A tale of two Sprints: What if Sprint severs ties with Clearwire?

On Tuesday, we looked at one of Sprint’s possible futures—the future sans Clearwire that Sprint is publicly claiming to now pursue. If Sprint were to go that route it, eschewing Clearwire’s massive spectrum assets, it would start out with an immediate and sizable disadvantage versus its competitors (CP: A Tale of two Sprints, part 1).

Sprint’s new LTE network would suffer in terms of speed and capacity compared to those of AT&T and Verizon Wireless, as Sprint tries to balance its limited spectrum between its 3G and 4G networks and between its iPhone and new 4G smartphone customers. If Sprint were to sacrifice its sacred unlimited plans and made liberal use of carrier Wi-Fi technologies, the operator might have the capacity to make it through 2014 as it is projecting. If LightSquared prevails in its efforts to build a hosted LTE network on Sprint’s Network Vision architecture, then Sprint’s chances improve considerably (CP: Sprint confirms LightSquared tie-up). But one thing is certain: Sprint will face a spectrum crisis far sooner than its competitors and will have to go the markets to find what limited useful mobile broadband frequencies are available. Given the huge demand for airwaves these days, Sprint is sure to pay a stiff premium.

The irony of that situation is that a vast pool of licenses is sitting right at Sprint’s feet, but it’s opting not to dive in. Clearwire owns more than 100 MHz of 2.5 GHz frequencies in all markets, and in many larger markets has more than 130 MHz. That’s more than double the 52.5 MHz of PCS and 800 MHz that Sprint holds independently on average. Sprint would be hard pressed to walk away from such a goldmine, which leads one to suspect it has other plans it’s not revealing publicly. Its current stance may be a negotiating tactic, but no one knows what exactly it’s negotiating for. Some of the possibilities include an outright buyout of Clearwire, a move to take over control of the Clearwire board (Sprint is majority stakeholder, but doesn’t control the board) or taking back the spectrum it originally ceded to Clearwire either in or out of bankruptcy.

If any of those scenarios happen, it’s clear that Sprint would be a very different company than the one it currently says it will become. Sprint would have the single largest holdings of untapped mobile broadband frequencies in the U.S., enough to get it through the decade if not further while its competitors scramble to find more spectrum. It could use those frequencies to build some of the most robust LTE networks in the country, putting the AT&T and Verizon Wireless LTE networks to shame in speed races. Let’s look at what Sprint’s future 4G rollout would like if it could tap into those assets.

Sprint, Version 2: The 4G monster

One of the reasons Sprint has been publicly posturing is its unhappiness with the Clearwire with which it’s forced to contend today. And it doesn’t seem to have the power to change Clearwire’s direction. WiMAX as a mobile broadband technology is in its death throes. Clearwire’s time-division-LTE technology plans are an improvement since they do conform to a growing global standard, but Sprint still would face the daunting prospect of building a global TD-LTE ecosystem with foreign operators (CP: Can Clearwire build a TD-LTE ecosystem?). Sprint tried to do that WiMAX and has probably had enough of such antics. Finally, Clearwire’s national WiMAX expansion stalled early this year—stalling Sprint’s 4G launch with it—and it has no plans to resume it until it gets more cash.

Here’s what Sprint would likely do differently if it took control. It would first scarp Clearwire’s TD-LTE strategy. That plan works for Clearwire because it already has time division duplexing (TDD) networks in place which it could re-use for TD-LTE (CP: CTO explains Clearwire’s unique flavor of LTE). Sprint, however, would move the LTE onto its own Network Vision architecture (CP: Sprint lays out a vague path to LTE with $5B network modernization), which would allow it to adopt the frequency division duplexing (FDD) LTE configuration common in the industry and near universal in North America. Doing FD-LTE isn’t the most efficient use of that spectrum as 2.5 GHZ was never designed with separate uplink and downlink channels. Sprint would have to create those channels artificially by designated ad hoc ‘guard bands’ between them. It would lose a few megahertz here or there, but Sprint would have so much spectrum to play with I doubt it cares.

Sprint then could deploy the mother of all of mobile broadband networks—a single 20 MHz-by-20 MHz LTE carrier. That’s four times larger than the 5x5 MHz configuration it could hope to achieve if it pursued its current LTE-over-PCS plans and twice as large as the 10x10 configurations launched by Verizon and AT&T. Judging by the tests of 20x20 MHz deployments in Europe, Sprint’s 4G network could clock speeds in excess of 50 Mb/s. That network would provide all the capacity Sprint needs to get through the next few years, but if it did start filling up that bandwidth it could easily launch a 20x20 carrier, and possibly even a third in major markets. Once LTE-Advanced technologies become commercially available, Sprint could even start aggregating those carriers into massive wireless broadband pipes providing fiber-optic level speeds to the most data hungry devices.

Sprint would keep Clearwire’s WiMAX network running for several years to support its current 4G install base, but as it shifts those customers to LTE it could shut down the WiMAX network, freeing up its 30 MHz of capacity for LTE. Meanwhile, Sprint could use the new voice efficiencies of Network Vision to switch CDMA 1X carriers on PCS to EV-DO, creating more 3G capacity for its new bandwidth-hogging iPhone customers. As it would face no pressure to re-farm PCS for LTE, Sprint could keep its unlimited data plans going for the time being, at least from an operational standpoint. From a business standpoint the economics might make less sense if average iPhone data usage continues to skyrocket.

As Sprint winds down its iDEN network, it can start using those 800 MHz frequencies for data services. Its plan is to deploy CDMA in that band, but moving LTE to 2.5 GHz will relieve the pressure on CDMA at PCS. If the CDMA network is still going strong in the next two years—despite the iPhone strain—Sprint can immediately allocate 800 MHz for LTE, gaining its own ‘beachfront’ spectrum in the lower bands. That would allow it deploy bigger cells in non-urban areas and in general build a more consistent coverage footprint.

The benefits of keeping Clearwire in the Sprint fold are enormous. The question is whether the messy financial and structural relationship Sprint has with Clearwire is worth maintaining. Sprint is publicly stating that it’s not worth the bother, but privately it may have other ideas. We’ll probably find out what Sprint is really up to in the next few months.