April 17, 2005

Japan Inc. Redux

Thomas Bleha must love the movie Rising Sun. There's an awful lot of Japan envy in that film, with bumbling Americans always one step behind the wily Japanese. There was a lot of that sort of thinking in the 70s and 80s. Japan was cleaning our clocks, their bureaucrats were promoting key industrial players, managing competition, and increasing efficiency. We were doomed to play second fiddle to the land of the rising sun.

It didn't work out that way.

Here's an article that fits right in with those pessimistic times. The Japanese are picking winners, subsidizing broadband, and will reap the fruits of all that fiber in the ground to gain a new broadband future with lots more economic growth. And you know what? He might be right. The managed competition method, like a stopped clock, does work every once-in-awhile. The benefits of the model are and have always been highly visible. It's the costs that are hidden and remain hidden.

The costs of subsidizing bandwidth vary by population density and the US has one of the lowest population densities of the entire 1st world. Australia beats us out but nobody really talks about their amazing broadband national coverage and it's for the same reasons.

Aside from a throwaway line ("The United States' vastness no doubt complicates the task, but it is no excuse for not undertaking the job") the problem of wiring the US remains largely unexamined. Canada has the happy circumstance that its population is largely concentrated in a band on its southern border. US population is concentrated in several noncontiguous megalopoli, the NE corridor that starts around Boston and runs along the coast to around DC, the West Coast Megalopolis that runs largely between San Francisco and Los Angeles, The Great Lakes megalopolis starts in Wisconsin and stretches all the way to western NY. There is also a South Florida one and several other wannabees like the emerging Texas "Triangle Cities" one.

With clumpy population separated by large distances, the cost of wiring a national infrastructure goes up while the relative benefit goes down. Network effects mean that the larger your network, the better your results. We're trying to cover 10 times the area to gain twice to three times the population node connectivity of Japan. That means that the cost to benefit ratio is somewhere between three to five times better for them.

The natural result of this is that we're going to lag. We're going to wait for costs to drop and use cheaper networking methods as they arrive as our . Goosing progress by dumping money into a national network infrastructure initiative is not only risky in that it will tend to give us bad habits, it is also going to cost us more for less benefit.

The geography of our nation, the distribution of our population, all mean that we're forced to play the "second mover advantage" game when it comes to broadband, at least until we figure out how to take a couple of zeroes out of lift costs per kg or some other sort of way to get data networking platforms up above us cheaply.

We can go expensive now or we can adopt a multi-platform competitive approach and try to encourage new entrants so that they quickly improve the cost to deploy and maintain bandwidth infrastructure. This, essentially is our current position with the FCC trying to spur innovation so that a national broadband rollout makes sense. Waiting patiently has never been a US strong suit but we're doing it now and actually playing the game pretty well. It would be a shame to ruin things and merely imitate solutions that do not scale to our particular density profile.

Posted by TMLutas at April 17, 2005 07:57 AM