Jeff Stambovsky Avanex article.
The Avanex Advantage
by Jeff Stambovsky
Perhaps the most exciting breakthrough of the last decade in the practical end of the communications sciences has been the development of Wavelength Division Multiplexing. WDM enables network builders to send heretofore unimaginable amounts of data down hair-thin strands of optical fiber. Our broadband future -- high-definition teleconferencing, movies-on-demand, three-dimensional holographic TV, long distance virtual reality, Captain Kirk-ish "telepresence" -- largely depends on the continuing progress of two trends: Rapidly increasing the carrying capacity of fiber optic circuits, and switching and routing the light beams carried on those circuits ever more cheaply and efficiently. Avanex's breakthrough technology is crucial to both and the company's execution so far has been superb.
WDM is the art of placing more than one information-bearing light beam on a single fiber, which is achieved by using different wavelengths (actually bands of wavelengths, called "lambdas") of light as carriers. The trick is to keep the lambdas from drifting off center and interfering with each other. A measure of how difficult this is, is that when the first commercial WDM systems were introduced, barely five years ago, they placed only two or four lambdas on each fiber. Eighteen months ago, 32 was state of the art. The Avanex engineering team, led by Chief Technology Officer Dr. Simon Cao, is pushing that number to a thousand, and in the lab, to a hundred thousand.
Not only does that sort of scale hugely increase the carrying capacity of a single fiber, it will eventually change the entire architecture of data networking. Today virtually all data networks require different users to share the same circuits. That requires lots of expensive electronic processing. But Avanex's breakthroughs hold out the promise of a return to classic circuit switching, with every user having his own waveband, dramatically speeding the network, and cutting costs.
A three-year-old prodigy that just went public last February, Avanex surprised Wall Street when it turned its first profit in the most recent fiscal quarter, handily beating the expectations of most analysts. Consensus estimates call for annual sales exceeding $200 million in the coming fiscal year. Industry consultant RHK, Inc. projects that the market for WDM equipment, presently estimated at $5 billion annually, will grow to $24 billion by 2004. That's a 50 percent annualized growth rate. Assuming that Avanex merely holds its own, the $1 billion sales mark should be in reach within the next three or four years.
Expect the company to do far better than hold its own, however. Not only is Avanex's patented WDM technology far superior to competitors in the number of lambdas it can place on a fiber, it is cheaper. On a per lambda basis, Avanex has driven down prices an order of magnitude in a little over a year. And the Avanex devices are easier to manufacture for the same reason they work better: Whereas competing designs can require the precision alignment of hundreds of elements, the Avanex multiplexer (called the "PowerMux") has only two and is self-aligning. Expect the company to gain market share rapidly and dominate the WDM market within a few years.
Still isn't 50 percent annual sales growth from now to 2004, and beyond, a bit of a stretch? Not necessarily. Cisco grew its top line 50 percent last year, and it is a much larger and older company in a much more mature market. Avanex's historical growth rate, to the extent that the number is even meaningful, has actually been well in excess of 50 percent: Quarterly sales went from $4.4 to $34.8 million on a year-over-year basis, and from $19.3 million to the same $34.8 million sequentially.
So how much is Avanex worth, or will it be worth by say 2004, compared to early January's stock price of around $62? What is a company with a sustainable 50 percent growth rate and $1 billion in sales (with healthy margins!) likely to be worth? Assuming that the game is nowhere close to over in 2004 and that Avanex continues to execute, a multiple of 20 times sales seems reasonable. Consider some of the multiples prevailing in early January, even after a nine-month process of compression, the worst year in NASDAQ history and the worst performing Dow in 19 years: CIEN-30, BRCM-28, SCMR-30, JDSU-21, CSCO-15.
At 20 times sales, the 2004 Avanex would be worth $20 billion or about $307 per share. Even at CSCO's price to sales multiple of 15, Avanex would be worth almost $230 a share, up nicely from early January's $62.
Beam me up, Simon.
--------------------------------------------------------------------------------
Jeff Stambovsky is a senior analyst with Gilder Publishing, publishers of the Gilder Technology Report and other strategic publications for technology investors. He may hold positions in stocks discussed in this column. Spectator readers who use this information for investment decisions do so at their own risk.
This article also appears in the February 2001 issue of The American Spectator.
(Posted 2/21/01 |
