The FCC says it, the CTIA says it, the CEA says it, and the network operators say it: If we are to provide sufficient wireless access for broadband, we need more spectrum. Some have estimated that we will need another 700 MHz of broadband spectrum to meet all of the demand going forward. Is there that much spectrum available? If so, who will get it and who will build out the systems to use it?

The FCC is already looking toward the TV broadcast industry as a "donor." After clearing TV channels 52 to 69 and auctioning most of the spectrum (24 MHz was given to public safety), the FCC is now floating a trial balloon to push broadcasters ever further and take some more spectrum away from them. The National Association of Broadcasters (NAB) has already opposed this idea and, of course, those who want to use some of the spectrum for unlicensed broadband will not be happy. But I think this is a good place to pick up some additional frequencies in a portion of the spectrum that is ideal for mobile broadband services.

The latest number I have seen is that only 20% of the U.S. population still receives TV transmissions via an outside antenna as opposed to cable and satellite. Some of these people do not have access to cable or satellite (they have to be able to see the sky in an easterly direction) and some have chosen not to pay for cable or satellite service. Many are in rural America, but some are in cities and suburbs. If we do away with over-the-air broadcasting completely, many of these people will lose access to TV. Is it fair to take away a free service we have had for 30 or more years to provide more spectrum for broadband wireless services?

If we took channels 40-51 back, we would have another 66 MHz of spectrum available (each TV channel has 6 MHz of spectrum). If we took channels 20-51 back, we would have 186 MHz of spectrum and would probably be able to provide over-the-air TV services using the lower channels. While 186 MHz is a lot in contrast to what is available today, it is still nowhere near the 700 MHz we are told we really need. Much of the rest of the spectrum would have to come from the government side that is managed by the National Telecommunications and Information Administration (NTIA), which controls more spectrum than the FCC. Is 700 MHz the right number? How do we know when we don't know how efficient our new technologies will be or what customers will want to be able to access on their mobile devices?

Whatever amount of spectrum we steal from somewhere else for broadband-and we will have to steal it since there is no unused spectrum in the United States-it will probably not be enough. We need to understand that even if we had broadband systems using every bit of spectrum that is technologically usable, and even if the technologies had two to three times the capacity they do today, there is no way we would have enough to replace the wired infrastructure. We have what we have and the way the Internet is growing, it appears as though we will always be short of spectrum since demand for wireless broadband, especially video services, keeps growing. Recently, 3UK, the smallest of the network operators in the UK, stated that today 94% of its network traffic is data, up from 15% just four years ago.

More Spectrum, But for Whom?

There is no doubt that we need the spectrum, and we need it sooner rather than later given the growth rate of wireless data services and the fact that more than 40% of all wired Internet traffic is now video. But who will buy it at auction, who will build out the networks, and who will manage them?

I am sure various groups will have ideas for how this spectrum should be allocated. The first will be those who favor unlicensed spectrum. Give it to us, they will say, and we will get more broadband built out for those who really need it and at a price that is more affordable than what we are paying today. Next will be the group that says it wants the spectrum auctioned to companies that can build it out on a regional or nationwide basis. But they don't want incumbent operators to be able to bid on it, or most of it, and want to make sure the winning bidders are new companies that will come in and challenge the incumbents, driving down the price of broadband access. Internet companies (Google?) might promise free wireless access based on an ad revenue model, some existing TV stations might want to enter the broadband business, and there are probably others.

The incumbent operators already providing broadband services and those networks are carrying more data than ever before, will be the other group that will want access to this "new" spectrum. These operators will say that since they already have the towers, back-end infrastructure, and billing systems in place, they could put this new spectrum into operation faster and more efficiently than any of the other groups.

The Fight Will Be On

Meanwhile, the public safety community is waiting to see what Congress and the FCC will do with its request to re-allocate the D Block to them so they can combine it with their own 5 MHz of broadband spectrum, now licensed to the Public Safety Spectrum Trust (PSST). The FCC will also be putting the finishing touches on its broadband report that is due into the executive branch by February 2010, and meanwhile, the stimulus funds ($7.2 billion) continue to sit in Washington until someone can decide which grants to fund and for how much.

So, we ask, will we get more spectrum, when will we get it, who will be entitled to bid on it, and what will be the build-out requirements? It will be a few years before any of this spectrum can be released and there will be lobbying efforts from a variety of interests. If Congress and the FCC continue to believe that the Internet is the most important technological advancement of all times, the results can be guessed now. The spectrum will be geared toward the Internet community and made available at auction to only companies that do not already have a wireless network in operation. This will make the DC crowd feel great knowing it is spurring more competition. But more competition will result in a market expansion followed within a few years by a market contraction.

Sometimes it is difficult to understand why people think that the more competitors we have the better it is for everyone. The U.S. government seems to want more competitors in the wireless space and keeps pointing to Europe as a leader in technology and openness. In reality, the United States is far ahead of Europe when it comes to competition and pricing. There are some countries that have more broadband penetration than we do, but that is because the government helped make it happen and opened the treasury doors to help pay for it.

What would it cost, today, to build out another complete network across the United States? The guesstimates run from about $8 billion to more than $30 billion in capital expense and $billions more in operating expenses. I have to wonder what the payback period is for a network that costs $15 billion to build and $3 billion a year to run when it is the seventh, eighth, or ninth network in a given area. I guess I must have missed something in my economics classes, where I thought I was taught that in a free market, the number of companies that survive is based on the amount of demand for a service or product, the number of providers of that service or product, how much profit they can make by providing it, and how they treat their customers.

When those who have the ball, and therefore are in charge, look at the U.S. wireless market, they see four nationwide networks and seem to believe there should be six or more. What they are not seeing is the total number of network operators serving a given region. When we include MetroPCS, Leap Wireless, other second and third tier networks, and Mobile Virtual Network Operators (MVNOs), the number is not four, it is usually seven or eight, all trying to stay alive with a population base that cannot support that many.

It won't be any different this time around. The government will expand the number of networks providing services, the market will respond, and some of the networks will be bought or will simply go out of business, leaving their customers stranded. We have been through this before and you would think that we would learn. China, which is already a huge wireless market and has the potential to grow for many years to come, decided last year to REDUCE the number of network operators from six to three. I wonder what China knows that U.S. policymakers don't.

If you disagree with me and think that enabling more network operators will drive wireless prices down, look at Clearwire and Sprint and their "4G" network. They keep opening new cities but they had fewer new customers sign up last quarter than in the quarter before. And even though there are half-price initial offers, their standard pricing is about the same as everyone else's. Where are the price reductions caused by this new competition that claims 4G data speeds?

LTE, or even WiMAX-M, will be different. They will be faster and have more capacity, and it will be less expensive to deliver more data with these technologies. But every operator will be installing LTE systems eventually and they will all have the same data speeds. Since we are making sure that all of our networks will be open to any device (with caveats) and any data, what is left for these network operators to use to out-market their competitors? Pricing? Perhaps, but how low can they afford to go before they start losing money? The wireless industry can't charge extra for checked bags or a bag lunch.

I hope that if the Internet crowd does manage to win some of the new spectrum at auction, it actually builds out its own networks and finally learns what it takes to provide 99.999% reliable service. And if they do actually give away voice and data services and pay for the network with ad revenue, we will learn more about what customers want, what they can get, and what they will tolerate.

The biggest threat to all of us is losing network operators that care about building networks close to 99.999% reliable. You can reduce the cost of a network by $billions if you are willing to build out to 80% reliability, and if this is where we are headed, instead of more and better service, we will get less and poorer service. But as long as it's free, perhaps we won't have a right to expect it to work all of the time.

Andrew M. Seybold