companies.
Strong chip demand also bodes well for semiconductor equipment makers. In the capital equipment arena, Applied Materials
[AMAT] gets top billing from Morgan. He notes that the producer of chip gear guided for 57% growth, well above the industry
consensus. He also likes ASM Lithography [ASML], the Europe-based maker of the systems used to imprint circuitry patterns
onto silicon wafers. Year-over-year sales growth of 54% in 1999 elevated ASML into the second position in its market, between the
leader Nikon and Canon [CANNY]. Morgan's research reveals that Nikon has been giving up market share to ASML for the past four
years.
Undoubtedly, the semiconductor sector ranks high on the vulnerability--and thus the volatility--scale. This provides terrific
opportunities to traders. If Daniel Morgan's assertions are correct, the semiconductor up-cycle will sustain exceptional growth
beyond the year 2001. Timing calls aside, there's no shortage of experts who agree with him. |
