Magical Reasoning About Title II Regulation

With all the hoopla in today’s news cycle about payments to consultants, the hiring of lobbyists, and the drama over the Senate Democrats’ CRA resolution, you may not have noticed the letter from House Democrats to the FCC about broadband mapping. The letter is a lot like the ongoing complaints about the FCC’s failure to verify the identities of commenters on net neutrality inquires: pointless.

The FCC is required to take comments from the public the course of making or altering rules; the requirement is to give all ideas due weight. This generally works best of they don’t give any weight at all to who created the idea. Ideas are either good or bad, regardless of who creates them.

Similarly, the data collected by the FCC for the purpose of accelerating broadband deployment is more important as a time series than as a snapshot. Isn’t it in the nature of acceleration to be about time and speed?

The Democrat’s Letter is Misguided

The letter, signed by House Energy and Commerce Committee Ranking Member Pallone and Communications and Technology Subcommittee Ranking Member Doyle, claims the Form 477 data used by the FCC to assess the Internet buildout is faulty. It implies that the FCC’s Restoring Internet Freedom Order is invalid to the extent that it was based on the 477 data because they (may) overstate the extent of broadband competition.

There is no doubt that the 477 data are imperfect; the question is whether they’re good enough, when combined with all the other sources of data on broadband progress, to form a rational estimate. And even if we assume they aren’t, questions remain about how important the progress estimate was for the Title II repeal.

This is all a bit rich given that the Wheeler FCC raised the guideline for broadband competition from 10 to 25 Mbps. This move appeared arbitrary because it was inconsistent with FCC data on the importance of speed to applications. The only thing it really did was create the illusion that the US has considerably less broadband competition that we really do.

The Democrats are Inconsistent

In the last Internet Access Services report issued by the Wheeler FCC (based on Dec. 2014 477 data) , 89% of Americans had two or more wired broadband options at 10 Mbps. This was a disaster for the Title II program because it showed healthy progress under Title I.

But the FCC found that only 23% had two or more options at 25 Mbps, a supportive number for massive regulatory upheaval. This is based on 477 data, which were apparently good enough for assessing competition in 2016 even though they apparently weren’t a year later.

How many times have we heard Title II supporters claim that only 20% of Americans have broadband choice? I hear it a lot. But what happened to the 477 data when the most recent FCC Internet access report was issued?

The Demand for Better Data Appears to be a Stalling Tactic

The most recent FCC report on Internet access service was issued in February, 2018; it was based on 477 data from December, 2016. All of a sudden, that 23% with choice precisely doubled to 56%. So we jumped from most have no choice to most do in just two years.

What good does it do to arbitrarily fudge the definition of broadband if network services are just going to keep getting better regardless? In another year or two, we’ll have as much broadband choice 1t 25 Mbps as we had for 10 in 2014.

Congressmen Pallone and Doyle have correctly perceived that the FCC’s competition data are never going to support their program. Not only are wired networks getting faster, in two or three years, fixed and mobile 5G will add even more sunshine to the data. The average 4G network speed already exceeds Wheeler’s 25 Mbps benchmark, after all.

In Praise of Data-Free Public Policy

No matter how good the data, underlying realities of technology will always show good progress. Hence, policy makers who want government to have a larger role in broadband are best served by an absence of data.

This is the spirit of the Pallone/Doyle letter in terms of its omissions as well as its demands. Democrats weren’t complaining about the 477 data when Wheeler was ion charge, but now they’re diabolical.

Let me suggest an approach that might be more productive. Let’s stop asking ISPs to file 477 reports in favor of more sophisticated sampling. Broadband coverage in cities is unimportant because they’re saturated, so we can drop it and focus on the problem areas.

Let’s Get Serious

We probably don’t know enough about rural areas, and we should since they’re subsidized. We need a program that tells us where the opportunities are for new network builds in rural America.

There must be a way to get more sophisticated sampling in rural America. If this kind of data were available, it could very well pay for itself by encouraging new builds financed with private money. Money seeks opportunity, doesn’t it?

The trouble with 477 is that providers can only report on the areas they cover, while the real questions are about the areas they don’t. It may be that the best way to get the data we need is through the Census. It deserves some investigation even though Pallone and Doyle didn’t raise the question.