What Broadband Competition Means In A 100Mbps Nation

"Competition" is a key buzzword in telecom policy. Existing providers think there's plenty, while many public interest groups think we need more. But getting in the way of effective policymaking is that the truth lies somewhere in between.

If you only consider the broadband speeds of today then in some areas the marketplace can look pretty competitive. Between cable, DSL, possibly fiber, and a variety of wireless technologies, consumers do often have multiple choices for 1Mbps+ service.

And as new wireless technologies are deployed the possibility of a competitive 5Mbps marketplace isn't out of reach over the next few years.

But let's look forward to 2015. That's the year that Senator Rockefeller and others have set out as the goal for achieving a 100Mbps Nation. What will competition look like then?

First off, DSL and BPL are out of the picture. BPL in particular may find more targeted uses than delivering generic connectivity, but neither of these technologies are able to deliver 100Mbps, and assuming 100Mbps is universally available DSL is likely to go the way of the dinosaur.

Secondly, while wireless 4G technology promises speeds at 100Mbps, there's no guarantee these networks will be universally available in the US, their claims of 100Mbps refer to the capacity of each tower not how much each user will get, and a number of factors can contribute to lowering that number further, like rain and fog. No other existing wireless technology can reach 100Mbps speeds besides point-to-point systems that don't offer the advantages of mobility.

So what are we left with? DOCSIS 3.0-enabled cable and full fiber networks. That's it.

Assuming consumers will need and demand reliable 100Mbps speeds by 2015, that means the most competition we can hope for is between cable and fiber.

Taking that a step further, the only chance we'll have any broadband competition at 100Mbps is if we have a fiber pipe to every home, otherwise we'll be left with a cable monopoly, which is less reliable with fiber.

This doesn't mean wireless goes away, just that as overall demand for bandwidth outstrips the capacity of wireless it will become increasingly clear that it doesn't compete with wireline, it's complementary to it. Though within the wireless world there's likely to be competition between companies and technologies for quite some time.

What this all means is that contrary to industry opinion, without someone deploying fiber we either won't have any competition or we won't be able to achieve a 100Mbps Nation. But contrary to public interest opinion, the best competition we can hope for is a duopoly in a 100Mbps Nation.

Of course there's always a chance a new technology will come about that will enable 100Mbps competition either over wireless or copper. But not only does it need to be invented it will have to be deployed, which will be expensive and take a lot of time. Because of this we can't afford to wait around for something to show up that may never arrive.

Instead a pragmatic national broadband strategy must acknowledge the truth about competition in a 100Mbps Nation, and make one of two choices:

- Specifically support the deployment of both DOCSIS 3.0 and fiber in the hope of preserving competition between technologies.
- Or focus on establishing competition between services on the same strand of fiber.

This second option is important to understand and consider as with fiber everywhere it would be relatively easy to enable a world where any number of service providers could compete over the same infrastructure. In fact, our only hope for insuring competition beyond a duopoly in a 100Mbps Nation is if we pursue this model.

Looking further ahead there will come a time when we'll need a 1Gbps or even 10Gbps Nation and beyond, the outlook for competition between last-mile access technologies looks increasingly bleak.

At some point in the not-too-distant future demand for bandwidth will outstrip the supply that copper cable networks can offer. The total capacity of a cable plant today is 5Gbps if you took out TV. But fiber has no real limitation. While cable's testing 300Mbps service in the labs today, fiber's being tested at 6Tbps, or 20,000 times higher than cable.

And the odds of any other technology coming along that can keep up with fiber are essentially zero as fiber transmits data at the speed of light. So this will eventually lead us into fiber being a natural monopoly.

The only way we can prevent this is by either getting multiple fiber pipes to every house or encouraging competition between services on the same pipe. But given the high cost of deploying fiber and its ability to accommodate any number of service providers, the former of these approaches doesn't appear to be the most pragmatic relative to concentrating investment on establishing the latter.

But the most important takeaway from this post isn't necessarily that we need to be moving full-steam ahead to a One Pipe, Multi-Service World, which may take decades to fully realize, but instead that we must acknowledge the shifting dynamics of broadband competition that will be taking place over the next few years.

We can't afford to continue taking a "all broadband is created equal" approach and assume that all broadband will be roughly equivalent for all time. If we do then we risk not even having the competition of a cable/fiber duopoly.

And this isn't a discussion about some time in the distant future, we're talking about less than a decade. Policymakers need to understand that the broadband they're subsidizing today will be in place for at least a decade. So while these may sound like way out there issues, the decisions being made today must take into consideration where the market's going tomorrow otherwise we'll hamstring our country's ability to continue evolving its digital economy in the 21st century.