Network Neutrality in Practice

Larry Lessig periodically links to this 2000 article in the Prospect about network neutrality. In it, he makes the closest thing I’ve seen to a convincing argument that network neutrality regulation was responsible the Internet’s growth:

But there is one part of the Internet where end-to-end is more than just a norm. Here the principle has the force of law, and the network owner cannot favor one kind of content over another or prefer one form of service over another. Instead the network owner must keep its network open for any application or use the customers might demand. Competitors must be allowed to interconnect; consumers must be allowed to try new uses. In this part of the Internet, “open access” is the rule.

This part of the Internet is–ironically enough–the telephone network, where because of increasing regulation imposed by the D.C. Circuit Court of Appeals in the 1970s–leading to a breakup of AT&T by the Justice Department in 1984 and culminating with the Telecommunications Act of 1996–the old telephone network has been replaced with a new one over which the owner has very little control. Instead, the FCC spends an extraordinary amount of effort making sure the telephone lines remain open to innovators and consumers on terms analogous to the terms required by an end-to-end principle: nondiscrimination and a right to access.

The argument here is that by ensuring that any consumer could call any ISP in his or her area code, the FCC’s regulation of the telephone network had the unintended consequence of imposing a de facto network neutrality rule on telecom companies, thereby ensuring that the Baby Bells couldn’t leverage their ownership of phone lines to control the Internet.

This isn’t a crazy argument. I’m rather annoyed that my local telco, SBC (now AT&T) requires me to get Yahoo-branded Internet service, even if I’d rather connect via another ISP. The fact that anybody could become an ISP by connecting to the public phone network indisputably had a positive impact on competition among ISPs.

Still, this argument doesn’t strike me as being quite right.

The problem is that common carrier regualtions impose a relatively static model on telecom companies. If all we wanted was for the Baby Bells and cable companies to continue providing the services they’ve always provided, a network neutrality rule would work pretty well. If our primary fear is that the quality of broadband service will degrade over time, then it might be possible to craft regualtions that require them to continue providing the services they’ve traditionally provided. Bureacracies are pretty good at getting in the way and slowing down the pace of change.

But I don’t think anyone would be satisfied with merely maintaining the status quo. And when you’re trying to promote innovation and new investment, bureaucracies’ penchant for getting in the way isn’t an advantage any more. When a telecom company is building new infrastructure, it’s not obvious what the appropriate non-discriminatory policies and prices for access to the new infrastrcture might be. And regulatory uncertainty dampens new investment.

Here’s a quick and dirty chart that illustrates the problem:

I should emphasize that this is a very rough effort, which I whipped up in an hour or so. The usual caveats about statistics and lying apply. The pre-1997 data is based on this Wikipedia article (along with a couple of educated guesses for the late 1980s, for which the article didn’t have specific dates). The post-1997 data is based on my personal experience: I had a 56k modem in 1998. I got a 256k DSL line in 2000, a 1 Mbit cable modem in 2002, and a 5 Mbit cable modem connection in 2004. I used 15 Mbit for 2006 because Verizon started to roll out 15 MBit FiOS connections last year, and AT&T is planning to roll out similar fiber service this year.

Now, I was in high school when the 1996 telecom bill was passed, so I’m not going to claim there’s a causal connection between the abandonment of the common carrier regime and the increase in Internet connection speeds. Doubtless one factor was simply that consumer demand for broadband grew rapidly as the Internet went mainstream starting in 1995. Still, I think these data are suggestive. Consider this: between 1981 and 1997, a period of 16 years, the typical Internet connection speed increased by a factor of about 200, from 300 bps to 56k bps. Today, after only 9 years, AT&T and Verizon are rolling out fiber connections that boast speeds of 10-30 million bps–that’s another factor of 200. In other words, the pace of innovation almost doubled,

I don’t think it’s a coincidence that cable modems, the speed leaders for the last 5 years or so, have been classified as “information services” and have therefore been exempt from common carrier regulations since the late 1990s. And it’s also probably not a coincidence that the Baby Bells stepped up their investment in new fiber capacity only after they got regulatory assurances that those facilities, too, would be exempt from common carrier regulations.

In fact, I think you could even argue that the speed increases of the 1981-1994 period understate the extent to which common carrier regulations impeded faster data networks. Keep in mind that the speed increases were almost entirely the result of the ingenuity of modem companies who figured out how to squeeze more and more data over a voice channel. The 28.8k modems of 1994 were transmitting data over a line that was little different from the line used by 300 baud modems in 1981. Only in the mid-1990s did the Baby Bells start making infrastructure improvements, first modestly (when they introduced quasi-digital phone lines that allowed 56k modems) and then with a vengeance with the introduction of DSL services.

Are there other explanations I’m not thinking of for the above trend? Can anybody see any problems with my numbers?

SHARE: 24 comments