The proposed $39 billion merger of T-Mobile into AT&T started a firestorm of criticism from all directions, mostly on antitrust grounds. My bet is that the deal is not approved. The question on the table is “then what”? The demand for wireless bandwidth is growing exponentially. AT&T is trying to get more bandwidth by buying T-Mobile’s. Without more bandwidth divided by demand across the wireless carriers, the mobile revolution starts faltering by 2015 when bandwidth supply constraints hits exponential demand growth. My solution: consider applying the U.S. electricity generation regulatory model to wireless communications.
“It’s the Bandwidth, Stupid”
To me, it’s a given that wireless mobile devices (gadgets) are a key growth industry for the global economy. The reason Apple has the #2 market cap right now rests firmly on the shoulders of iPhone and now iPad growth. Or check out this video of Corning’s view of the future. All the tech companies are fixated on this rapidly growing market, with tablet sales alone projected at over 100 million units in a couple of years. So, huge, growing demand in devices that need mobile wireless bandwidth.
Second, the gadgets are consuming more data per device. That’s why Verizon and AT&T capped monthly wireless bandwidth last year for new subscribers. The growth of wireless video, video chat, navigation maps, e-book downloading, digital news and more are rapidly growing the wireless data consumption.
How fast does AT&T see the wireless network demands growing over the next five years? Think exponential.
AT&T projects an order of magnitude growth in bandwidth demand by 2015 to 50 petabytes a month.
Meanwhile, the wireless bandwidth spectrum is not growing. Oh, more cell towers are going up in poor reception areas and rural areas, but the basic supply of wireless electromagnetic spectrum is pretty much fixed (unless the government frees up more of its own reserved spectrum, but don’t count on that).
Which leads to point number three: fixed spectrum supply is about to be overwhelmed, inundated, creamed, smothered, drowned by a sea of gadget demand for wireless spectrum. Yes, 4G technologies will use that spectrum more efficiently, but there are limits to annual cap ex spending and nothing can ameliorate a 10x increase in demand over the next five years.
But the wireless spectrum is fragmented across multiple carriers by law in order to support market competition. AT&T customers compete with Verizon, T-Mobile, Sprint and other wireless spectrum lessees, reducing the spectrum for any one carrier customer’s. It’s like being trapped in the slow TSA line while other lines might be going faster. Point four: there is no legal basis for pooling the finite wireless spectrum, so every wireless provider does the best they can with the spectrum they lease.
Why Did AT&T Go After T-Mobile Now?
I am convinced AT&T went after T-Mobile for its bandwidth, plain and simple. There is no way AT&T can build enough cell towers in San Francisco or New York City to meet today’s demand, let alone an order of magnitude more bandwidth demand in five years. Telephone is betting it can beat the antitrust odds and gain more precious spectrum.
What If The Merger Does Not Go Through?
But what if politics and antitrust correctness deny the merger? The status quo starts rolling towards 2015:
- 10X demand still happens
- Wireless carriers put fingers in the dike
- Everybody is mad at the national disgrace of a wireless system with horrible response times
- Disgusted, users first slow usage and then don’t upgrade devices. This kills off tech innovation.
Don’t think that could happen? It already has before. CB radio in the 1970s: crushed by too much demand, too few channels, and foul-mouthed truckers. Dead as a mass market in two years.
Is There an Alternative to Impending Disaster?
I am a free-market thinker. However, in this case, it’s hard not to look at finite wireless spectrum as a natural monopoly. Let’s open up a public discussion.
My thoughts since the AT&T announcement Sunday have led me to the U.S. electricity industry, which is the 1990s was broken up into two components: electricity generators and transmission companies; and electricity marketing and delivery companies.
- The electricity producer/wholesalers are highly regulated and make a return on capital. They are economically encouraged to grow low-cost and peak demand generation capacity. They band together into grids to deal with spikes in demand and outages;
- The electricity retailers are in the less-regulated and profit from a markup on wholesale electricity. They compete on price, service, and support (not much, but there is competition).
I argue that the best national interest would be served if the wireless network function was given to a regulated monopoly charged with providing as much bandwidth as possible and as inexpensively as possible.
I’d take away public service spectrum and replace public service equipment like in-car police dispatching terminals with wireless digital-compatible equipment. Create an infrastructure boom that frees up lots of wall-penetrating spectrum. Voila, more very useful spectrum.
Since we wouldn’t have five or more wireless companies fighting for spectrum in every market, the local wireless network monopoly could focus on quality of service and improving gadget throughput.
The wireless service companies would sell price, service and support as retailers. They’s also sell phones and gadgets. AT&T and Verizon could keep their landline businesses.
Yes, there are some big issues like which wireless network technology to support, or whether to support them all going forward. But I think I’ve stirred the pot enough for today.
As usual, comments and feedback appreciated.