Wednesday, May 22, 2013

"Build Where the Demand is Greatest" Even if "Universal Service" Suffers


One key innovation Google Fiber has brought to the construction of fiber to home networks is the “build first where there is greatest demand” principle.

Veterans of the telecom and cable TV business immediately will recognize that this flies in the face of established precedent that “universal service right now” is more the legacy principle. But proponents of gigabit networks already have moved to embrace the idea of “building where you can, right now” as a way to stimulate the building of gigabit networks on a wider basis.

That might now be the thinking of analysts at Point Topic, looking at building of new 30-Mbps networks throughout the European Union. In other words, the greatest progress, at the lowest cost, will happen when urban networks get built first, rather than giving priority to rural areas.

But analysts at Point Topic also do estimate it will cost far less than previously estimated to provide 30 Mbps service on a ubiquitous basis across the European Union.

Point Topic estimates it could cost €82 billion, though other estimates have ranged as high as that produced by the European Commission of €180 to €270 billion.

The FTTH Council quotes an estimate of €202 billion as the total cost of meeting the Digital Agenda targets with fiber-to-the-home networks.

The Point Topic estimate is dominated by the €52 billion cost for reaching rural areas, defined as those areas with a population density of less than 100 persons per square kilometer.

Covering the semi-rural areas, home to 15 percent of the EU’s population, would cost another €22 billion.

Completing coverage in the urban areas, those with a population density of 600 per square kilometre or more, would cost only €8 billion and reach 71 percent of the population.

That new estimate illustrates the problem all fixed networks face, namely the high cost of building networks in areas of low population density.

The cost of rural networks also accounts for the country by country costs of construction.

France has the biggest requirement of all at €17.5 billion, whereas the United Kingdom, similar in population but with only 37 percent of the land area, needs only €7.5 billion to build its national network. The reason is the higher percentage of rural areas in France.


The key insight here is perhaps not that the actual cost of building a fiber to home network has changed significantly. But there is a seemingly growing practical realization that high-capacity networks need a business model, and that model probably only works well in some neighborhoods in any given city.

Other initiatives such as Gig.U use the same principle, recognizing that gigabit networks cannot be build, or sustained, everywhere, right now.

But Google Fiber will test whether it is possible to operate at dramatically lower costs, or with a new business model, depending on take rates for a disruptive value proposition.



No comments:

Many Winners and Losers from Generative AI

Perhaps there is no contradiction between low historical total factor annual productivity gains and high expected generative artificial inte...