Net Neutrality Tail Wags Broadband Dog

by on March 11, 2010 · 6 comments

I published an opinion piece today for CNET arguing against recent calls to reclassify broadband Internet as a “telecommunications service” under Title II of the Communications Act.

The push to do so comes as supporters of the FCC’s proposed Net Neutrality rules fear that the agency’s authority to adopt them under its so-called “ancillary jurisdiction” won’t fly in the courts.  In January, the U.S. Court of Appeals for the D.C. Circuit heard arguments in Comcast’s appeal of sanctions levied against the cable company for violations of the neutrality principles (not yet adopted under a formal rulemaking).  The three-judge panel expressed considerable doubt about the FCC’s jurisdiction in issuing the sanctions during oral arguments.  Only the published opinion (forthcoming) will matter, of course, but anxiety is growing.

Solving the Net Neutrality jurisdiction problem with a return to Title II regulation is a staggeringly bad idea, and a counter-productive one at that.  My article describes the parallel developments in “telecommunications services” and the largely unregulated “information services” (aka Title I) since the 1996 Communications Act, making the point that life for consumers has been far more exciting—and has generated far more wealth–under the latter than the former.

Under Title I, in short, we’ve had the Internet revolution.  Under Title II, we’ve had the decline and fall of basic wireline phone service, boom and bust in the arbitraging competitive local exchange market, massive fraud in the bloated e-Rate program, and the continued corruption of local licensing authorities holding applications hostage for legal and illegal bribes.

But the FCC has not ruled out the idea of reclassification.  Indeed, just as the piece was being published, FCC Chairman Julius Genachowski was testifying before a Senate committee considering Comcast’s proposed merger with NBC Universal.  When asked whether the agency was considering reclassification, the Chairman responded:  “We are defending the position that Title I gives us the authority we need. We’ll continue to assert that position and hope we will get a favorable decision. If the court does something that requires us to reassess, we’ll do that.”

I leave for another day a detailed discussion of whether the FCC could in fact reclassify broadband Internet as a telecommunications service without explicit authority to do so from Congress.  It was the Commission, after all, who argued successfully in the Brand X case that broadband Internet clearly fit the definition of information service.  (I criticized the Brand X case when it was decided as not going far enough, see “Cure for the Common Carrier,” CIO Insight, April 2005)

Under the Chevron Doctrine, the U.S. Supreme Court gives great deference to agencies in the interpretation of their governing statutes.  Brand X held that the two definitions were ambiguous and that the FCC’s resolution of that ambiguity was reasonable.  Under Chevron, that ends the involvement of the courts.  To reclassify broadband, the FCC would have to argue that its interpretation was not in fact reasonable.

Left out for reasons of length was a discussion of the history of the two terms and the source of the definitions given for them in the 1996 Act.  That history demonstrates even more clearly, I think, that regulation of the Internet cannot and should not be governed by Title II.

Title II telecommunications services are subject to the “common carrier” provisions—including unbundling, rate oversight, and the Universal Service Fund—that were created over the years to manage the legal monopoly held by AT&T.  Under the 1913 Kingsbury Agreement, AT&T was able to maintain its control over most aspects of U.S. telecommunications in exchange for accepting government oversight.  In 1934 Congress created the Federal Communications Commission, which addressed itself to regulating AT&T’s interstate business.  State and local authorities handled the intrastate business.

Over the last thirty years, the AT&T monopoly has been largely broken up by a combination of disruptive technologies (alternate transport including cable, satellite, and cellular, and new architectures including the Internet) and government interventions.  The equipment monopoly of AT&T’s Western Electric subsidiary went first, followed by the separation of local and long distance in 1984.  Competition in long distance was followed in 1996 by the forced opening of local services.  What was left of the original AT&T was acquired by some of its former subsidiaries in 2005.

Meanwhile, since the 1950’s the computer revolution has been busily transforming business and life.  In the early days of standalone computers, there was no overlap between computing and telephony, but with the advent of time-sharing and later interactive and distributed computing, private data communications began to develop using much of the infrastructure invented for voice.  A simple—and I think, entirely reasonable—way to distinguish between “information services” and “telecommunications services” is to understand that historically “information services” meant private data communications and “telecommunications services” meant voice.

For the most part, information services have remained outside of the FCC’s regulatory clutches for the simple reason that AT&T, until 1980, was banned from offering data communications.   (It’s a long story, but that ban is the reason Linux—derived from AT&T’s UNIX operating system—is open source.)  Since the regulated monopoly wasn’t involved in the computer or data communications business, there was no basis for subjecting it to common carrier rules.  That’s a fortunate accident of history, of course, because leaving computing unregulated has meant all the difference in its dramatic growth.

The legal border between Title I and Title II goes back at least to 1956 and an earlier attempt to break up AT&T.  The effort failed, but a Faustian bargain was struck.  AT&T and its subsidiaries (including Western Electric and Bell Labs) agreed to stay out of the computer business (“information” or “enhanced services” as it was called at the time) and remained under the regulatory control of the FCC for traditional telephony (“telecommunications services”).  In exchange, the government let AT&T stay together.

At the time, commercial computing was in its infancy.  There was no data communications and certainly no Internet.

In 1980, AT&T was released from its pledge not to offer data services.  But four years later, before they could really do very much with their freedom, the company was broken up by Judge Harold Greene, who then ran the communications industry from his chambers until Congress finally passed the 1996 Act.

In any case, by the 1980’s the computer industry had evolved dramatically.  The provisioning of private data services was a fast-growing business.  IBM, DEC, and other providers had developed proprietary networks and proprietary standards, and used them to lock customers into their hardware and software products.

All of that had changed by 1996.  For the first time, it was not only businesses but consumers who were using information services.  Public data networks operated by America On-Line, CompuServe and Prodigy had millions of subscribers and were growing rapidly.  Netscape Navigator unleashed the full potential of the World Wide Web as a non-proprietary networking standard, creating new industries and services which continue to evolve at a staggering pace.  The protocols that made up the Internet had shifted from a government and university use to public use.

Given this history and the great uncertainty in 1996 as to where commercial and consumer computing was headed, Congress decided that the fast-growing provision of data communications should remain outside the control of the FCC.  Hence “information services” remained unregulated and “telecommunications services” remained regulated.  The terms and their definitions in the 1996 Act were lifted, for better and for worse, straight out of the 1984 decree in the AT&T antitrust case.

(Much of this history is recounted in detail by Prof. Susan Crawford in a 2009 article in the Boston University Law Review, “Transporting Communications.”   Though I don’t share Prof. Crawford’s conclusions, the story is told quite well.)

One could argue (I have) that the need for Title II regulations even as applied to wireline voice communications (what’s known in the industry as POTS – Plain Old Telephone Service) makes little sense now that nearly everything has converged under the Internet—including voice.  Or put another way, the continued application of common carrier rules are introducing more costs to consumers than any inefficiencies that might exist in an unregulated world.

But it’s much harder to make the case that we should actually move data communications over to Title II in addition to POTS, especially if the only reason behind doing so is to salvage the Net Neutrality rulemaking.  To mix some metaphors, that’s both the tail wagging the dog and killing goose that lays the golden eggs.  It also flies in the face of the history of the two titles, the consumer experience of life under Title I, and common sense.

  • john b.

    You mention that “Public data networks operated by America On-Line, CompuServe and Prodigy had millions of subscribers and were growing rapidly.” Do you think their growth would have been different, had they had to negotiate with the phone company for access to the phone wires? After all, dialup ISPs all communicated with their customers via Title II pipes.

    I agree with you that computers and the Internet took off in part because of unregulated innovation–but I agree with Crawford that a predicate for that was access to nondiscriminatory general purpose communications networks. I doubt that the consumer Internet would take the form it has today, had it not been for access to Title II data transport in its infancy. My first ISP certainly had to pay Bell Atlantic dearly for its lines, but it (a free BBS with for-pay shell accounts) would not have been in a position to outbid the big guns, or compete with an in-house offering by the phone company.

    I was disappointed with your description of Brand X in the CNet article; I think you characterize it more fairly here–finding a decision within an agency's discretion is very different from “endorsing” the agency's decision. However, under the Chevron framework, an administrative agency doesn't have to find its past course of action unreasonable; it merely needs to give reasons for adopting the new course of action. It can be the case that several mutually-exclusive courses of action are all “reasonable,” and, like it or not, “we can regulate one way and not the other” probably counts as a valid reason for picking one course over the other.

  • georgeou

    Excellent article Larry.

  • http://www.facebook.com/RichardBennett.Cal Richard Bennett

    The Internet was already a world-wide success before it was opened to the general public, john b., so the Title II status of POTS was neither her nor there with respect to its acceptance by John Q. Public. Without Title II POTS, DOCSIS and DSL would have been accepted sooner.

  • http://bennett.com/blog Richard Bennett

    The Internet was already a world-wide success before it was opened to the general public, john b., so the Title II status of POTS was neither her nor there with respect to its acceptance by John Q. Public. Without Title II POTS, DOCSIS and DSL would have been accepted sooner.

  • Pingback: TechCrunch TV on “Net Neutrality”

  • Pingback: Net Neutrality and the Inconvenient Constitution | Larry Downes

Previous post:

Next post: