Google Fiber Costs Do Not Appear to Have Been Materially Better than Any Other Telco

One key issue since the advent of Google Fiber, as well as market entry by any number of other independent Internet service providers, is whether Google Fiber had uncovered some cost advantage over all other leading providers that would allow it to make a profit selling gigabit Internet access connections at $70 a month, when other major ISPs were selling services operating at far lower speeds, at comparable prices.

To be sure, getting streamlined permitting processes from cities arguably helped a bit. Building networks neighborhood by neighborhood was an important innovation municipal regulators decided to allow.

But it never was clear that Google Fiber had material advantages in construction costs that represent perhaps two thirds of the total cost of building a new fiber to home network.

A decade has passed since the first FTTH network deployments, yet the cost of building
a network remains the primary obstacle to ubiquitous fiber connectivity for every household,” says Commscope.

From 2005 to 2015, the cost per home passed dropped from $1,021 to just under $700, Commscope notes. Those costs likely are fairly standard, no matter how big or small a firm might be.

The problem is that most of the cost of building a fiber-to-home network comes from civil engineering, not network elements.

Construction, civil works engineering, obtaining permits and right-of-ways account for roughly 67 percent of total cost, while the equipment accounts for about 33 percent.

So while GPON and fiber equipment costs have indeed fallen, skilled labor rates have risen.

In other words, a fiber-to-home network mostly represents construction costs, not network element cost.

My simple way of explaining this is that most of FTTH cost comes from “digging holes, then closing the holes back up.”

If so, then the cost of FTTH cannot be reduced too much more.

That rather suggests that Google Fiber has no particular business advantage in construction costs.

Consider that Dycom Industries, whose main business is network construction for tier-one telecommunications providers, counts AT&T, Comcast, CenturyLink, Verizon and a “customer who has chosen to remain anonymous” among its top-five customers.

Most everyone believes the unnamed customer is Alphabet (Google Fiber).

If so, it is unlikely Google Fiber has material advantages in either network elements or construction cost. It might have some marginal advantages in permitting and other sorts of make-ready work, but those are not the primary cost elements.

Perhaps Google Fiber has saved a bit by making its own set-top boxes for video, as well as network interfaces for Internet access services. But not necessarily. At low volumes, Google Fiber might well have spent as much, or more, than it would have spent buying gear off the shelf.

Nor is there any particular reason to believe Google Fiber has gotten network element prices very different from what AT&T, Comcast or Verizon might pay. In fact, if volume discounts apply, then Google Fiber might be paying higher prices than AT&T, Comcast and Verizon.

With rumors that Google Fiber has fallen quite short of its subscriber forecasts, and might be getting ready to cut its workforce in half, it might be reasonable to assume that whatever else might be the case, Google Fiber did not uncover some new cost-saving way of building a fiber to home network.

One might have hoped for lower overhead costs, something that seems key to success for small, independent ISPs. But Google Fiber probably did not have overhead costs materially better than Verizon or AT&T, and perhaps had overhead higher than that of Comcast.

Even if Google Fiber had some marginal cost advantages in a few areas, it does not appear that the cost side of the network build was materially different from any other bigger providers.