There’s no question we need a lot more data. It’s going to be important to evaluate the industry over time. For example, we know very little about prices and demand. The 477 Report (an annual report broadband providers submit to the FCC) might not necessarily need to collect it. But there are other tools we can use, like the Census Bureau may start asking broadband usage questions. And the Bureau of Labor Statistics. We can think about ways to modify the data.
Q: The $350 billion number has been tossed around quite a bit since the FCC’s status report a month ago. That number is the estimated investment needed speeds of 100 megabits per second. Does the average consumer need that kind of speed to get the benefits of broadband?
We should look at types of applications available today. We don’t care about broadband per se, but care about it as a means to doing many other tings. It’s a general purpose technology that enables all kinds of other activities. So it’s the activities that matter.
There just really aren’t applications right now that need 100 megabits per second. That’s not to say there won’t be, of course, and the presence of the infrastructure that can handle that is important. But everything is about tradeoffs and there are very limited resources available.
Q: Blair Levin has also emphasized the need for more spectrum. Where would that spectrum come from?
We’re going to have to look and see where spectrum is used for things that aren’t particularly valuable, where it wasn’t acquired through an auction mechanism. Another place to look is government. That’s going to be hardest place because agencies aren’t going to want to give it up. It’s hardest to create incentive for them to do that. But there are huge swaths of spectrum there.
{mosads}I also think there’s potential in the secondary spectrum markets (in which a spectrum owner leases the use of the spectrum to another party). That’s a market where already there’s more activity than most people think, but not nearly as much as there should be. Already, about as much PCS spectrum changes hands every year than was released in AWS spectrum auction.
There are barriers to making that market work better, like fixing the FCC database, the Universal Licensing System, which is so complicated few people know how to use it. A well-functioning market requires information. A market can’t work if you don’t know who owns what.
The FCC has kept records of spectrum for as long as there have been records to keep. One of the problems with the ULS is, every time there was a new change to spectrum, that was added to database. So it’s a database of many modules that feels like its being held together with duct tape and coat hangers. They’ve done a great job making it work, given that they’ve never been given the resources to do it right.
Sometimes companies buy another company just to make it easier to get to that spectrum. If you’re a buyer, you need to know who owns the spectrum you want. There have to be easy ways to do it. There’s even a company called Spectrum Bridge that want to become a platform for trades and put together a spectrum price index.
Another barrier to secondary markets is, while the FCC has moved toward this, the spectrum needs to be unencumbered. The user needs to be able to do whatever they want with it. If the usage is set, even if it’s the right thing at the time, within a few years some new service will be available, and the spectrum use needs to be able to change.