The Broadband forum said this week that global broadband subscribers hit the 400 million mark.
Interestingly the company that gave them that information was Point Topic, a UK research house that does a good job of counting broadband and looking at what drives revenues on broadband.
Back in 2003 there were just 63 million broadband subscribers according to the same source, and Point Topic produced a projection of where broadband would be over the coming years. Our own projection listed below was a simple linear calculation working out how fast broadband installation was accelerating and then extrapolating. Although at the time our numbers were higher than anyone else’s, they were not high enough and instead of hitting 400 million in the middle of 2010, we’ve hit it almost two years earlier.
(see pf version for table)
Most broadband watchers, Point Topic among them, have revised their forecasts each year, but there is little point in getting to the absolute number. The rate of install is clearly above the 43 million ADSL plus the Cable 9 million (both numbers were too low, but for different reasons) ‘ at a combined rate of around 52 million a year, which we predicted, perhaps closer to 70 million new lines a year. We needed to factor in the 45 million fiber lines, and the Ethernet direct connects and the WiMAX subscriber homes and those connected by Powerline or straightforward Wi-Fi, and now we have to think about broadband wireless ‘ simple HSDPA, and the beginning of a process by which broadband lines may start to erode in favor of portable broadband – cellular or WiMAX, depending upon the technology preference in any given region.
Back in 2003 Point Topic estimated that the ceiling of demand might be about 30 to 40 lines per 100 people and on that basis calculated that broadband could grow to ten times the 2003 number so around 630 million connections globally.
But history tells us this will be way too small and not only are the total number of broadband homes likely to be way higher than we anticipate, but the total bandwidth will be at least ten times that which we all predicted back then, as the bandwidth of access technologies have climbed from the prevalent 500 Kbps to more than 100 Mbps for some technologies
Point Topic at the time posited the idea that GNP of a given country had some bearing on penetration of broadband lines, perhaps what it meant was the rate at which broadband lines were installed. Naturally a recession will have an impact on the rate of slowdown on this installation rate and in fact it is likely to be a key indicator of just when the recession ends.
But the number of broadband lines may well begin to ceiling once we run out of actual physical phone lines, which is around 1.3 billion, which is more or less every fixed phone line in the world. Obviously the rate at which this happens will slow and slow until it is a crawl and we are picking up only a tiny amount of total growth. At the current rate of broadband roll out this might take 20 years, whereas actually it will take much longer.
Another problem is that with HSDPA currently offering around 3.6 Mbps broadband access shared across a base station sector, and this moving on to 7.2 Mbps and 14.4 Mbps over the next few iterations, mobile broadband will begin to ‘interfere’ with these numbers. In some parts of the world there will begin the process of wireless fixed broadband substitution, while other parts of the world will still be growing rapidly.
Point Topic said ‘The emphasis is now going to shift to providing high bandwidth, high quality broadband, that can deliver multiple, steady, pin-sharp images for applications. To do this, we’ve got to have advanced global standards for end-to-end delivery of broadband data streams.’ Which is what the Broadband Forum would want them to say since that’s what they do.