January 2006

I’m kind of baffled by the network neutrality debate. Take this quote for example:

At the end of the day, Google’s Davidson says that his biggest worry is not for Google but for the prospect of bringing fresh innovation to the Internet. After all, if worse comes to worst, Google can pay AT&T or BellSouth to maintain its role as the Internet’s dominant search engine. But the bright young start-up with the next big innovative idea won’t have that option.

That’s the last paragraph from a long article about the network neutrality debate. The article implies that the outcome of the network neutrality debate could determine whether the telcos end up in control of the Internet.

Except that it doesn’t make a lot of sense. Take the paragraph I just quoted. I’m having trouble even imagining a plausible scenario in which the above could happen. Think about it: let’s say that Verizon wanted to get every website on the Internet to pay it a fee for access to its customer base. How would they do it?

The first step presumably, would be to send an email to the webmaster of each site, informing them that they’d be disconnected if they didn’t cough up the required fee. But it’s likely that most web site owners–especially the smaller ones–would ignore the notices. A lot of sites, such as this one, are run by volunteers and don’t have a lot of money to spend on tolls to Baby Bells. Others would calculate that it’s in their strategic interest not to pony up the cash–especially since they still have plenty of customers from other parts of the country. So then what would the Baby Bells do? Simply blocking access to the sites would be financial suicide. The vast majority of Internet users access at least a handful of small websites on a regular basis. Hence, if you shut off all non-paying websites at once, you’d piss off virtually all of your customers. You’d spend years repairing the financial and PR damage from a stunt like that.

They’d have the same kind of problem if they chose to degrade the performance of non-paying web sites rather than cutting them off entirely. That’s more likely to just make it look like their own Internet service sucks than harm the targetted sites. I would be sad if my TLF readers had to wait longer to read my posts, but not sad enough to cough up money to the company responsible to get them to stop. And you can bet we’d add a banner at the top that says “Verizon customers: is this site loading slowly? Click here to find out why.”

A slightly less draconian plan would be to allow everyone to use the existing bandwidth, but to limit access to additional bandwidth to those that paid a special fee. But even here, it’s hard to imagine them getting very far. I mean, imagine if Verizon approached Apple’s Steve Jobs and told him that they wanted him to cough up some money if he wanted to be allowed to use their new fiber capacity for his iTunes video store. You know what he’d do? First, he’d laugh at them. Then he’d point out that if they don’t give him access to their bandwidth, he can cause every copy of iTunes to pop up a little notification during long downloads that explains that it’s Verizon’s fault the download is slow and explains how to switch to another ISP. I’ll give you three guesses who would run out of the room with their tails between their legs.

In addition to having content consumers want, content providers also have a direct connection with their customers–in many cases much stronger than the connection the customers have with their cable or phone company. If content providers and telcos began engaging in brinkmanship, the outcome would depend crucially on who consumers blamed for the mess. In that kind of PR battle, the telcos wouldn’t have a prayer, given that essentially the whole Internet would be lined up against them, and people increasingly get their news and information from sources on the Internet.

So I don’t see how a Baby Bell would go about extorting money from web site owners without shooting themselves in the foot. I’m having trouble even envisioning a plausible scenario, much less one that is likely to actually happen. The idea of a balkanized Internet sounds ominous in the abstract, but when you try to imagine how it would actually happen, it doesn’t make a lot of sense. Telcos need content providers at least as much as content providers need telcos. Content providers know it.

The telcos are wrong about their ability to charge content providers for the use of their “pipes,” they’re right about one thing: “network neutrality” regulation is a solution in search of a problem.

Update: Although he’s more ambivalent about network neutrality regulation than I am, Ed Felton highlights some of the same problems.

SHARE:

There’s a great conversation going on over at Marginal Revolution about net neutrality. As a card carrying free-marketeer I feel I’m expected to support Verizon, AT&T and the rest when they demand payment for use of their pipes. But I haven’t made up my mind yet. While net neutrality looks like forced access redux, I think it’s actually a much more complicated issue.

I am skeptical of regulation or legislation to enforce neutrality; preemptive regulation hardly ever works out they way it is intended. However, as Tyler Cowen points out, tiering the internet would change the nature of online content:

The beauty of the status quo is that web sites compete on the basis of consumer surplus alone. The bandwidth costs end up as a fixed charge on net access as a whole; I suspect this hits many inelastic demanders, a’la the Ramsey rules for optimal taxation. Admittedly it may be a bad deal for the poor who cannot afford to connect, but the overall arrangement enhances the long-run “competition of ideas” feature of the net.

It seems to me that the obvious solution to this problem is to get rid of flat-rate access charges and move to variable prices based on bandwidth usage. Sadly, consumers have historically resisted per-unit access charges, even when they would have come out ahead. They like the idea of not being rushed to disconnect or feeling pressure to monitor and cap their use.

What’s more, bandwidth consumption in itself is not the problem. Bandwidth consumption at peak times, causing congestion, is the real problem. But, as Arnold King points out, the internet is not suited to incorporate congestion-based pricing:

Think of Internet packets as envelopes with very exact formats for the address. The format does not provide for a way to designate the envelope as “high priority.” Even if it did, the cost of reading the “priority bit” on every packet header would almost surely exceed the benefits of congestion pricing.

Even if we did want to go the route of a two-tiered internet anyway, with one tier getting preferred delivery, it’s not clear how we can do this without breaking what makes the net unique. Ed Felten clarifies that “although the two-tier network is sometimes explained as if there were two tiers of network infrastructure, the obvious and efficient implementation in practice would be to have a single fast network, and to impose deliberate delay or bandwidth throttling on non-preferred traffic.”

So, I guess what I’m ultimately arriving at is that while I’m far from sold on net neutrality regulation, I like the neutral nature of the internet. And, given that it would be technically difficult and unpopular with consumers to tier the net, I’d like to think that a free market would preserve neutrality. What we need to ensure is robust competition so that a small number of ISPs can’t push this down consumers throats. Jeff Pulver is right that what Google, Apple, and Yahoo need to do is call the telcos’ bluff and make it clear they’re not going to consider paying the ISPs. What’s are the ISPs going to do? Tell their customers they can no longer access Google or iTunes?

SHARE: 2 comments