The list of recent of accomplishments at the Federal Communications Commission is pretty short. Although President Obama and his FCC chief, Julius Genachowski, took office with a lot of bold talk, very little has happened in the past five years. One reason is that the FCC has gotten mired in the same partisanship that has crippled policymaking in general.
But the FCC has wasted a tremendous amount of time and energy fighting old battles that no one was willing to let go of. Should anyone really care about rules governing the concentration of broadcast ownership at a time when online media are exploding and broadcasters are losing their relevancy? And the FCC’s network neutrality rules, which are likely to be struck down by the D.C. Circuit Court of Appeals, deal with a threat that, after 20 years of the commercial internet, remains much debated but almost entirely theoretical.
A savvy insider. Things may be about to change. After his confirmation was held up for months for no particular reason, Tom Wheeler has taken over as FCC chair. A savvy Washington insider who has won the respect of most of the interest groups that shape communications policy, Wheeler is off to a fast start. And he is setting an agenda of issues that matter for the future.
The two items at the top of Wheeler’s list both deal with the way internet and wireless technologies have revolutionized communications and lest policy gasping to catch up. The one that certainly has the tech industry’s attention is the need for more spectrum to support the burgeoning use of wireless data. The other, which has been largely off the public radar but which is vitally important to the future of communications, is what to do about the expensive and increasingly obsolete public switched telephone network. Wheeler declared his interest in taking on the telephone issue in a blog post in which he announced it is time for what he calls a“Fourth Network Revolution.”
AT&T Gets the Ball Rolling. The process was actually set in motion last year when AT&T formally petitioned the FCC for permission to abandon its traditional wireline phone network. Wheeler plans to begin formal consideration of this in January. There is little doubt that the public switched telephone network (PSTN) and much of the regulatory regime that have grown up around it over more than a century are relics of an earlier technology.
There’s a widely held misconception that the phone system is analog. In fact, AT&T (the old AT&T, not the company that now uses the name) began experimenting with digitizing truck calls in the 1930s. Today, only the “local loop,” the bit of the system that connects “plain old telephone service” ((This is actually a standard industry term describing traditional residential service.)) subscribers is analog. All the links between switches that form the backbone of the system,as well as most business connections, are digital. The question is, what kind of digital?
Packets or Circuits? The internet runs on a technology called packet switching. Messages, including digital voice services such as Skype or Vonage, are broken up into short packets. Each packet finds its way to its destination independently; the packets need not follow the same path, nor need they arrive in order.. A TCP/IP network is based on what is known technically as “best effort delivery”; the network will do its bet to deliver each packet in a timely fashion, but no promises.
The PSTN works very differently. When you make a traditional call (or when your wireless call is connected to the PSTN), an SS7 switch at your central office contacts a chain of SS& switches at other central offices along the way to create a dedicated circuit for the call. This circuit, which has a bandwidth of 64 kilobits per second, is devoted exclusively to you call until you hang up. ((The circuit is not actually a physical wire. AT&T developed a technology called time division multiplexing decades ago to allow multiple calls to share a wire. And these days, the “wires” are almost all optical fibers that can handle thousands of calls.)) The legacy landline telephone system delivers very high reliability and good voice quality (though the sound fidelity is artificially degraded by frequency limits that go back decades), but makes very inefficient use of the network.
Although the number of land lines in the U.S. has dropped considerably from its peak, the decline has flattened out in recent years and there are still about 150 million lines in use (chart.) A fair number of these lines are already IP-based. If you get a land line from a cable company, AT&T Uverse, or Verizon FiOS, you already have IP phone service.
What Sort of Regulation? The biggest question facing the FCC as it considers the IP transition is what sort of regulatory regime should apply to the new system. Even most libertarians will agree that some things will still need to be regulated, such as 9-1-1 emergency services. Universal service, the idea that telephone service should be available to every American and, if necessary, at a subsidized price that even the poorest can afford, is a political reality that will not go away. There will be arguments, however, about just what sort of service is required. There have already been disputes about Verizon’ efforts to substitute wireless service for landlines in some isolated areas in the wake of hurricane Sandy.
What should the rules be for interconnections between phone systems? Should they be like the unregulated market for internet peering, or like the heavily regulated system of traditional phone interconnects? How reliable does the system have to be? The legendary “five nines” of the Bell System, 99.999% reliability, allowed for an average of only five minutes of downtime per year. No internet provider offers a service level anything like that because the cost of doing so is so high. How much reliability is enough in an era when nearly everyone can pick up a mobile phone when the landline is down?
Questions like these, and many, many others, will dominate the debate over the IP transition, and the answers will shape U.S. telecommunications policy for decades to come. It is probably the most important question the FCC will face for a long time, and it’s a good thing that Wheeler is giving it a very high priority.