One of the chief benefits of LEDs, more generally known as solid-state lighting, is that they last a very long time—in some cases, orders of magnitude longer than incandescent or fluorescent devices. Eventually though, everything fades, and few things fade faster than profit margins in high tech.
After years of torrid growth, commoditization is coming to high-brightness LEDs, the current darling of the solid-state lighting industry. New entrants are driving average sales prices down and—adding insult to injury—industry experts are predicting softening demand. In response, the industry is looking for new applications and individual businesses are splitting and recombining, looking for economies of scale.
The solid-state market will continue to grow, just more slowly that in the past. Recent reports from analyst firms Strategies Unlimited and iSuppli predict growth in the 10 to 15% range for the next few years, which is slow only in comparison to the 46% annual growth that the high-brightness LED industry has enjoyed since 1995, according to Strategies Unlimited.
What's more, LEDs are expected to eventually dominate both automotive and general lighting. It will take several years for those markets to mature, but given the size of the prize, the R&D competition is predictably fierce. The next few years may not be caviar and champagne for solid-state lighting, but nobody is going to be eating dog food either.
High-brightness actually identifies the middle of three categories into which observers divide the solid-state lighting market. Standard-brightness LEDs are the twinkling lights that festoon more and more devices, such as the power and hard-drive activity indicator lights on your computer or notebook. Available by the thousand for less than five dollars, these lights are heavily commoditized, with the total market hovering around $2 billion per year.
High-brightness LEDs are where the action has been recently. More expensive but also far more durable and long-lasting than either incandescent or fluorescent lighting, high-brightness LEDs have been adopted where they can be used in batches: traffic lights as well as running lights, brake lights and turn signals for trucks and automobiles. Over the past two years, those uses have paled beside the volumes generated by the cellular handset market, where they are used as backlights for keypads and displays and—more recently—as flashbulbs for camera phones. According to iSuppli, total sales for high-brightness LEDs reached $2.7 billion in 2004 (see "High-Brightness LED Market Growth," below).
But the fat days for high-brightness LEDs are over, in part because demand from the handset market is softening, according to Jagdish Rebello, principal analyst for iSuppli. "Growth rates are slowing for handsets, down from 17% to 6% this year," he says. LED adoption in the automotive market is increasing, but even if manufacturers can sell more LEDs per vehicle, the differences in volume are just too great. "Seven hundred million handsets versus sixty million cars and trucks per year isn't even close," says Rebello.
And competition will continue to intensify. Nichia and Toyoda Gosei (Japan), Cree (United States) and Osram (Europe) once dominated the market. But since 2003 new suppliers in Taiwan, South Korea and China have entered the market, driving average sales prices down by as much as 15% each year. Chinese competitors are likely to become even more of a factor in the years to come: In June 2004, the Chinese government announced a national project to develop its semiconductor lighting sector. According to a report in China's People's Daily Online, China plans to invest 10 billion yuan (roughly $1.2 billion) to kick the project off.
These dynamics are spurring some market consolidation with an eye toward economies of scale. In August 2005, Agilent shed its LED-manufacturing unit as part of the sale of its semiconductor products group to Kohlberg Kravis Roberts and Silver Lake Partners. Later that month, two of Taiwan's largest LED manufacturers—Epistar and United Epitaxy (UEC)—announced plans to merge under the Epistar name, citing size and production capacity as key strengths of the new combination. Rebello believes that price erosion will lead the group that acquired Agilent's chip business to "eventually sell Agilent's former LED manufacturing unit to an existing LED manufacturer in the Far East."
Other companies are breaking into new technology. Agilent also sold its share of Lumileds, an LED lighting joint venture, to its partner Philips. Lumileds is one of a handful of companies (along with Nichia, Cree, Toyoda Gosei and Osram) with significant ultrahigh-brightness (UHB) expertise. UHB lamps will be as much as 100 times as bright as high-brightness LEDs and will eventually compete with both incandescent and fluorescent light sources for everything from automobile headlights to indoor and outdoor lighting.
For instance, an automobile headlamp built with as few as nine UHB lights would require less power than existing incandescent lamps and would allow beam modulation tricks that are practically impossible with incandescents. "With nine discrete light sources, you have independent control over intensity, you can focus beams and use multiple levels of brightness," says Rebello. However, he adds, many technical challenges remain. "UHB LEDs generate tremendous amounts of heat that must be dissipated, and they're still very expensive." Not to mention that the products currently in the lab don't put out enough light—yet.
It's hard to gauge whether consolidation or new technologies will reignite LED market growth. Most observers don't expect solid-state lighting to compete with incandescent or fluorescent lighting in the general illumination market until the end of the decade, which means that LED manufacturers are facing a few relatively thin years. On the other hand, it's hard to be too sympathetic when the low estimate for growth is still at 10 percent. "There is some consolidation going on, but the future is not dim," says Rebello. "This is sustainable long-term growth."
