Gaping hole in models for using spectrum efficiently

In February, the FCC’s Office of Strategic Planning and Policy Analysis published three studies (1, 2, 3) on spectrum licensing and spectrum utilization.  Thanks to Nick Ruark for pointing them out.

Since the papers discuss tradeoffs between licensed and various “unlicensed” approaches to spectrum utilization, I printed them out and threw them in my luggage, finally reading them today.  These are theoretical studies that assume you are familiar with Nash equilibrium (named after John Nash, the mathematician made famous by the movie A Beautiful Mind) and mechanism design (a part of game theory that is directly applicable to designing auctions).  Heavy stuff indeed!

Reading on I was struck by a gaping hole in their assumptions.  Of course an academic study can start from any assusmptions the authors prefer, but it helps if there is some relationship to reality.

What’s missing in these papers is any thought of future innovations.  The key feature of the currently available slivers of “license exempt” spectrum (in bands just above 900 MHz, 2.4 GHz and 5 GHz) is that they provide options for future innovation.  None of what we know of as WiFi or Bluetooth, either technology or business models, was in existence when the FCC originally opened up the 2.4 GHz band — a band that was previously devoted to (and is still used by) microwave ovens and industrial heating equipment.  Indeed, the principal reason why the FCC is studying spectrum allocation models today, is the innovation that resulted from opening up license exempt spectrum.

So the goal isn’t to figure out how to make current users of unlicensed spectrum use it more efficiently with today’s technology.  The goal is to motivate people to find new ways to get more utility out of the spectrum in question.  How do we foster innovation?  How do we increase the opportunities for innovators to find new and improved ways of using wireless spectrum?

Of course, I shouldn’t be too hard on the authors.  After all, the economics profession only began to model innovation in the past 20 years or so.  For the previous 200 years, innovation was lumped in with other “exogenous” factors, i.e. factors that didn’t fit the model.