Few issues in the global long-haul bandwidth business are as perennial as capacity growth and pricing trends. The former always is of interest as an indicator of sales growth and capital investment requirements; the latter is of interest as a key component to profit margins in a business that generally expects falling prices per bit over time. 

On both of those scores, recent trends over the last couple of years have been reassuringly stable. Annual global bandwidth growth has exceeded 50 percent over the past four years, according to TeleGeography (News - Alert) Research, driven as you might expect by more broadband access subscribers, more bandwidth consumed by most of those customers, and greater consumption of video content. 

Trends such of those naturally lead to questions about how long it might be until capacity ceilings begin to be reached. At the moment, only about 25 percent of the subsea capacity across the Atlantic, for example, actually is lit, says Tim Stronge, TeleGeography VP. At foreseeable rates of growth, of perhaps 40 percent annually, capacity could reach saturation by about 2015.

“For most routes, there’s the appearance that there’s lots and lots of spare capacity, lots of free wavelengths, so no new worries about growth, but appearance can be somewhat deceiving,” Stronge said. 

But there's nothing as certain as continued increases in long-haul optical fiber capacity over time, allowing network operators to upgrade without building new facilities. 

Global Crossing’s (News - Alert) Atlantic Crossing-1, for example, was one of the first private cables to go into the water in the Atlantic, and the maximum capacity of the cable originally was thought to be 360 Gbps, but now they think it is capable of doing nearly 3 Tbps, an 800 percent increase,” Stronge said. 

Of course, some will predict that if capacity increases about 20 percent annually while demand grows at 40 percent to 50 percent annually, there could be a capacity crunch at some point, even with wavelengths running routinely at 40 Gbps. Stronge estimates a capacity issue by about 2017. 

Most observers think the next round of increases, to 100 Gbps per wavelength, will require new construction, though. None of that would be terribly surprising, unusual or unexpected. All communications infrastructure is upgraded and replaced periodically. Physical obsolescence, however, is less an issue than technological obsolescence. 

Will we really run out of bandwidth, anywhere, by 2017 or by any other year? Not likely. Additional demand, so long as the business model is reasonably rational, will always create additional revenue to support new construction. And no trend continues in a linear fashion, forever. Nor can we be completely sure that consumer behavior will be unaffected by service provider business policies, packaging and pricing. Both supply and demand parameters are likely to evolve over time. 

Register for a replay of the webcast at http://tiny.cc/j9r1k.