Net Neutrality Is a Bad Idea That's Run Its Course

When the FCC isn't protecting us from bad language, it concerns itself with markets created by and for communication networks. It allocates the airwaves used by old-school television, talk radio, mobile phones and Wi-Fi; it oversees mergers and acquisitions among communications companies such as NBC Universal, Comcast, AT&T, and Sprint; and in the current century it has expended considerable resources on micro-managing the technical operations and business models of broadband Internet Service Providers.

While the agency would seem to be plenty busy carrying out its statutory responsibilities with respect to spectrum and mergers, it has chosen to become embroiled in an extra-curricular affair of its own making, the "net neutrality" controversy. This kerfuffle dates back to philosophical meditations on regulation and innovation before the turn of the current century.

It got real in 2007 when self-styled public interest groups filed a complaint with the FCC alleging that Comcast was picking on piracy-oriented BitTorrent networks to protect its TV business. Although Comcast was actually protecting voice competitor Vonage, it stopped using the offending system as soon as it had a higher-quality alternative. The FCC rapped Comcast's knuckles anyway, which led the company to give the FCC a shellacking in court. This in turn caused the agency to devise a new set of Internet regulations in 2010, only to have them vacated by the court this January.

Somewhere along the way, most net neutrality wonks stopped caring whether it was good policy for innovators or even what the term means: now it's all about winning. The FCC has decided it can't passively accept the status quo and has issued 100 pages of questions on various approaches it might take to satisfy the warring clans in the Internet economy's Game of Thrones, none of which is broadly popular.

At the heart of the conflict lies a misconception about how the Internet works; this naturally leads to a series of counter-productive prescriptions. The very first of the net neutrality papers, "Network Neutrality, Broadband Discrimination" by then-Virginia law professor Tim Wu, imagined a magical Black Box connecting ISPs with the Internet. Wu realized that the Internet is rife with "discrimination", content and services offered at various prices with widely divergent levels of quality and utility. He also recognized that neighborhood broadband networks do a number of different things that depend on discrimination: in addition to connecting to the Internet, they supply cable TV and furnish telephone service.

Wu feared ISPs had incentives to degrade competitors, especially video and voice services that went toe-to-toe with core elements of their business model. So he took the unusual step of granting an effective monopoly to the ISPs for voice and video by making the Black Box favor web surfing over other uses. Wu may have sought to design a system that would make ISPs structurally incapable of bad behavior, but he ended up favoring the Web over emerging Internet applications. Banishing the devils has a way of eliminating the angels as well.

Given that it's committed to making new rules for the Internet, the FCC has a choice between basing its authority on a terse direction in the law allowing it to promote investment in advanced networks (Section 706) or on another portion pertaining to the traditional telephone network, Title II of the Telecom Act. In either case, the agency seems convinced that the Black Box is a winner, at least at the ballot box.

Networks that can't discriminate are incapable of supporting the wide range of uses that more agile networks can handle. A Black Box network must necessarily be tuned to a single, dominant application instead of being responsive to a diverse pool of uses. Whatever else we know about the future, it's certain that the Internet will be expected to do more things for more people ten years from now as it was ten years ago.

If we're going to have a robust and growing market for network applications in the future that improve quality of life and grow the economy, we're going to need networks that can move information quickly or cheaply, reliably or pervasively, securely or accurately and in several other modes as well.

Consequently, the locus of concern for regulatory policy must shift from preventing the bad to promoting the good. The FCC can do this by drafting rules consistent with the desire to promote meaningful competition, network investment, and service diversity.

Most of the content we get from the Internet comes to us through a kind of fast lane known as a Content Delivery Network that accelerates our access by placing duplicate copies of the content around the web. It's a law of engineering that short distances can always be crossed more quickly than long ones. It's also the case that sensitivity to the fundamental elements of network service quality - information loss and delay - depends on the application in use. Backing up a hard drive is less time sensitive, more loss averse, and more data volume-intensive than making a phone call. Network systems such as Wi-Fi, Ethernet, and 4G/LTE wireless recognize this fact with built-in mechanisms to match network service to application needs.

The Black Box rules these adaptations out of bounds, effectively forcing applications to adapt to the whims of policy makers and an arbitrary network. This approach compromises innovation and economic growth, and ultimately erodes quality of life.

The business practices of network industries need the same sort of anti-trust scrutiny that every industry faces, but they do not need precautionary prescriptions that throw the baby out with the bath water. Twenty years of experience with the commercial Internet has proved that fast-lane services like CDNs are beneficial, so we should be looking for ways to grow the Internet economy by creating more services like them.

Network neutrality is simply a bad idea that has run its course.