The Blog

Congress and the Federal Communications Commission have divided all communications using wires or radio waves into two categories:  telecommunications services or information services.

Telecommunications services are defined as “basic” transmission services in which there is no “protocol conversion” or re-stating of the communication in a different technical form.  Telecommunications services are subject to “common carriage” regulation, in which no discrimination is allowed and everyone must be served on a similar basis.  In 1956, AT&T was told in a consent decree with the Department of Justice that it could provide only “common carriage” services.

Information services, on the other hand, do involve these protocol conversions, and are supposed to be “unregulated”  – left alone by regulators save for the imposition of particular “social policies” that are deemed to be worth the regulatory candle.

The idea is that everything fits within one or another of these categories – regulated or unregulated – but not both.  The theory is that monopoly-provided basic telecommunications services should be subject to common carriage obligations, and one of the ideas behind the AT&T breakup in the 1970s and 1980s was that regulated common carriage providers (the local telephone companies) should not be allowed to cross-subsidize unregulated information services businesses.  Similarly, the FCC and Congress take the position that businesses providing unregulated information services should not be subject to common carriage obligations.   At bottom, the FCC takes the view that the regulator’s job is to decide how a particular form of communication should be categorized.  From that categorization everything else flows.

Network operators providing access to the Internet through coaxial cables, fiber-optic lines, and wireless transmissions have persuaded the FCC and the Supreme Court that everything they do falls into the “information services” category.  There is very little common carriage “telecommunications services” provision going on any more.

At this point, the division between “telecommunications services” and “information services” seems like an immutable principle.  It seems inconceivable that we could abandon our commitment to this American understanding:  telecommunications services on one side, information services on the other, and a hermetic seal between the two.

Yet this fundamental slicing of the world is (or should be) difficult to maintain.  Application of the words “information service” to highspeed internet access (which certainly feels just like basic transport to everyone other than communications lobbyists) is extraordinarily difficult. Even though the Supreme Court managed to defer to the Commission’s interpretation on this point, its opinion was easily parodied by Justice Scalia in dissent.  The complete absence of basic common carriage telecommunications services has unmoored the notion of “information services” – rather than “information services” being something made possible by basic transmission, “information services” are now everything.

Although the telecommunications/information service dichotomy is a basic building block of communications law, the details of the categorizations that are currently carried out using these words make little sense.  The reason the “information services” category is incoherent (particularly when it floats alone, unaccompanied by any notion of underlying common carriage) is that we have forgotten where it came from.
The origins of “information services” are found in the pre-Internet era, when we were concerned that telephone companies would use their bottleneck control over communications services to control access to early computing services such as data processing carried out on enormous mainframes. We assumed the presence of common carriage communications, and we were worried about giving those carriers the power to choke off these new computing services.  The idea of “information services” was designed to protect the computing industry from the depredations of the carriers.

Since the 1950s, the computing industry has morphed into the human-computer-interaction that we know as Internet applications.  Meanwhile, the notion of “information services” has been pressed into duty by the carriers themselves to define what it is they do so that they can protect themselves from regulation.  But this idea was never designed to protect the carriers, nor even to serve as a definition of a particular form of transport. It was designed to protect computing.

The consequences of the deracination of “information services” are profound. Now that all network operators can style themselves as providing “information services,” there is no legal constraint on their ability to discriminate against particular uses of their networks.  Yet this discretion to discriminate is exactly what the original thinking behind the “information services” category sought to avoid.  Also, loss of the idea of common carriage that is behind the “communications services” category (now itself a kind of lost positive) has been destructive to communications provision and even, perhaps, to economic growth in the U.S.