Thus one should immediately be cautious when we have talk of a 35x increase in demand over the next five years, especially when the consequences of this demand growth in terms of the need for additional spectrum can be changed significantly by virtue of a simple mathematical error in the FCC’s calculations.

Spectrum demand growth could be significantly different to that projected in the FCC’s analysis with only small changes in behavior and/or pricing. For example, many people may well want to give up their fixed line DSL or cable connection for mobile broadband. However, if we still have pricing of $10 per Gbyte, as AT&T currently charges, not many people will be prepared to see their data usage double every year (as is implied by a 35x increase in demand over 5 years) reducing the need for more spectrum. On the other hand, if spectrum does end up becoming scarcer, then it simply won’t be in cellular operators’ interests to promote low margin (in terms of dollars per MHz) applications like fixed broadband replacement.

Its certainly not the case that there is a bubble in spectrum prices to anything like the same extent as there was in the 2001 3G auctions, but that was another example of where expectations of demand growth were out of step with reality (as was LMDS spectrum pricing). However, I would argue that when everyone is saying there is a spectrum crisis, based on huge demand growth, the smart money should be on the opposite side of that trade. As they say “buy to the sound of cannons, sell to the sound of trumpets”.

Terrestrial CMRS services also saw their share of improvements in spectral efficiencies from the air interface over the last decade, but capacity gains and implementation of advanced waveforms have usually been demand driven and implemented over a period of time through standardization. These improvements have been implemented because they are ultimately more cost effective than acquiring additional spectrum. Today, we employ advanced frequency-reuse techniques by deploying advance antenna beam-forming technologies, and by deploying base stations with a smaller coverage footprint, (pico cells) in order to effectively reuse the spectrum available. All of this is done because the technological improvements are ultimately cheaper than the acquisition of additional spectrum (if available at all in some markets).

The problem with spectrum is there is no like substitute. There is no technology that will show up overnight and ultimately create “more spectrum” at a given frequency, that can be licensed. The spectrum is licensed and secured with payment of a fee. If 1996 were revisited today and the FCC were to declare that all the CMRS providers must wholesale spectrum to any interested party that came along. Then I would assure you that I would be purchasing NOKIA stock as soon as the market opened on Monday morning.

Unlike a CLEC that raised a couple of million dollars and purchased a Cisco switch in 1998 to compete in a Local Exchange Market in which mobility of the phone number was not even a consideration. In the CMRS space, an operator must provide wide ranging roaming capabilities which will require O&O spectrum in the most highly populated markets in order to compete effectively. Those are the major markets where the spectrum crisis is upon us.

Just as we have seen a consumer market transition away from the POTS landline telephone business model toward a “mobile only” telephony service, we will also see a transition away from the terrestrially provisioned broadband connection to a “Mobile-Only” broadband connection. I personally know of three households that have terminated their home based internet connection in lieu of a mobile phone only connection via their Iphones. The desktop computer market is in a steep decline in the consumer space, all you have to do is walk into a Best Buy and look at the change in the “product mix” on display as compared to 5 years ago.

I personally have no home broadband connection anymore, I use a Verizon MiFi for internet access, it is mobile, and travels with me anywhere I go.

There is no unwinding of a bubble, as the bubble has not even begun to form. As a matter of fact. If you went back and performed an analysis of spectrum capacity Mpbs, vs Cost per Mhz/POP vs Propagation Characteristics. You would find that the latest FCC CMRS auction spectrum provided spectrum at a significant discount to historical prices. in Auction 5 which ended in 1996, and Auction 73 which ended in early 2008. You will find that spectrum actually sells at a steep discount to 1996 levels using 1996 dollars. (Nextwave bid nearly 1 Billion dollars for 30Mhz at 1.9Ghz for the NYC BTA in Auction 5 which excluded Non-Designated Entities bidders vs. Verizon’s winning bid of $429MM for 12 Mhz at .7Ghz for the NYC BTA in Auction 73). If you want to call a bubble. That would have been the European UMTS auctions in 1999.

ORBITRAX