The next broadband battleground

Net neutrality, also called network neutrality, has three components. First, users should have access to any content they choose while being able to attach any device to the network--so long as it will not harm the network. Secondly, there's the issue of the management of the broadband network. This involves practices such as shaping traffic, reserving capacity or spectrum, or blocking viruses and harmful programs. The third component involves initiatives designed to convince applications and content providers to make new types of payments for premium services, such as faster delivery or caching content closer to the edge of the network.

The initial federal policies on these components are clear. Consumers have already won on the first component--the FCC or Congress will step in if incumbent networks are blocking or discriminating in ways that consumers can observe. But network owners will initially prevail on the rest. Until there is demonstrated competitive harm, the government is unlikely to limit network owners' flexibility to manage their networks or strike private deals for faster or different services.

Proponents of open networks should adopt a market-based strategy of strategic bypass.

The policy debate is effectively over on the federal level. But the battle is not over. Proponents of open networks--where all three of the Net neutrality components are implemented in ways favorable to the consumer or application provider--should adopt a market-based strategy of strategic bypass.

Two premises underpin this strategy. First, consumers prefer open networks. Many years of experience in U.S. markets show that if one network is open, competing networks will also be open. Both cable and wireless broadband networks have to date been open despite having no regulatory requirement. Why? Because it was the only way they could compete with the telco networks. The second premise is that it will be impossible for a network owner to have different "openness" or Net neutrality policies for different geographic regions in its network. Such variable policies might be technically possible, but they would be politically indefensible.

This leads to an interesting conclusion: If one broadband network in a geographic market remains open, its competitors will also have to remain open as a business imperative--and each of them will have to remain open across their entire networks.

This means you don't need to build a third pipe throughout the U.S. in order to force open broadband networks. Instead, you can engage in strategic bypass, constructing a "third pipe" in a handful of economically attractive, geographically dispersed markets. The third pipe would have to be a true, open broadband network (probably above 2Mbps symmetrical), and would have to be available to both residential and business customers throughout the market. The technology used is irrelevant--it could be BPL, fiber, Wi-Fi, WiMax or any other fixed or mobile wireless technology.

Bypass-market incumbents would have to either maintain their openness policies or close their networks and cede most of each market to the bypass technology. The second option is probably not viable. By ceding market share, incumbents would also be confirming that their ability to impose closed network policies in non-bypass markets resulted from their market power in the last mile, and not from efficiency, consumer demand or other acceptable market forces.

Markets presenting excellent characteristics for bypass would be Philadelphia, San Francisco, Denver, Milwaukee (or Chicago), Miami (or Atlanta), and New York (or Washington DC/Montgomery County, Md.). A third pipe in six (and possibly as few as four) of those markets would force seven of the eight largest incumbent RBOCs and cable MSOs to choose one of these two unpalatable options. In addition, the two largest wireless broadband networks (Verizon Wireless and Sprint) are in all of those markets and would probably also have to retain their existing openness policies.

Here's where it gets very interesting. Seven of the cities have issued RFPs or announced plans for citywide municipal Wi-Fi networks. The cost of putting in a Wi-Fi network will be between $10 and $20 MM per city. Annual operating costs would be only a couple of million dollars a year. So for somewhere between $75-120 MM over the next two years, content and application providers and other open network supporters could effectively guarantee open networks over the long term.

Google, EarthLink, Microsoft and others may have already figured out this strategy. If they haven't, they will soon. The battle over municipal Wi-Fi is now a battle over open networks, not municipal finances or the fairness of municipal competition.