This article was originally published in Maclean's Magazine on March 27, 2000
JDS Uniphase Corporation
Jozef Straus has a reputation for being a bit, well, eccentric. There's that funky black beret he wears practically everywhere but in the shower - his company, Nepean, Ont.-based JDS Uniphase Corp., has even taken to handing copies of it out to visitors. Then there are the watches: the Czechoslovakian-born physicist keeps one strapped to each wrist. The first is set to his current time zone, the second to the time in the city he intends to visit next.
Actually, when you think about it, that one kind of makes sense. But now look down at Straus's feet: on this particular day in early March, his navy pin-striped suit is absent-mindedly paired with two different shoes - one a full brogue, the other a half brogue, both black and, if the truth be told, badly in need of polish.
No matter. People are in the habit of forgiving Straus his little quirks these days, especially if they are fortunate enough to be JDS Uniphase shareholders. In Toronto recently, one grateful investor walked up to Straus in the lobby of his hotel and asked if he could kiss him on the cheek. The stranger had evidently just finished paying for a new kitchen with the profits he had made from investing in JDS Uniphase stock. The kiss, which Straus accepted, was simply the shareholder's way of thanking the 53-year-old president and co-chairman for a job well done.
Well done may be an understatement. Since last June, when the fibre-optics equipment maker formerly known as JDS Fitel Inc. merged with Uniphase Corp. of San Jose, Calif., the company's shares have rocketed from $28.75 to a recent high of $219, giving the company a market value of $64.2 billion - equal to the combined value of Canada's three largest banks. (Last Friday, the shares closed at $189.) Revenues in the most recent quarter, which ended on Dec. 31, were 119 per cent higher than the combined sales of the two companies in the same period a year earlier. The bottom line: if an investor had been given the choice a year ago of buying shares in either Straus's company or that larger and far better-known symbol of Canadian high-technology prowess, Nortel Networks Corp., he or she would have done twice as well taking a flyer on JDS.
Like many of today's stock market darlings, JDS Uniphase owes its good fortune to the explosive growth of the Internet and electronic commerce. But don't confuse Straus's company - with more than 10,000 employees and $1.6 billion in annual revenues - with all those dot-com start-ups that appear out of nowhere and make their founders rich on the strength of a half-written business plan and generous servings of hype. Far from being an overnight success, JDS has been plugging away for 19 years, ever since Straus and three of his co-workers at Bell-Northern Research Ltd., the defunct research arm of what is now Nortel, decided to go into business for themselves making components for fibre-optics networks. At the beginning, they ran the company out of a basement, never dreaming that one day the technology they were working on would become one of the linchpins of a communications revolution that touches almost every aspect of modern life, from home entertainment to international finance.
As it turned out, their timing was perfect. In the early 1980s, telephone companies around the world were just beginning to grasp the potential impact of fibre-optic communications - the use of light rather than electricity to transmit large quantities of information - on their industry. Although the technology had been around in a variety of experimental forms since the 1930s, it was not until 1977 that GTE Corp. of Stamford, Conn., installed the world's first commercial fibre-optic telephone circuit, carrying long-distance calls between two suburbs of Los Angeles. Gradually, other phone companies followed suit, using fibre to replace older, less efficient copper wires. Today, there are more than 230 million kilometres of fibre-optic cable around the world handling 80 per cent of all long-distance voice traffic.
At Bell-Northern Research in the late 1970s, Straus was one of a team of physicists and engineers whose job was to design components for Nortel's first generation of fibre-optic systems. They soon realized there was a gap in the company's R and D operations. "Our job was to develop a few samples and then go on to something different, but the people in the labs who were developing the systems needed a continuous supply of these components," says Gary Duck, one of Straus's co-workers at BNR. "In those days, you couldn't buy them off the shelf, so we started working overtime to meet the demand." Eventually Straus, Duck and two other colleagues - Philip Garel-Jones and William Sinclair - decided to put their experience to better use by setting up their own company to design and manufacture parts for optical networks. They called it JDS after the initials of their last names, omitting the second S because, to their ears, three letters sounded better than four. "After all," says Straus, "they didn't call it IBMM."
None of the founders had run a business before, so at first things were pretty informal. "This was not a classic Harvard school of management situation where you write a business plan and a profit and loss statement," Straus recalls. "God forbid - we never did. We were more like little elves. If some company had a problem, they'd call JDS. Today, there are far smarter people competing with us, but in those days we were the problem-solvers. It was lots of fun."
By 1989, JDS Fitel had grown to 70 employees and a respectable $7 million a year in sales. Revenues were rising by about 35 per cent a year and the future seemed bright, but even then none of the founders foresaw just how big the fibre-optics market was destined to become. One reason is that, until the mid-1990s, the growth in demand for telecommunications capacity, or bandwidth, was more or less linear. Every year, telephone companies installed more lines and their customers placed more long-distance calls, with the result that the industry's overall traffic levels were increasing at between five and eight per cent a year. The widespread introduction of fax machines in the late 1980s and the growing use of corporate computer networks added modestly to the demand, but didn't fundamentally alter the industry's need for capacity. Neither did the Internet, which in those days was used almost exclusively by university-based researchers. In such a predictable environment, phone companies typically bought capacity on intercontinental routes 20 or 25 years ahead of time, confident that when the time came they would have enough bandwidth to handle their needs.
What changed everything was the invention of the World Wide Web in 1990 by Tim Berners-Lee, a British computer scientist then working in Switzerland. Until then, virtually all of the traffic over the Internet had consisted of things like e-mail and other text documents. The Web, however, made it possible to combine text with photos and other graphics, the transmission of which typically requires far more bandwidth. More recently, Web page designers have begun to make extensive use of audio and video files, increasing by orders of magnitude the amount of network capacity required to download information from those sites. A simple 15-second video clip, for example, can easily consume as much bandwidth as several full-length novels.
The other major change that has fuelled demand for capacity on the world's telecommunications networks is, of course, the rapidly growing number of people and organizations using the Internet. Until the late 1980s, the Net was effectively off-limits to most computer users. The volume of online traffic picked up after the appearance of the first public-access Internet service providers in 1989, but what really caused Internet usage to soar was the introduction of the first commercial Web browser, Netscape Navigator, in 1994. In that year, there were an estimated 13.5 million people around the world hooked up to the Internet. Now, there are 130 million active users, and the number is expected to hit 350 million by 2003. And instead of swapping simple text files, many of those people are exchanging things like digitized songs and even pirated full-length Hollywood movies - driving bandwidth consumption to previously unimagined levels. (Ironically, Straus himself is not a regular Internet user. "I get on once in a while, but actually I'm a computer illiterate," he says.)
The Web is proving to be a boon for the entire fibre-optics industry, but particularly for companies such as JDS. That's because its specialty is a technology known as wavelength division multiplexing, or WDM. Fibre-optics systems work by turning digitized information - essentially a stream of ones and zeros - into rapid-fire pulses of invisible infrared light generated by a laser beam. Travelling at the speed of light, 300,000 km per second, the pulses move along a hair-thin fibre made of glass so pure that a person could look through a wall of it 70 km thick and still see a light bulb glowing on the other side. The first fibre-optic networks in the 1970s used a single beam of light, but before long scientists figured out how to split that beam into two colours, or wavelengths, thereby doubling capacity. In recent years, as demand for bandwidth has increased exponentially, researchers have developed new systems that can split light into many more wavelengths, each carrying a separate stream of data.
The latest multiplexers from JDS Uniphase are capable of dividing light into 160 wavelengths, allowing phone companies to massively increase the capacity of their fibre networks without having to install new cables. The equipment to do so isn't cheap - such upgrades can run into many millions of dollars - but even that is far less expensive than the traditional approach of digging a trench and installing new lines. The rule of thumb in the industry is that the cost of laying new fibre from one city to another works out to about $300,000 per kilometre. What's more, a project like that generally stretches out over many months, whereas an existing line can be upgraded with the latest WDM technology in as little as a few weeks. "The growth in bandwidth demand today is very fast and WDM is really the only solution," says Joseph Ip, senior vice-president for product strategy at JDS Uniphase. "Any other method will take longer and probably be more costly."
One of Straus's favourite sayings is that nobody has a monopoly on brains, and it's true that JDS owes much of its success to luck: the company is in the right place at the right time, with products that the telecommunications industry can't do without. But Straus also deserves credit for the skilful way he has steered the company. In particular, analysts praise his decision last year to merge JDS Fitel with California's Uniphase, a similarly fast-growing company that was the leading supplier of so-called active fibre components, including lasers, to telecom equipment providers such as Nortel, Lucent Technologies Inc. of Murray Hill, N.J., and Alcatel SA of France. JDS didn't make active components, but was facing pressure to get into that market so that it could supply its customers with the full range of fibre-optic hardware. For the same reason, Uniphase was thinking of entering the market for the passive components that JDS made. By choosing instead to join forces, the two companies avoided what could have been a costly head-to-head battle.
By all accounts, the corporate marriage has gone smoothly. It was Straus who initiated the courtship, phoning Uniphase CEO Kevin Kalkhoven in the summer of 1998 to propose that the two men go hiking alone in the Rockies to talk business and see how well they got along - a "classic male-bonding exercise," in the words of Zita Cobb, a longtime JDS executive who is now senior vice-president for strategy and integration. They spent three days together in British Columbia's Yoho National Park, a trip that has since become part of company lore. "Really, the famous hike was a stroll," says Straus, laughing. "We went about half a mile into the mountains and it started to get too steep so I said, 'Let's go for lunch.' " Still, in every other respect the outing was a success. Five months later, the merger was announced, a $4.7-billion union of equals that at the time was the second-biggest deal in Canadian high-tech history.
Just as the JDS and Uniphase product lines dovetailed perfectly with each other, so, too, do Straus's talents seem to complement those of his new partner. Kalkhoven, a 54-year-old Australian with a background in marketing, operates from the company's U.S. head office in San Jose. As CEO and co-chairman, he has overall responsibility for running the business. Straus lives with his wife and their three children in Ottawa, site of most of the company's R and D and manufacturing operations, but spends about half his time travelling around the world meeting customers and visiting JDS Uniphase facilities in the United States, Europe, Australia and Asia. (Two of the other founders, Duck and Garel-Jones continue to work at the company, while Sinclair sits on the board of directors.)
The pace is hectic, with growth the only constant. In the past six months, the company has announced seven takeovers, worth more than $27 billion. When those deals all close, the company's workforce will swell to more than 12,500, of which more than 7,000 will be located in the Ottawa area, spread among 17 buildings. To make room for further expansion, the company's 46,500-square-metre main building will soon undergo a 33,500-square-metre expansion. When that space is fully occupied, there are plans to develop an adjacent 21-hectare site that is now just an open field. "If you're managing growth of 35 or 40 per cent a year, it's challenging, but you're not in auto-panic," says Cobb. "But when you're doubling in size every year, as we have for the last several years, it just never lets up."
Right now, the company's biggest problem may be keeping up with the towering expectations of its shareholders. Currently, JDS Uniphase has a price-earnings multiple (the ratio between its stock price and its annual per-share profits) of 330, nearly double the average for companies in the telecommunications equipment sector and 13 times the average for all companies in the U.S. Standard & Poor's 500 index. Straus and his colleagues are heroes now, but they're keenly aware that investor sentiment can swing wildly from one extreme to another when companies fail to meet the market's expectations for future growth. "We know that lots of people are reasonably content now and we hope it continues," Straus says. "But if success is defined by this, what would happen if there wasn't success? It wouldn't be a pretty sight."
Not that Straus sounds particularly worried. It's often said by scientists who work in the field of fibre-optics that the industry is now at a stage equivalent to where electronics was in the 1950s, before the invention of the integrated circuit and the microprocessor. If that's true, the next two or three decades should bring a succession of technological breakthroughs, creating new opportunities for companies such as JDS Uniphase and greatly increasing the size of the market. "It's for us to screw it up, but the industry's moving," Straus says. "And if the industry's moving, we should be able to move with it." Moving, that is, at the speed of light.
Maclean's March 27, 2000