Commissioner Pai Remarks Before the Internet Innovation Alliance
REMARKS OF FCC COMMISSIONER AJIT PAI
BEFORE THE INTERNET INNOVATION ALLIANCE:
“THE IP TRANSITION: GREAT EXPECTATIONS OR BLEAK HOUSE?”
JULY 24, 2014
In debates about economic policy, there has long been tension between those whose primary
concern is redistribution and those whose priority is growth. Or, to put it another way, some focus on
how to divide the pie that we already have while others are more interested in making the pie larger.
We are now witnessing a similar phenomenon in the field of communications policy. Internet
regulation has been the talk of the town recently. You know the buzzwords: Title II, Section 706, net
neutrality, open Internet, paid prioritization, two-sided market, slow lanes, commercially reasonable. I’m
surprised that no one has created a Net Neutrality Mad Libs game yet. So before today’s event, I asked
one of my staffers to name a verb with apocalyptic overtones, a network management practice, and a
provision of the Communications Act. That generated the following sentence: The FCC must eliminate
usage-based pricing using Section 215(b) of the Communications Act or else the Internet as we know it
Not to put too fine a point on it, as Charles Dickens first said—but this whole debate is a
distraction. Instead of repeating the seemingly never-ending arguments about how to manage (or,
depending on your perspective, micromanage) the networks that we currently have, we should prioritize
policies that will encourage the private sector to expand and upgrade high-speed broadband networks.
And that means that we need to concentrate on expediting the Internet Protocol (IP) Transition.
When it comes to the status of the IP Transition, I’m reminded of A Tale of Two Cities: “It was
the best of times, it was the worst of times.” In the private sector, the IP Transition is proceeding apace.
Packet switching has usurped circuit switching. Carriers are pushing fiber further into their networks and
upgrading from DSL to IP-based technologies like carrier-grade Ethernet. Mobile providers have almost
nationwide 4G footprints, each vying to improve upon LTE’s baseline for greater speed and resiliency.
Cable operators are on DOCSIS version 3.1—a standard promising speeds of up to 10 Gbps.
And the usual suspects aren’t the only ones investing. Small entrepreneurs have built wireless
ISPs from the ground up, using otherwise fallow spectrum to reach consumers in rural America. Satellite
companies have followed suit, offering 12 Mbps service to anyone with a clear view of the southern sky.
Electric utilities are beginning to invest in fiber projects that span their footprints. And online companies
are exploring new ways to get their content to consumers, reportedly even bouncing signals off high-
In short, we’re in the midst of a competitive explosion in the broadband marketplace with 1,712
broadband providers investing in the United States. That’s approximately 1,710 more providers than
there were 40 years ago when Ma Bell was a legal monopoly and ARPANET a government project.
The result is that high-speed Internet access is more universal than ever. In 2012, wireless
broadband was available to 99.8 percent of Americans and fixed broadband available to 96 percent of
Americans. Subscription rates are jumping as well. In the first half of 2013, Americans signed up for an
additional 34 million broadband connections—bringing the total to 163 million, or about one and a half
per household. And that’s only for connections that meet the FCC’s definition of broadband. If you
include connections good enough for email, web browsing, or VoIP, the number soars to 276 million.
The fact that the marketplace is embracing the IP Transition means we’re in the best of times.
Unfortunately, I can’t say the same about the public sector. The government seems to be stuck in neutral
when it comes to modernizing its regulatory framework to account for the IP Transition. And it’s unclear
whether its next step will be to drive forward or to put the IP Transition into reverse.
What’s sidetracked us? Our efforts have been plagued by three critical mistakes. To continue the
Dickens theme, the first involves the past, the second involves the present, and the third involves the
future. All three mistakes are coming together in the Title II debate. But before getting to Title II, let me
discuss each of these errors.
First, we are bogged down in brouhahas from the past. The FCC obviously has plenty to do
when it comes to preparing for the all-IP future. So we don’t have time to waste on yesterday’s battles.
But that’s precisely what we’re doing. Consider the FCC’s review of the enterprise data services market,
better known as the special access proceeding.
Almost two years ago, the Commission suspended its deregulatory special access rules, promising
that it would reach “final conclusions on the need for overall reform of the special access marketplace . . .
in 2013.” Back then, I said that for this to be true, the Office of Management and Budget (OMB) would
need to approve our special access data collection by March 2013. It’s sixteen months later, and we are
still waiting for that approval. I’m not surprised; as I said at the time, the proffered “timeframe [was] just
Part of the problem, no doubt, is the breathtaking scope of the endeavor. The Commission sought
comment on reregulating the special access market in every conceivable way. It put on the table such
things as regulating prices on a location-by-location basis or even mandating different prices on different
floors of the same building. To justify these types of intrusive regulations, it teed up a one-time, complex
market analysis that would require “panel regressions.” These regressions, in turn, would incorporate at
least thirteen different factors. And to collect the data for this analysis, the Commission proposed a
nationwide data collection—seeking information about every cell tower, office building, school, college
campus, factory, farm, and other enterprise facility in the country.
The numbers prove how daunting this task will be. According to OMB, this one-time data
collection will require 6,400 responses, with respondents estimated to take 134 hours per response.
That’s 934,400 burden-hours in total, which translates into tens of millions of dollars in administrative
costs—again, for a one-time data collection. And that’s after the Wireline Competition Bureau purported
to streamline the collection last year! It all reminds me of Dickens’ aphorism that “The one great
principle of the English law is to make business for itself.”
Assuming that a data collection is eventually approved by OMB, how much time will
Commission staff spend analyzing it? How much time will it take to formulate a proposed framework for
reregulating the special access market? I can’t quantify it, but I do feel confident in two answers: a lot
and too much.
We should step back and ask what the point of all this is. After all, the most purchased and
popular special access circuit is the T1 line, which offers subscribers just 1.5 Mbps. In other words, we
have spent countless hours debating whether to suspend our rules, what data to collect, how to analyze
that data, and whether we should reregulate the market . . . all for a product that does not even meet the
FCC’s definition of broadband. It took us six months to agree to set aside $75 million to see whether
providers will offer 25 Mbps service in rural America—and we’re still mulling a data collection that will
cost industry almost as much in order to study services 16 times slower.
The special access proceeding teaches us that we can’t become mired in the disputes of the past.
If we want to facilitate broadband deployment and competition, then we should focus on facilitating
broadband deployment and competition. Collecting every last bit of data on last-generation services for
the purpose of reregulating that market—that shouldn’t even be on the agenda.
Second, we are failing to seize the present. When it comes to embracing the IP Transition, we
can’t keep putting off until the unspecified future the things we can and should do today, starting with
modernizing our rules.
The Code of Federal Regulations remains replete with legacy regulations. Many of them are as
tedious as they are pointless in the competitive, digital age. For instance, Part 61 contains 39 pages of
tariffing rules. Those rules specify how carriers must calculate tariffs using CCLs, EUCLs, APIs, and
CMTs. They outline how those calculations might differ for interLATA or intraLATA services. They
even dictate the placement of a tariff’s effective date in the lower right-hand corner of each page, as well
as other minutiae. Now, way back in 1842, Dickens in real life wrote, “I am quite serious when I say that
I do not believe there are, on the whole earth besides, so many intensified bores as in these United States.”
Can you imagine what he’d say if he had read Part 61?
We have the tools at our disposal to uproot these outdated regulations. But for some reason, we
haven’t been willing to use them. It reminds me of President Lincoln’s famous comment on General
McClellan during the Civil War: “If McClellan does not want to use the Army, I would like to borrow it
for a time.”
Consider the FCC’s reluctance to exercise its forbearance authority under section 10 of the
Communications Act. This statutory innovation gave the FCC much-needed flexibility to modernize its
rules. But the FCC has been stingy about using it. And in recent years, it has actually made it harder, not
easier, to grant forbearance from economic regulations.
In the Qwest-Phoenix Order, for instance, the Commission essentially presumed that all existing
regulations are necessary and in the public interest, and it imposed high evidentiary hurdles on any party
seeking relief. Perhaps unsurprisingly, the Commission denied Qwest relief despite that fact that Qwest
faced direct competition from cable operators and over-the-top VoIP providers, along with wireless
substitution. The Qwest decision rejects that wise Dickensian dictum: “No one is useless in this world
who lightens the burden of it to anyone else.” I have no doubt that Charles Dickens had in mind a
forward-thinking view of forbearance.
Unfortunately, it appears that the FCC is not going to make itself useful in this regard anytime
soon. Earlier this year, CenturyLink petitioned for forbearance from dominant carrier regulation and
tariffing for its enterprise broadband services. This is the same relief that the FCC granted Qwest (now a
CenturyLink subsidiary) just a few years ago.
The logic of the CenturyLink petition is straightforward: Neither Verizon nor AT&T nor
Comcast nor Time Warner Cable nor Google nor Level 3 nor XO nor even CenturyLink within its Qwest
territories has to comply with these legacy regulations, so why should CenturyLink be treated any
differently in its non-Qwest territories? To ask the question should be to answer it.
Nevertheless, the FCC is gearing up for a data request so that it can graft a Qwest-Phoenix market
analysis onto CenturyLink’s petition. When you hear folks complain that the enterprise broadband
market isn’t robust enough, point them to this proceeding, which makes it harder for CenturyLink and
others to compete.
There are also other areas critical to the IP Transition that are starving for attention but, like
Oliver Twist, may be left begging “Please, FCC, I want some more.”
One of them is numbering. Numbering may be arcane, but it’s important. Our telephone
numbering system needs to end its dependence on the public-switched telephone network (PSTN). Last
year, we commenced a rulemaking to do just that. We proposed to review our cost allocation rules for
numbering. We proposed to revise our numbering databases so that they would be ready for the all-IP
future. And most importantly, we proposed to end our decades-old prohibition on VoIP providers getting
direct access to numbers.
We even beta-tested the direct assignment of numbers to VoIP providers to see if any technical
issues would arise. None did. Five VoIP providers participated in the six-month trial, and the Wireline
Competition Bureau reported an unambiguous verdict: “[I]t is technically feasible for interconnected
VoIP providers to obtain telephone numbers directly from the numbering administrators.” In other
words, the tests could “not identify technical problems” to solve.
That means we’re ready to complete this rulemaking.
But have we? Not yet. Will we? It’s
Another critical priority should be to make consumers aware that the IP Transition means better
services. High-quality voice is a good example. Few people realize that the old PSTN is hardwired to
capture and transmit only a fraction of the frequencies used by the human voice. HD Voice captures a
much broader range of frequencies and incorporates new technologies like noise-cancelling dual
microphones so that callers can have a conversation even when they are standing beside a freeway. I tried
it out for myself at the FCC, courtesy of my friend Daniel Berninger, and I’m telling you, the difference
was remarkable. Plenty of others will, too: Every improvement in voice quality will make it easier for
hard-of-hearing Americans to converse without using a relay service, for example.
Now, some VoIP providers and wireless companies have begun rolling out HD Voice already, but
the FCC seems to be slow to recognize this progress. We can remedy that delay by issuing a Notice of
Inquiry on how to facilitate the nationwide transition to HD Voice. That’s important for all consumers—
God bless us, every one.
Speaking of A Christmas Carol, third, we are becoming crippled by fear of the future.
Embracing creative destruction means that something must be abandoned—and that something is the
PSTN. Over 7 million households, or about one seventh of households that still had a copper landline,
dropped it over the last year. Almost 40 percent of U.S. households have dropped landlines entirely, and
only about a third are still on the PSTN.
This should be celebrated. For years, we have known that fiber is the fastest, most reliable way to
transport data over long distances. What makes up the Internet backbone? Fiber. When companies
invest millions to trench an undersea cable linking the continents, what do they use? Fiber. What’s the
best way to improve a customer’s broadband Internet experience? Dig up the copper and deploy fiber to
the node, to the curb, or to the home.
And yet, the loss of copper is causing consternation for some. Indeed, the fading copper field—
see what I did there?—is bemoaned in much the same way as some film aficionados still lament the
advent of color.
Alarmingly, there’s a pending proceeding at the FCC that could require the maintenance of
copper networks in perpetuity, or at least so long as a tiny subset of customers wants to use them at
subsidized rates. And I’ve read about state commissions examining their copper retirement policies on
the notion that replacing copper with fiber somehow harms the consumer.
Here’s the simple truth: Fiber is better than copper. Twisted copper pairs are fragile. They
corrode. They fail in adverse weather. They transmit information electrically, which increases their
energy demands, limits the length of copper loops, and reduces the maximum speed of transmission.
Fiber, by contrast, is resilient. I’ve seen numbers that suggest fiber networks fail at only one-
eighth the rate of copper networks. And fiber transmits data at the speed of light with reliability unheard
of in the copper world. The mayor of Hummelstown, Pennsylvania—a 4,300-strong hamlet near
Harrisburg which will soon become the state’s first “all-fiber” town—knows all this. As he put it, “This
new technological advancement helps Hummelstown position itself better for the future.”
The FCC knows all this, too. In fact, it’s known it for more than a decade. That’s why our rules
aim to spur the deployment of fiber.
But here’s the thing. Running two networks is expensive. You cannot induce carriers to
maximize fiber deployment if they have to maintain copper. And so the FCC’s copper retirement rules
are a necessary component of our fiber-deployment strategy.
Under those rules, carriers need not maintain their copper networks once they’ve been replaced
with fiber and once the carrier has notified customers about the reconfigured networks. It’s this
regulatory certainty—the knowledge that investing in fiber means no more maintenance costs for aging
copper infrastructure—that has spurred fiber deployment throughout the United States.
Any signal that we will backtrack on our copper retirement rules will harm broadband
investment. We can’t let fear of the future deter investment in the networks of the future. For if we do,
we risk becoming the Circumlocution Office. Not familiar with this Dickensian bureaucracy? With your
indulgence, and from Little Dorrit:
The Circumlocution Office was . . . the most important Department
under Government. No public business of any kind could possibly be
done at any time without the acquiescence of the Circumlocution
Office. . . . It was equally impossible to do the plainest right and to undo
the plainest wrong without the express authority of the Circumlocution
Office. . . . Whatever was required to be done, the Circumlocution Office
[worked] beforehand with all the public departments in the art of
perceiving—HOW NOT TO DO IT.
* * *
So, moving back to Title II, how does the ongoing debate about Title II regulation exemplify each
of the mistakes that I’ve outlined above?
First, we are getting bogged down in a debate of the past. Indeed, it is a debate that harkens back
to nineteenth-century railroad regulation. This debate is no fresher now than it was when the Supreme
Court decided Interstate Commerce Commission v. Baltimore & Ohio Railroad Co., 145 U.S. 263 (1892).
The only difference is that we now know, or should, how common-carriage regulation can take us off
Second, we are failing to seize the present. Countless staff hours, meetings, and phone calls are
being devoted to net neutrality. That’s effort that isn’t being devoted toward modernizing many of our
rules to expedite the IP Transition.
And third, we are becoming crippled by fear of the future. In particular, the Title II discussion is
being driven by a parade of horribles that is entirely hypothetical. But, as Dickens put it in Great
Expectations, “Take nothing on its looks; take everything on evidence. There’s no better rule.” And the
evidence in the Internet marketplace does not justify dramatic regulatory change solely for the purpose of
assuaging fears that have not materialized.
The irony is we already know that the public-utility model, so beloved by Title II proponents,
would deter innovation and drain private investment in broadband. The European experience is our
benchmark. Europe chose to regulate broadband operators like public utilities; we did not. Today, only
54 percent of Europeans have access to 25 Mbps broadband speeds and per-household broadband
investment is $244. In America, those numbers are 82 percent and $562, respectively. The European
Commission itself has recognized that “Europe is losing the global race to build fast fixed broadband
connections.” But if we embrace Title II, our future would look a lot like Europe’s present.
In sum, it is as true in the digital age as it was in Dickens’ time that we should “reflect upon [our]
present blessings—of which every man has many—not on [our] past misfortunes, of which all men have
some.” The U.S. broadband market is a success story worth celebrating. And that success was enabled
by light-touch regulation. Where there are concrete “misfortunes,” we can take targeted action. But
otherwise, we should recognize net neutrality and the Title II debate for what it is—a great distraction.
And we should seize the moment to embrace the IP Transition and our all-IP future.
Note: We are currently transitioning our documents into web compatible formats for easier reading. We have done our best to supply this content to you in a presentable form, but there may be some formatting issues while we improve the technology. The original version of the document is available as a PDF, Word Document, or as plain text.