Former Google Chief Executive Eric Schmidt recently penned an opinion piece for the Financial Times where he basically argued that the current model used by the U.S. government to auction off wireless spectrum was detrimental to the ongoing deployment of 5G networks.

Schmidt based his argument on the Federal Communications Commission’s recently completed auction of so-called C-Band spectrum that sits in what’s considered the mid-band of wireless spectrum resources. That auction involved most of the country’s largest wireless operators and generated more than $80 billion in winning bids.

WARNING: Boring spectrum history lesson

The U.S. government has unleashed spectrum in this auction manner for decades because it generates significant revenue for the federal government from what are seen as the “people’s airwaves.” Most operators have become used to this model, and while they obviously don’t like to spend more than they need to, they have built the process into their business models.

The other way governments have distributed wireless spectrum to support telecommunication services has been by just allocating that spectrum to certain companies. This removes the benefit of gaining monetary compensation from public companies that are given exclusive access to a common asset, but theoretically frees up capital for that public company to more quickly (and profitably) launch services that benefit consumers.

The FCC did, in fact, use this model for its first telecom spectrum distribution process more than 35 years ago in a move that spawned today’s wireless telecommunications market.

Schmidt’s argument is that operators will have to fork over so much money to pair for the airwaves to transport 5G signals that they won’t have enough money left over to actually deploy the physical parts of that network.

Arguments for both models have been ongoing for decades, and different countries have at different times used both models. Whichever model has been used, it’s safe to say that wireless telecommunication services are generally robust, and the companies operating wireless networks are generally profitable.

But this is not the part of Schmidt’s piece that I found most interesting.

Boring spectrum history lesson over

The part that I thought was more interesting was the part where Schmidt claims that it will require 1 million new cell sites to provide 5G service using that spectrum to 80% of the country.

Putting those numbers into some perspective: 80% of the country covered still leaves out 60 million people. And if history is any indication, that 20% will continue to be those on the wrong side of the digital divide.

Additionally, that 1 million new cell sites is more than double the number of cell sites that dot this country today. If you think city council meetings are overflowing today with citizens complaining about so-and-so carrier wanting to put in a new 5G cell site today, just wait until those carriers are looking to install three-times as many sites.

New cell sites are required to support true 5G services. Not the services we have today where we can post all of our daily activities to every social media outlet, but the 5G services that come from vastly higher speeds, lower latency, and near-ubiquitous coverage. You know, real 5G.

These services will require more antennas to beam out and receive signals, each of those antennas somehow connected to power (green power?), and some way to transport that traffic back to the internet. That will require these antennas to be either located somewhere that already has these connections or considerable construction to install those connections.

(And if you think the number of people that will pack city council meetings complaining about new 5G antennas will be an issue, just wait until you add those folks that want to complain about construction tearing up the streets/causing traffic/generating noise.)

True cost of 5G: physical vs. monetary

These challenges don’t take into account all of the technological 5G issues that are ongoing, which while considerable, can be overcome with enough money and time. The physical requirements of actually putting 5G into use are where the true cost of 5G challenges lay.

Some may point to satellite services as being a way to deal with those physical challenges. But ask anyone with a current satellite radio service how well that service works when there is anything but blue sky overhead and it’s obvious that is not a real answer.

I instead view the roll out of 5G service as being a bifurcated process where certain controlled locations and environments like buildings and stadiums will be infused with all of 5G’s possibilities, and most physical locations will just have faster access to their favorite cat video.

Perhaps this is not a bad thing, and if done with a modest view of social good, could produce true societal benefits. But the hurdles to this are large and local in nature, which could prove to be more a more important measure of the true cost of 5G than the tens of billions of dollars in expenses companies worth hundreds of billions of dollars will need to spend.