On the plane to Mexico I had a chance to catch up on some reading, including Susan Crawford’s most recent paper, The Internet and the Project of Communications Law. ~50 pages, but excellent!, well worth the time.
Part one is a comprehensive history of Internet related communications law and scholarship, with footnotes that include all the interesting papers and commentary I’m aware of, plus many more I now need to pursue.
In part two she argues that current communications law theory and discussions fail to capture what is really valuable about the Internet. To explain that real value, she first provides a tutorial on Paul Romer and economic growth theory, thus hitting on another interest of mine. :-)
Romer gets credit for bringing ideas into mainstream economic models. Historically, economists focused on physical goods, leaving ideas as exogenous factors, i.e. fudge factors outside their models. But ideas are the root of economic growth and need to be incorporated into economic models. Importantly, ideas are fundamentally different from physical goods, as ideas can be used by multiple people without reducing their value. It’s the difference between bits and atoms. Bits can be copied, almost for free while atoms can only be moved around, combined and separated, bought and sold. Yet the real value of my laptop is the information on its hard drive (the ideas, held as bits), not the physical device itself.
Susan’s central point is
given that economic growth is created by the emergence of new ideas, the proper role of government should be to support the diversity of complex social interactions online. Scholars who argue about wealth effects on telecommunications providers of particular regimes are focused on a small subset of the story. It is the large and diverse online world of interactions and ideas out there that matters …
What matters are communications themselves and the increasingly diverse and valuable ideas they produce.
I love this. In my view, the value of the Internet is in (eventually) allowing connections between the 6.5 billion minds on the planet. In this, the Internet is the fourth notable breakthrough in the history of mankind — the earlier ones being speech, writing and printing. But I’ve only bandied around grand statements without articulating the justification for those statements. Thank you Professor Crawford!
Where my views diverge, at least a bit, is in some of her “implications for policy.”
Universal Service. We agree that universal service fund (USF) in the US is a completely broken program that is pursuing the wrong goal (telephone service rather than Internet connectivity). But I can’t agree that replacing today’s USF with an Internet focused USF would work.
- Globally, there are many different USF efforts attempting to bring telephone service to remote areas. With the possible exception of Uganda, they are utter failures that are delaying the advent of telephone service, not fostering it. Without understanding these failures it’s hard to see using USF ideas to foster Internet connectivity.
- USF includes the word “service” but what we seek is connectivity. I could see a government program to subsidize rural co-ops purchasing and operating infrastructure that connects them to a legitimate Internet exchange (one where 3 or more Internet “service” providers are present). But any program where government money goes for the purchase of services is just a gift to the incumbents.
Divestiture. We agree on the need, but perhaps not on exactly what is to be divested. In the end, I don’t want a government run, or regulated, IP service. IP (and the routers and electronics it’s built upon) is too complex and too rapidly evolving for any government program. To explain my rational, it’s necessary to back up a few steps… Consider the service life of various elements.
- Dark fiber (and dark copper and dark coax), 20+ years. Poles and conduits, 20–50 years. Electromagnetic spectrum, decades to indefinite. Rights-of-way, centuries to indefinite.
- Electronics to light fiber, modems to power copper or coax, and radios to light portions of the spectrum — all become obsolete within 2–5 years.
Now consider the life of telecom legislation.
- Roughly 15 years between major telecom laws
- Similar gaps between major FCC rulings (Computer I, II, …).
- Significant course change take longer — Ronald Coase and Leo Herzel proposed spectrum auctions in 1959 leading to the first auction in 1994 (35 years).
The fundamental reason we’re having any of this discussion is that there’s limited right-of-way for first mile access. Given the useful life of conduits and poles and of dark fiber (or dark copper or dark coax), it’s practical to manage these via government processes. But government involvement in anything above the dark fiber should be viewed as tactical, not strategic. I spoke of this at Spring VON a year ago.
Stockholm is an example of government processes working extremely well. The government has funded a company, Stokab AB, to install dark fiber. Stokab is restricted from offering any services above dark fiber, they are allowed to use long term (20 year?) accounting, and they have to break even. Within those constraints they endevour to install at much fiber as possible and offer that fiber equally to all comers. The result has been vibrant competition above the dark fiber layer. No, Stockholm does not yet have universal, very high speed (100 Mbps or more) access, but they are so far ahead of New York City that there is no comparison.
I don’t have pat answers, but there are plenty of international examples to study. Bill St. Arnaud has published extensively, runs an email list and blogs on dark fiber. When he ran Stokab, Anders Comstedt spoke widely (ppt) on the subject.
Network Neutrality. Here we have less agreement. I’d really like to see a national strategy to get as much of the population on dark fiber as possible. If that fiber is user-owned (perhaps through condominiums) or owned by local governments or co-ops, but there is competition above, then network neutrality is moot. In places where the long term strategy hasn’t been or can’t be realized, then some concept of network neutrality or common carriage is essential. What about tying it to significant market power (SMP) a current EU focus for regulation? Again, I don’t have a pat answer, but I know the current network neutrality discussion is distracting us from our strategic interests, i.e. user-owned or controlled fiber, with rampant competition at all higher levels.
Comments