Sprint eyes T-Mobile acquisition
American wireless still in upheaval
Sprint is reportedly in talks with Deutsche Telekom about buying up T-Mobile USA, creating a single network part-owned by Deutsche but able to compete with AT&T and Verizon.
The talks, which Bloomberg reckons have been going on over the last few weeks, could result in T-Mobile becoming part of Sprint with Deutsche Telekom owning as much as half of the resulting company, though Bloomberg also suggests that T-Mobile buying up a chunk of Clearwire spectrum is an alternative plan.
T-Mobile certainly needs to do something, it's been bleeding customers for the last few years and hasn't the radio spectrum, or the funds, to build a 4G network to compete with those being offered by AT&T and Verizon. Deutsche Telekom did a similar deal in the UK, ending up with half ownership of Everything Everywhere, which was created from T-Mobile UK and Orange (the latter being France Telecom's UK operation).
Given its inability to build a 4G network, there has been much speculation that T-Mobile USA would be a flagship customer for LightSquared. LightSquared plans to build a national LTE (4G) network in radio spectrum usually reserved for satellite communications, and then wholesale that network to operators – with T-Mobile being the most obvious potential customer. But in fact it has been Sprint who has been talking to LightSquared about sharing cell sites, and even discussing the possibility of using LightSquared's network in preference to the Clearwire one in which Sprint has ownership.
That's probably just Sprint trying to drive down the rate it pays Clearwire for network access, beating it with a LightSquared-shaped stick with no intention of actually switching networks. Making the switch would mean migrating users to new handsets and all sorts of complexity, and with Sprint being the majority owner of Clearwire, it has no interest in beating the network provider too hard.
Merging with T-Mobile, on the other hand, would achieve economies of scale and remove the requirement for an additional 4G network (which would be bad news for LightSquared). Bloomberg reports that T-Mobile's valuation is the sticking point, which shouldn't be any surprise, but a deal would seem to make sense if an agreement can be reached. ®
CDMA is cheaper to deploy, more spectrally efficient, and has better range than GSM. So, no, it would be quite senseless to tear down a CDMA network in favor of GSM. As for UMTS/HSDPA/etc., AT&T has had real problems deploying it, as the carriers in the US did not get "fresh" spectrum to deploy it in as they did in Europe, and the 5mhz channel width made it hard to deploy in the existing spectrum. Really, the sensible solution would be to deploy LTE, and phase out both CDMA and GSM over time.
As for having to deploy a 4G network -- well, Sprint & Clearwire are using Wimax.. which when they started deploying it made perfectly good sense, but now means they have to use specialized phones for 4G (CDMA+Wimax hybrid phones). If T-Mobile used Clearwire for 4G, then, they would also need custom handsets, just as they would if they went with LightSquared. However, T-Mobile is no stranger to this -- they have deployed their HSDPA in the AWS band * -- so these AWS 3G phones are specific to T-Mobile and IWireless (the plain 2G GSM uses the standard bands though.)
Side note -- first T-Mobile (US), then AT&T, have started falsely claiming their HSPA+ networks are 4G (so they can "keep up" with Sprint's Wimax and Verizon's LTE deployments without actually deploying anything.). This will be especially confusing for AT&T customers; AT&T plans to roll out LTE towards the end of the year, so since they are falsely claiming their upgraded 3G network is 4G, it'll be like "Oh, here's some *other* 4G". Yeah. T-Mobile takes the cake, they have had recent ads specifically targeting AT&T and IPhone 4, pointing out IPhone 4 is not 4G (which is true) and then listing a bunch of T-Mobile phones that have "4G" in the names and claiming they have the largest 4G network in the US. News flash T-Mobile -- those phones are not 4G either, and you have a fast 3G network, but no 4G whatsoever.
* Second side note -- AWS is 1700mhz phone to site, and 2100mhz site to phone. This is not the European 2100mhz band, but I think the 2100mhz site-to-phone was selected intentionally so a world phone would not have to have so many antennas.
What about the technology?
Sprint is CDMA, t-mobile GSM. A combined company and technology would make sense how?
More standards in the soup pot
So Sprint is CDMA, and they acquired Nextel which uses iDEN, and that went over so very well indeed. Sprint is still struggling with the legacy iDEN network as there are dedicated users who do not want to switch, and they don't plan to shut it down until 2013. Of course, this task is made easier by the former Nextel customers who fled due to Sprint's customer support.
And, as we know, Sprint decided to do their own thing and use WiMax for 4G and not LTE, thinking they'd get a jump on the market as LTE was still gelling. But delays in building out their network and lukewarm hardware support (everyone else is going LTE, why make a WiMax handset?) has kept growth slow. And with Verizon and AT&T building out LTE, and able to select from more handset offerings, now Sprint is facing an increasing headwind in attracting users to their 4G network. Enough that Sprint has increasingly warmed up to the possibility of switching to LTE in their public comments. I think it is inevitable that they will, and fortunately for them a lot of WiMax and LTE base station gear has commonalities, so hopefully they'll be able to upgrade their netwoek without a total forklift replacement. But that's still going to be a royal mess.
Now they're looking at acquiring T-Mobile, a GSM carrier. That seems to be just what Sprint needs, yet another network protocol in the mix.
From a technical standpoint, if that were to happen, migrating their network to GSM+LTE would seem to make the most sense. Verizon is on a trajectory toward CDMA+LTE with designs on pure LTE down the road. Going GSM+LTE would give Sprint access to a wide variety of handsets, and with Sprint's bandwidth they might be able to move away from AWS and have some commonality with the handsets used by AT&T.
Of course, Sprint is larger than T-Mobile and they have more CDMA users than T-Mobile has GSM users, so if one network had to go it'd probably be GSM. That'd leave GSM users without a lot of options - AT&T would be the only major US GSM carrier. But how long would this take? Sprint will be dealing with the iDEN cutover for the next couple of years, do they need to add a GSM migration on top of that? And if they do come to their senses and switch to LTE for 4G, how will that migration work?
Are we going to see one carrier juggling CDMA, iDEN, GSM, WiMax, and LTE? And that's leaving out HSPA+, EVDO, etc. And even if Sprint & T-Mobile were to merge the resulting Frankencarrier would have roughly 27% of the US market, they'd still be number three.
Sprint - Jack of all Trades, Master of None
T-Mobile today at least has the advantage of being seen as kind of the scrappy underdog with innovative pricing and some nice hardware - and they don't load up their phones with the bloatware the other carriers lard on their units. On the Android front they've also been good about keeping their handsets hack friendly without locked down boot loaders and the like. And while they have bent the truth pretty sharply with their '4G' claims, their HSPA+ network has actually tested out faster than the 'real' 4G WiMax/LTE network performance from other carriers today. The big caveat is the limited coverage area. As the LTE networks are upgraded I know HSPA+ will be left behind, but right now it is fast if you have the coverage. (I put 'real' in quotes because of the controversy over calling today's networks '4G' when the speeds are still no higher than the high end '3G' speeds.)
I switched from AT&T to Verizon when the Droid launched. I'm up for renewal in July, and I will seriously consider switching carriers again to get the best handset. My Droid is rooted and running a custom ROM, which Verizon is making increasingly hard by locking down newer handsets. Depending on the state of things this summer, T-Mobile is actually a tempting carrier due to their policies. But not Sprint.