So, I think the net neutrality issue is very difficult. I think it’s a lose-lose. It’s a good idea in theory because it basically appeals to this very powerful idea of permissionless innovation. But at the same time, I think that a pure net neutrality view is difficult to sustain if you also want to have continued investment in broadband networks. If you’re a large telco right now, you spend on the order of $20 billion a year on capex. You need to know how you’re going to get a return on that investment. If you have these pure net neutrality rules where you can never charge a company like Netflix anything, you’re not ever going to get a return on continued network investment — which means you’ll stop investing in the network. And I would not want to be sitting here 10 or 20 years from now with the same broadband speeds we’re getting today. So the challenge, I think, is to accommodate both of those goals, which is a very difficult thing to do. And I don’t envy the FCC and the complexity of what they’re trying to do.
The ultimate answer would be if you had three or four or five broadband providers to every house. And I think you actually have the potential for that depending on how things play out from here. You’ve got the cable companies; you’ve got the telcos. Google Fiber is expanding very fast, and I think it’s going to be a very serious nationwide and maybe ultimately worldwide effort. I think that’s going to be a much bigger scale in five years.
So, you can imagine a world in which there are five competitors to every home for broadband: telcos, cable, Google Fiber, mobile carriers and unlicensed spectrum. In that world, net neutrality is a much less central issue, because if you’ve got competition, if one of your providers started to screw with you, you’d just switch to another one of your providers.
This covers I think the central concern very well, though it’s surprising to me that Andreessen acts as if there’s no way to reconcile the twin goals of ensuring the investors in infrastructure can make a return and ensuring the infrastructure they build will be designed according to rules that serve the public interest. Of course there is — government subsidy. This is not a new idea. It’s been used in the past to build railroads, the phone network, etc.
If Comcast and Verizon (and Google and anyone else with a credible plan to build) gets $15 billion a year (say) in subsidies to build out a neutral network infrastructure, what’s the problem? Investors are now only in the hole $5 billion a year instead of $20 billion so they are much more likely to be able to make a reasonable return, society gets faster network speeds and permissionless innovation (such as the kind that allowed Andreessen to create one of the first web browsers, Mosaic, and to market it as Netscape, back before he became a VC). This seems to me like a “win-win.”
Further, Andreessen neglects to mention the best reason for network neutrality: it’s a better network design. It’s hard to anticipate all future uses of a given network. The phone network was originally designed for voice and yet it gave rise to the consumer internet; cable was originally designed for television but it’s now being repurposed for data usage. We don’t know if today’s heavy traffic users (i.e. streaming video) are going to be anything like tomorrow’s, so it’s better to build a network that can handle any type of data, rather than optimizing the network for particular uses which may not be as important as we assume.