A Journey Into the Glasshouse
123Jump.com, 12/4/2000 4:33:00 PM
New York, Dec 04, 2000 (123Jump via COMTEX) -- The Michelson-Morley experiment (1887) demonstrated that the speed of light is always the same, and that it is not possible for an object or particle to travel faster than a determined 300,000 km/s. This gave rise to Einstein's assertions, beginning in 1905, that later revealed the laws governing electric and magnetic fields, and the laws that govern the motion of bodies - combined in a four-dimensional object called space-time. In 1851, roughly a half-century earlier, a glass business was established in the state of New York and later incorporated, December 1936, under the name of Corning Inc (GLW).
To associate theoretical physics with the age-old method of producing glass, at that time, would have seemed a senseless stretch of the imagination. Today, the melding of those two concepts drive the efficiencies behind the $1.131-trillion dollar telecommunications industry - an industry that accounts for one-seventh of the gross domestic product of the largest economy in the world.
Perhaps it was the $3.6-billion acquisition of Pirelli's components unit - an unprofitable company with one customer and a scant $25 million in annual sales - that dragged Corning shares from its 52-week high of $113. Or possibly sectoral weakness and industry-wide concerns could be blamed for bringing Corning's market capitalization down to just $50 billion. Whatever the reason, the matter at hand is to determine whether the company is accurately valued, and if not - what will drive its share price higher?
The company's three broadly based business segments,(telecommunications, advanced materials and information display) are enjoying terrific rates of development and expansion - an inverted trend to Corning's stock price - driven by robust new-information age demand for these products. The overarching success of Corning, however, has so far been driven by its aptitude for answering the additional needs of customers for its optical cables. These needs essentially aggregate into the production of high-end photonics components that efficiently use all of the glass that they sell. Today, 78% of sales come from new products, as opposed to 57% in 1998.
A recent Ryan, Hankin & Kent report is certainly encouraging for the sector; forecasting a quintupling of the market for Dense Wavelength Division Multiplexing (DWDM) components to $25 billion by 2005, amid damaging capital expenditure reassessments by service providers for next year. Those reassessments are currently all downward, yet RHK advises that no market within the DWDM market will grow less than 80% next year. Even HSBC is saying that sales will grow as fast, or even faster, in 2001 as they did this year - as the dark fiber is lit. The bottom line is some segments will grow rapidly and others will not. It is, therefore, key to understand which segments of the networking communications market will thrive, by exposing the real needs of those who will be spending the cash - the network owners.
From Glassworks to Gigabits
Optical cabling is currently a $10-billion industry with about 70 million kilometers purchased annually. The industry was conceived almost entirely by Corning in the 70's. Today, the market is dominated by Corning (85%) and Lucent (LU) (10%), and is powered by a robust build-out of communications networks in North America, Europe, and the Far East and through most of the world's oceans. It has become commonplace to take Internet connectivity for granted, few people are aware of the hair-thin optical threads that link the different continents and cater to a worldwide demand for 4,451 gigabits/second of data through pulses of light.
Known primarily to most households for its cookware brand, Corning has been involved in glass for technology applications since the last century - when the company produced the casing for Thomas Edison's first light bulb. The deregulation of telecoms, and the unbundling of the central switch, bred new carrier commitment to refurbish communication networks with fiber in place of copper. As a consequence of Corning's transmutation in the early 90's, the company now employs over 1,700 Ph.Ds and recognizes photonics components manufacturers such as JDS Uniphase (JDSU) and Lucent Technologies as its main competitors. Although not quite as drastic as Nokia Oyj's transformation from pulp and paper to communications devices, Corning's current trajectory is viewed as one of today's rare technology alteration success stories.
The drivers of growth have not been confined to the optical market, even though the company's fiber manufacturing capacity is pre-sold for months in advance. Information display, currently the smallest of its three business segments (which increased only 10% in 1999), has grown by over 63% in 2000 thus far. This segment includes flat-panel displays for devices such as laptops, flat-panel televisions and personal digital assistants, and is projected to grow to 1.2 billion in 2003 from about 400 million today. Corning is the worldwide leader in this market, having pioneered innovative technologies such as Active Matrix Liquid Crystal Displays, (AMLCDs).
Fiber, Still the Driver?
October 23 - after announcing third-quarter earnings that revealed $317 million (more than 100% year-on-year growth) in profits - a four-year contract to supply Level3 (LVLT) with 10 million km of fiber was disclosed. The strongest segment of the company's business is still fiber and cable; its phenomenal profitability perpetuates Corning's need to maintain control over this space.
Corning's LEAF fiber came under-the-gun when Lucent's competing product, AllWave, eradicated the attenuation spike at the 1400-nanometer range. Microscopic water molecules found in the fiber itself block infrared light and prevent a clear transmission of any wavelength of this size. Dense wavelength division multiplexing (DWDM) has enabled cramming of multiple streams of information down single threads of fiber, and the need (albeit a higher-end need) to counter the attenuation window at 1400nm is real. Also, it's not good for company morale to hear your competitors boast of their ability to jam 750 wavelengths at 50-gigahertz spacing down a single fiber while you're being harassed by H2O.
In its MetroCor fiber, a lower-end single-mode fiber, Corning labs have successfully eradicated the attenuation spike - thereby, increasing the passband by 40% and opening up the 1400nm window for transmission.
In development at Corning, is a solution to the problem that Erbium Doped Fiber Amplifiers (EDFA's) do not currently operate at 1400nm (Erbium does not have the atomic structure for amplifying light at those wavelengths). Once an amplifier has been developed, transmission distances of greater than 40km could be achieved and the product deployed in long-distance networks - currently where most of the wavelength cramming, or DWDM takes place. There would be significant cost advantages attached to the deployment of a fiber with a wider passband, demand for the product and the components associated with it would balloon. The global market for optical amplifiers is roughly $2.5 billion, and will grow by at least 80% next year. This may be a conservative estimate as current EDFA's, deployed by Sycamore Networks (SCMR) and Corvis (CORV), would eventually have to be replaced to fully capitalize on the newly-enabled capacity - the bottom line is that carriers who don't take advantage of this opportunity will lose out.
Given that the prospects for using plastic optical fiber (POF) seem to be diminishing - as neither Boston Optical nor Lucent can get around the problem of attenuation in POFs - and given the influx of contracts for optical fiber, the glass fiber market will continue to experience healthy growth. Of Level3's $255 million in third-quarter sales, $100 million was attributable to sales of dark fiber - broadly hinting that the pace of fiber being lit is comparable to the demand for those services carried over it.
Warning signs, however, include the fact that rates of lit fiber compared to dark, in both Europe and the United States, are still very low. Telecoms in Europe may have become slightly "overbuilt and underbooked", due to Nortel (NT) and Cisco's (CSCO) willingness to act as a credit facility to the world's emerging CLECs. Additionally, in stark contrast to Level3's results, it is uncertain whether Williams (WCG) will hit $100 million in dark-fiber sales for the whole year.
Cheap and Dirty Components
However, keeping Corning at the top tier of the optical industry, in terms of influence and innovation, is its aptitude for producing low cost photonic components. Only when a company has size, can it lead the way in transforming cottage-industry production into streamlined processes. With a new breed of startup touting futuristic microfabrication techniques, only entities that have mastered current methods will be able to reap benefits. In actual fact, Corning has as many patents for process as it does for its materials.
The terrestrial market for DWDM components, which is set to blossom from $4.5 billion this year to almost $25 billion by 2004, includes amplifiers, filters, couplers and lasers, among other odds and ends. In photonic components, Corning currently manufactures and sells DWDM modules, amplifier modules, thin-film filters, mutliclad couplers, fiber bragg gratings, pump and transmission lasers - not to mention switches and cross-connects, which aren't really lumped into the components category.
The next generation of amplification, that will accompany the PureGain product line, will be a hybrid Raman-EDFA amplifier. Driven by the prioritization of integration with other products in the network, a new smart card will be able to manipulate the gain on EDFAs and Raman amps, as different types of fibers require various types of customization. This trend toward integration, rather than spewing out proprietary components, is critical for product success and adoption.
The optical switch market - currently a nascent space - is forecast to grow from $25 million this year to $3.7 billion by 2004, driven by rigorous switching requirements at interconnection points in the new networks. Corning has bet on two different horses within this space. The first is based on liquid-crystal switching technology and has given birth to the Wavelength Selective Switch and the Dynamic Spectral Equalizer. Second, its complete acquisition of Intellisense has enabled it to demonstrate the use of MEMS (micro-electro-mechanical system) technology. Nortel's acquisition of Xros, for $3.25 billion, similarly gave it access to a MEMS-based all-optical switch. Using this approach, Corning claims it can use MEMS to create optical switch fabrics within cross-connects and other switching systems - and use the liquid-crystal approach to route light within switch fabrics.
The central idea that Corning has the capacity and capability of, firstly, coming up with an appropriate technology, and then generating an efficient process behind it, is clearly to the company's advantage - in comparison with Lucent or other smaller competitors. To use the example of thin-film filters used in DWDM modules; Corning has reduced cycle times and fashioned a 25% increase in production yields.
The Pros and Cons of Buying Progress
Nortel, Lucent and Cisco have all attempted, within the past year-and-a-half, to drive innovation through acquisition - in the case of Cisco, much of its growth in revenue has been a result of its 30-odd purchases since 1999. As valuations of the buyers inevitably fluctuate, the pursuit of acquisition-based strategies is limited to those periods when a company feels rich. Lucent has seen its shares slide to one-sixth this summer's average value. To a certain extent mismanagement of integration, (the Nexabit fiasco), is at fault. However, the company is certainly not in a position to aggressively continue the trend of buying technologies.
Corning has made some questionable moves as well, however they haven't been tremendously costly and the total bill since January 2000 is under $12 billion - given Corning's size that really isn't much. Broadcom (BRCM) for example, whose cap is now one-fifth that of Corning, has spent $6 billion this year on buying smaller companies. Nortel has bought a total of 16 companies since 1999, with a total price tag of almost $27 billion. The most recent acquisition announced by Corning was on November 30 and that was a mere $8.5-million purchase of a company called Silentor NoTox, based in Copenhagen. The company makes diesel filters using silicon carbide materials for cars, trucks and buses - not exactly oozing with overtones of connectivity.
The point is that Nortel's 40% revenue growth has been driven by its acquisitions, Cisco's 66% increase in revenues has been driven by acquisition, but Corning has been able to demonstrate a 54% year-on-year growth in revenue largely because of strong demand for its best-known product - fiber. It is, therefore, not as exposed to risk of a slowdown in growth (should a stock market aversion persist) as those companies buying their mass in exchange for equity.
Turbulent Times for Technology Stocks
With over 50% of the population having some sort of exposure to stocks, it is hardly surprising that, when the indexes head south, pointers such as consumer confidence and retail sales follow suit. However, just as the equity markets are driven wildly up by distortions in perception they are also dragged down by the same practice of misconception. We can, therefore, rest assured that the current journey southward is partially due to irrational disaffection - once irrational exuberance. And, before these sentiment-driven measures begin to influence lagging indicators such as the Misery Index (the rate of unemployment plus the rate of inflation), one expects that the current slide will rein in.
Naturally, Gateway Computers' (GTW) profit warning on November 29 compounded investor concern regarding a slowing technology sector, dragging the market to a new 52-week low. And, Nortel's third-quarter slow down in revenue growth and a decline in optical revenue have done nothing to aid sentiment. Furthermore, it is quite obvious that an expected $170 billion in telecommunications expenditures will most likely not be met. This, however, does not mean that carrier spending slow down is going to target every company or space in the sector. Unless there is a fundamental overshoot in the technology provided - as compared with the real needs of telecoms - the building blocks will continue to sell and competition will continue to spur innovation.
Whether one considers the hybrid amplifier or the all-optical switch based on liquid crystals (expected to outlive MEMS competitors because of their lack of moving parts or mechanics), Corning has displayed a knack for solving problems cheaply and intuitively. Furthermore, a wealth of value lies in the company's ability to leverage advances in one area and use the technology for others (an example being liquid-crystal displays) - reminiscent of Agilent's (A) application of bubble-jet technology to its photonic switch.
A show of forgiveness by the market is really a function of industry growth in the fourth quarter and, specific to Corning, its ability to turn Pirelli and its agreement with Cisco into success stories. Another curious issue lurks, however, and that is the issue of DWDM capability. The company began by innovating in this space, bringing with it one of the industry's first 32-channel systems. However, competition in this space is now being guided by new entrants, such as Avanex (AVNX) and Corvis, demonstrating miniscule spacing and blizzards of channels down single fibers. Corning's solution to this problem - assuming track record is something to go by - is eagerly awaited by all. |
