Many folks talk about broadband build out as if it were rural electrification. I do agree with this in one sense — it is a critical part of our infrastructure and links to our tradition of ensuring that we remain one country with access to vital services for all. In this respect, broadband is similar to telephone/voice, electric power, sewage, roads, and other other public utility/natural monopoly type investments. But it is fundamentally different from all of these in a fundamental way. Other public utilities have high initial construction cost, but then have very predictable maintenance and upgrade costs. This makes it possible to solve some problems with a huge one-time grant or, for the private sector to make a serious cap ex investment, but then budget for regular upgrades based on projected need and maintenance based on standard depreciation.
Not so broadband. As our technological capacity increases, we increase both the potential capacity for the network and our capacity to use the network in unpredictable ways. But we have neither public policy nor private sector models that acknowledge this — with the possible exception of Verizon, which solved the problem from their perspective by aggressively pulling fiber/overbuilding capacity as to current demand where profitable and aggressively selling off high cost rural regions. And, while that works for Verizon and its shareholders, it rather sucks from a public policy perspective.
I call this the Zeno’s Broadband Buildout Problem. No matter how much Achilles invests in build out, he will never catch up to the limit of possible upgrades. As I explain below, my tentative conclusion is that the right public policy result is a recognition that we don’t get to do a one time investment and go away, but need to continue to experiment to find sustainable models that factor in growth rather than simply look at build out followed by steady state. I’m not sure beyond that, other than my conviction that anyone who shrugs and says “that’s why the government shouldn’t do this at all” is definitely wrong.
OTOH, it also means I find the speed upgrade in the stimulus package — 45/15 for wireline and 3/1 for wireless — pretty good despite the fact that many of us want to reach the 100 mbps or even 1 gigabit/second capacity for future network needs. Broadband Achillies may not be able to catch up to Bandwidth Demand Tortois, but that doesn’t mean he gets to slack off either. A good swift stimulus in the patootie is actually a pretty good idea, given the open ended nature of the problem.
More below . . . .
When we think about public utilities, we generally think about electric grids as an archetypal example. You can go with private companies, municipally owned, co-ops, or even big honking Federal investment. Generally, you upgrade it based on predicted increases in energy demand. Sometimes there is new technology that you need to introduce, but it is an infrequent enough occurrence that you can budget for it and plan it. Even then, in the regulated world, it is a royal pain in the neck. Back when I did rate of return cases for stranded natural gas and electric generation clients about 10-15 years ago, we would have huge fights over whether this technical upgrade or that piece of technology was really necessary and how to fit it into our rate of return.
But a really good communications system is a matter of continuous non-stop upgrade. Even with fiber, you are not going to just pull glass and go away for 20 years with regular maintenance and easy to predict upgrades. It’s going to be dynamic. This is one of the things that made life so awful for cable, why AT&T is struggling and while VZ looks like evil geniuses for going straight to fiber and ditching the high-cost regions. Both the cable cos and AT&T tried stopgap measures to do short term savings and please investment analysts. Now they are either doing their best to squash customer demand (excuse, “manage bandwidth hogs”) or are totally rebuilding their networks for the second or third time in 10 years, or both. As the technology improves, the possible capacity of the system improves at pace that demands frequent upgrades and adjustments.
Worse — from the point of view of the network manager, better for society as a whole — that expanded capacity has real use. Build fatter pipes, and people build better and more productive applications for fatter pipes. Sure, you can get by with dial up or dinky FCC-style “broadband” for a lot of things even today. But is that really our measure of success?
This is another reason I am not too heartbroken that my stimulus Achilles is not going to catch up to the broadband tortoise anytime soon. It can’t happen in one bite, so I need to be prepared to constantly close the gap. Certainly the stimulus bill has its limits. It focuses primarily on rural build out (although I am hoping that other pots of urban agenda and health IT reform money can go toward getting our poor and unserved urban folks connected). It boosts speeds, but remains asymetric and, no surprise, most of us would like to see better than 3/1 for wireless or even 45 mbps for wireline.
But we will never be able to “future proof” a network. Worse, from my perspective as someone who believes that broadband in the home has huge transformative power, the good folks at the PEW Internet and American Life Project shows that the die hards on not getting internet in the home generally see no reason to buy the service. After all, they get email through dial up. Why would they want more?
A few months ago, I half-seriously proposed that Verizon give away FIOS for free for a year to get folks addicted to speed. I believe in a similar strategy here. I think we — at every level in the public and private sector, not just as a matter of federal policy — need to figure out how we can sustain a system where we will want to constantly increase the capacity of the network. The first step is getting people to consider what a new network could do for them.
For that, they need the first quantum leap from existing broadband speeds to something significantly higher — and all at once, not in the slow way the incumbents are handling these upgrades. Plenty of folks once thought they could manage with a horse and buggy. What changed their mind was not a dinky local trip at 15 miles an hour, but the ability to make a trip that used to take days in only a few hours. We need the same thing here. For what the folks still on dial up or offline at home use the network for now, dial up speeds are fine. The jump in capacity from 56 kbps to a promise of 5 mbps that really only delivers speeds of 500 kbps when folks are home in the evening and everyone is congesting the network is hardly noticeable, and certainly not worth an extra $20 or more a month. But give these same folks a bump up all the way to 45 mbps down/15 up, and the world really does change enough to notice.
So I don’t see the speed issues in the stimulus package as a problem or “making the good enough the enemy of the good” or anything like that. The problem of future upgrades is a systemic problem, one neither the private sector nor the public sector has figured out how to solve. But if Broadband Achilles is never going to catch the Bandwidth Demand Tortois, then we should not think we can use the stimulus to build a 100 mbps network. The smarter move is to use it as, well, stimulus, something that advances us to the next level and makes folks thirsty for more.
And on the plus side, there is no way City Mouse is going to sit still for Craptastic speeds of “up to” 15/really pathetic when Country Mouse has speeds of 45/15. So like it or not, the incumbents will need to explain why they can’t provide 45/15 and why they have to impose bandwidth caps, when Country Mouse system doesn’t. Not as good as direct competitors, perhaps, but a useful benchmark.
As always, we need to keep in mind that the stimulus package is not a cure all for everything. It is meant primarily as a Keynsian nudge to alter certain structural elements of the economy rather than a total makeover. We will still need to figure out how to solve the longer term problem of an open ended commitment to a vital element of infrastructure. For now, I am content to slip Broadband Achilles a few subsidy steroids so he can get back in the race.
Stay tuned . . . .