Hal Singer has discovered that total wireline broadband investment has declined 12% in the first half of 2015 compared to the first half of 2014. The net decrease was $3.3 billion across the six largest ISPs. As far as what could have caused this, the Federal Communications Commission’s Open Internet Order “is the best explanation for the capex meltdown,” Singer writes.
Despite numerous warnings from economists and other experts, the FCC confidently predicted in paragraph 40 of the Open Internet Order that “recent events have demonstrated that our rules will not disrupt capital markets or investment.”
Chairman Wheeler acknowledged that diminished investment in the network is unacceptable when the commission adopted the Open Internet Order by a partisan 3-2 vote. His statement said:
Our challenge is to achieve two equally important goals: ensure incentives for private investment in broadband infrastructure so the U.S. has world-leading networks and ensure that those networks are fast, fair, and open for all Americans. (emphasis added.)
The Open Internet Order achieves the first goal, he claimed, by “providing certainty for broadband providers and the online marketplace.” (emphasis added.)
Yet by asserting jurisdiction over interconnection for the first time and by adding a vague new catchall “general conduct” rule, the Order is a recipe for uncertainty. When asked at a February press conference to provide some examples of how the general conduct rule might be used to stop “new and novel threats” to the Internet, Wheeler admitted “we don’t really know…we don’t know where things go next…” This is not certainty.
As Singer points out, the FCC has speculated that the Open Internet rules would generate only $100 million in annual benefits for content providers compared to the reduction of investment in the network of at least $3.3 billion since last year. While the rules obviously won’t survive cost-benefit analysis, I’m not sure they will survive some preliminary questions and even get to a cost-benefit analysis stage.
The FCC has argued that the definitions of “telecommunications” and “information” services in the Telecommunications Act of 1996 are ambiguous and that “changed factual circumstances” justify the reclassification of broadband as a “telecommunications” offering that can be tightly regulated by the FCC—as if it were a dangerous monopoly subject to no competitive checks and balances whatsoever—and not as an “information” service that is otherwise subject to Federal Trade Commission supervision.
The Supreme Court agreed in the 2005 Brand X decision that there is statutory ambiguity with regard to the “last mile” connection between the customer’s computer and the broadband provider’s computer processing facilities, but the FCC has now chosen to regulate well beyond the last mile. According to Commissioner Ajit Pai,
It is not limited to the last-mile transmission service between a customer and an ISP’s point of presence. It extends into the ISP’s network all the way to “the exchange of traffic between a last-mile broadband provider and connecting networks”—a scope that necessarily extends onto the Internet’s backbone, since that’s where many networks interconnect. (citation omitted.)
As the industry trade associations and others who are appealing the order point out, all nine justices in Brand X agreed that broadband Internet access is an information service—the only disagreement concerned the last mile. If nothing else, the Open Internet Order may be extremely vulnerable to legal challenge as a result of this example of over-reaching.
With regard to the “changed factual circumstances,” the commission argues, first, that it is now possible to conceive of broadband as nothing more than a bare “connection link,” and, second, that competition for fixed broadband is insufficient.
One thing that is clear is that the statutory definitions of “telecommunications” and “information” services look to the nature of the service provided and not to the market structure. For this reason, whether the broadband market mirrors the textbook definition of “perfect competition” or not is completely irrelevant.
The Supreme Court agreed in 2005 with the commission’s original interpretation that broadband is an information service because—as the FCC argued and the Court agreed—it enables users to browse the World Wide Web, access email and Usenet groups and transfer files from file archives available on the Internet. It also facilitate access to third-party Web pages by offering consumers the ability to cache popular content on local servers and provides access to the Domain Name System (DNS) that matches Web page addresses with Internet Protocol (IP) addresses.
The commission now argues that things like DNS, caching and network security don’t count because they’re merely used for the management of a telecommunications system or service; and that things like email and web hosting shouldn’t count because “consumers are very likely to use their high-speed Internet connections to take advantage of competing services offered by third parties.” (emphasis added.) Once these exceptions are made, all that’s left—in the commission’s view—is a telecommunications service.
But is there is no third-party exception in the statutory definitions, and the petitioners who’ve brought suit in the D.C. Circuit Court of Appeals point out that the telecommunications management exception is extremely narrow. The commission clarified in 1985, they note, that the exception is limited to services that “facilitate use of the basic network without changing the nature of basic telephone service.” In the Supreme Court’s 2005 decision, the dissent similarly tried to argue that DNS doesn’t count—as the commission argues now—but the majority responded that this argument “begs the question because it assumes that Internet service is a “telecommunications system” or “service” that DNS manages…” The Supreme Court declined to resolve this issue in 2005, but perhaps it eventually will this time around.
Hal Singer’s observation that investment in broadband infrastructure is already declining is hardly surprising. Telecommunications and information services were distinguished for regulatory purposes in 1979-80 to protect regulation of communications while protecting computing from regulation. But it soon became painfully obvious that preventing the Bell System from participating in the market for information services had the unintended effect of restricting innovation in communications. The resulting bandwidth limitations in communications ultimately became a barrier to progress in computing. In 1996, Congress tried to preserve regulation of monopoly telecommunications services while competition developed for voice services, but it was determined to free advanced information services from legacy regulation. The Open Internet Order is an attempt to do exactly the opposite of what Congress intended.