Richard A. Epstein: What are internet services?

Much of the hullabaloo about the US Telecommunications Act of 1996 revolves around the preposterous claim that it introduced competition into telecommunications markets by stripping away the older system that granted a single carrier a monopoly position in each local exchange market. The Act did nothing of the sort. What it did was substitute one system of regulation for another. Out went the direct regulation of the rates that the single incumbent could charge and in came an elaborate system of rules under which the Federal Communications Commission and the state commissions could determine the charges that new entrants had to pay, either for interconnections with their own networks or for the purchase of network elements in the incumbent’s system.

The heavy burden of the new regulation is evident in the recent decision of the Ninth Circuit (for the western states) in <em>Brand X Internet Services v. FCC</em>. The case raised the daunting question of how to classify internet services under the Act. These services allow customers to hook up through broadband in order to download at great speed the full array of content available on the internet. The simple fact that the FCC and the court were duelling over this arcane act of classification shows how alive and well regulation is in the telecommunications sector. And given the rapid development of new technological possibilities, <em>Brand X</em> had to figure out where these fancy internet services belonged. One possibility was to treat them as a simple form of cable service, subject to various kinds of direct price controls under the various cable regulations. But both the FCC and the Court dismissed this possibility out of hand. Internet services are individualised and interactive; the information streams are tailored to each customer and run in both directions. Cable channels move only in one direction and offer the same content to all individuals. So far, so good.

The more difficult question of classification was whether these internet services should be regarded as telecommunications services, information services, or both. The issue arises both for broadband hook-ups that take place through phone lines and those that use cable, raising the disquieting possibility that eventually, direct competitors could be subject to different forms of regulation. Here the FCC took the sensible economic position that these services should be classified solely as information services, which had the effect of reducing the level of regulation to which they are subject. The key administrative law question that consumed the court was how much deference it should pay to the FCC’s determination given that its own prior decision had taken the view that these services were a mixture of telecommunications services for the pipeline hook-up and information services for the internet content. After much debate and some hesitation the court held that on matters of law, which this definitional struggle surely is, it could follow its own judgment, at least when it had handed down its decision prior to any FCC action. 

I think that this result is correct as a matter of principle and would indeed take the matter one step further. There is little doubt that the central question of administrative law in the US, not only in telecommunications but everywhere else, relates to the question of deference under the Supreme Court’s highly influential 1984 Chevron decision. Chevron deference seems appropriate when the FCC decides difficult questions of fact, where it acts more like a jury and less like judge. But the case seems far weaker on matters of definition and principle in the interpretation of statutes. Here the FCC has no super expertise in a task for which judges are at least as well suited as administrators.

Nor in this case do I think that the court erred in its ultimate judgment that the broadband hook-ups to the internet were a composite of both the telecommunications and information services, subject to dual regulation. After all, one would reach that result if the telecoms company supplied only connections and some independent internet service provider provided the content in question. The fusion of the two functions into a single company does not appear to change what the consumer receives. If the two functions are regulated differently when supplied by two separate companies, they can be regulated differently when supplied by only one.

The point here is of not little consequence, because what is at stake is the requirement that the supplier of the combined services act as a common carrier who is required to supply the interconnection to any and all ISPs. In and of itself that does not seem like an impossible position, except for two potent objections. The first relates to the pricing by which the new carrier obtains access to the system of the incumbent. If, as I have argued previously, the pricing rules are fatally flawed, then a sound decision on statutory construction exacerbates the difficulties with the entire system of forced sales under the 1996 Act. The net effect is to stifle innovation, because the company that innovates has to share the fruits of its innovation so long as it rivals pay it a paltry sum.

The second point exposes the rigidities of this, or perhaps, any comprehensive system of regulation. Here the basic justification for the 1996 Act was is that these re-sales and interconnections had to be allowed in order to introduce a much-needed dose of competition into the consumer market. In a deep sense that argument was time and technology bound, for it ignored the rapid decline in cost in mobile phone competition and the rise of internet telephony, both of which have undermined and continue to undermine the empirical assumption that the “last mile” of telephone is a technological barrier to competition.

A similar transformation is at work with internet services. Right now there is an extensive head-to-head competition between the cable and the telephone providers of broadband services. Under those circumstances, there seems little need to continue to push hard on the elaborate system of resale at, arguably, below market rates: competition now comes in the form of a struggle between two "unrelated" forms of communication. If this observation is correct, then it shows yet again a fundamental difficulty in the overall scheme of telecommunications regulation writ large. Regulation cannot keep pace with market configurations. The FCC sensed this when it issued its ruling, but the question before the Ninth Circuit in Brand X concerned language, not function, let alone efficiency. There is no ground to fault a court for doing correctly what it has to do. But there should be some recognition at the legislative level that political plodding only interferes with technological advance.

The writer is the James Parker Hall Distinguished Service professor of law at the University of Chicago and Peter and Kirsten Bedford Senior Fellow at the Hoover Institution.

Copyright The Financial Times Limited 2017. All rights reserved. You may share using our article tools. Please don't copy articles from FT.com and redistribute by email or post to the web.