If Comcast can’t make it in the wireless biz, who can?

Comcast(s cmcsa) tried hard to become a mobile operator, but it failed to find a way to make the business work — at least that’s what Comcast EVP of public policy David Cohen claimed in a fascinating blog post last week describing why it and its cable buddies decided to sell their spectrum and shack up with competitor Verizon Wireless(s vz)(s vod). That admission is a surprising one to say the least, and if Comcast is being forthright it also raises some troubling questions about the state of the U.S. wireless industry. If a company with a $71 billion market cap and deep roots in the telecom service provider business can’t make a go of the wireless business, what hope is there for any newcomer?

Comcast, Time Warner Cable(s twc) and Bright House Networks joined together as SpecrtumCo in 2006 to bid on and win licenses in the AWS spectrum auction, but according to Cohen, the venture couldn’t make the math work when it came to actually launching a commercial mobile service. SpectrumCo spent $20 million in prepping its frequencies for launch; tested equipment from Qualcomm(s qcom), Nokia (s nok) and Samsung; and explored multiple network technologies ranging from WiMAX and LTE to CDMA’s now long dead 4G standard. The group struck up a partnership with Sprint(s S) to share network costs and capacity, and it even invested $1.65 billion in Clearwire(s clwr) to gain access to its WiMAX network.

Despite all of those efforts, Cohen wrote, Comcast couldn’t make wireless work:

Ultimately, but only after a significant investment and years of substantial study and analysis, we determined that it was not feasible to use the AWS spectrum to launch our own wireless network because: (i) we did not have enough spectrum for anything other than an initial launch;(ii) the costs to solve for that problem were too substantial for us; (iii) the costs of building out a new national network were also too high; and (iv) there were significant marketplace hurdles in launching a new, competitive wireless service, which included access to devices at commercially beneficial terms and the need to obtain national roaming rights on attractive terms.

It’s important to note, though, that Comcast had an ulterior motive for its sudden soul-searching online. SpectrumCo’s license sale to Verizon isn’t yet final, and the Federal Communications Commission is paying particularly close attention on how this deal will affect wireline competition in the U.S.  The cable operators aren’t just selling their licenses, they’re also going into business with Verizon, splitting up the residential broadband market between them. The FCC has already asked to see their business plans in private, but Sprint and T-Mobile are demanding that the Verizon and the cable operators release them to the public.

All in all, the FCC probably will probably approve the deal, but Verizon and its SpecrtumCo sellers don’t want to give the FCC any reason to impose restrictions on the sale. Comcast’s problem: the loose lips of its CFO.

Why you don’t admit to being out for a quick spectrum buck

At a Citigroup(s c) investment conference earlier this month, Comcast CFO Michael Angelakis said that the cable provider never intended to use the AWS spectrum when it bought it, so it made perfect sense to sell it to Verizon. The mention didn’t get much attention at the time, but FCC Commissioner Robert McDowell noticed, and at CES, as Deadline reported, he questioned whether Comcast and SpectrumCo purchased their AWS licenses under false pretenses.

Spectrum licenses aren’t real estate – you’re not supposed to purchase frequencies and flip them for a profit. They’re held in the public trust and carriers purchase the right to use them. Part of the deal is those carriers agree to actually launch services over them within a certain time limit. At least that’s the theory.

In truth, companies buy and sell and spectrum like securities all of the time. ‘Carriers’ like Aloha Partners and Vulcan Spectrum basically exist as spectrum holding companies. Dish Network(s dish) may similarly be looking to cash in on the satellite spectrum it purchased from Terrestar and DBSD. Once you get a license, it’s very hard for the FCC to get it back.

But there’s one thing you don’t do. You don’t come out and publicly admit that you have no intention of building a network. That’s why Comcast is walking back Angelakis’s comments. SpectrumCo and Comcast have plenty of time on its licenses so there’s little chance of the FCC taking the licenses back, but generally it’s bad form to acknowledge to your regulator that you’re squatting on the airwaves — especially when you’re asking that regulator to approve a multi-billion sale of those assets.

The FCC is a political entity as well as a regulatory one. If the Commission feels Comcast acted in bad faith for all of these years, it might take it out on the deal, imposing strict timelines on Verizon’s deployment and other requirements that Verizon is sure to not like. Ultimately the FCC wouldn’t kill the deal — those airwaves would never see a network if it did — but it could force SpecrtumCo to lower its $3.6 billion price tag to keep Verizon interested. Comcast has every interest in walking Angelakis’s comments back.

Was Comcast just warehousing spectrum?

I don’t think it was Comcast’s intention to merely flip its spectrum for a quick profit. But I also don’t believe Comcast and its SpectrumCo bedfellows planned on launching their own independent competing wireless carrier — which is what I believe Angelakis was getting at when he delivered his ill-timed comments. At the time of the AWS auction, the cable operators were still facing a big threat from DSL and the “quad-play” services of Verizon and AT&T(s T). The cable companies figured they needed to work with a competitor to the Baby Bells in order to add wireless to their portfolios. Becoming a spectrum holder would ensure they would be equal partners in such a venture.

The vehicle for that wireless service, SpectrumCo and Sprint’s Pivot venture, stalled right after it left the lot. The cable operators’ big stake in Clearwire now looks like a bad investment. Meanwhile, they watched Cox Communications strike out on its own in mobile, only to see it flop. If becoming a wireless player was a mere feint for Comcast, they certainly spent a lot of money and wasted a lot of time keeping up that subterfuge. Comcast made several honest, though misguided, attempts to build a legitimate wireless business with its partners until it ultimately concluded the effort was hopeless.

There has been all sorts of speculation that some new wireless operator like Apple(s aapl) or Google(s goog) will emerge to challenge the dominance of AT&T and Verizon, but Comcast’s failure should serve as a wake-up call. Apple and Google have enormous resources to be sure, but so does Comcast — and it had the backing of Time Warner and Bright House as well. Additionally, Comcast has extensive experience as a telecom service provider, selling broadband, TV and telephone services to millions of customers. It has the customer service and billing infrastructure in place, and it has its own extensive fiber transport network to serve as the backbone of any wireless service. If any outsider was positioned to break into the wireless operator club, it was Comcast.

Over the last few weeks, I’ve been following as it shakes up the French wireless market (check out Om’s excellent profile of Free Mobile). Free’s success in offering dirt-cheap voice and data plans has made many of us covering the telecom industry wonder if a similar upstart could replicate its business model in the U.S. The truth is if any company was ripe to model itself on Free, it was Comcast. Free has built the foundation of its wireless network on 5 million Wi-Fi access points, which it has spent the last several years installing in its broadband customers’ homes. To become a viable challenger to France’s Big 3 wireless operators, Free first had to be successful residential broadband operator.

When it comes to residential broadband there’s no more successful company than Comcast. It had the network in place, the spectrum on hand, the established customer base and the competitive drive to rock the foundations of the U.S. wireless market in ways that a Clearwire or a LightSquared never could. Yet it failed. It may be a hard reality to face, but the wireless market we have today maybe the wireless market we’re stuck with.