I found yesterday’s news quite hard to believe. Even though AT&T / T-Mobile merger was talked about and discussed many times in the past, more recently it has faded primarily due to AT&T’s size and concerns about how US government would scrutinize such a deal. Considering the initial feedback, those concerns seem to be very valid. Probably every consumer protection group in the US will end up petitioning DOJ and FCC. Similarly many competitors including Verizon and Sprint will object to at least parts of this deal. In the last 4-5 years, T-Mobile has steadily moved in the direction of serving lower income groups who happen to be more cost-concious and less likely to be part of employer phone plans. It is likely that there will be a lot of opposition from those representing T-Mobile customers. I think the US government is going to look at this deal at geographical granularity and will make approvals on a Major Trading Area (MTA) / Business Trading Area (BTA) basis.
Following tables are from FCC’s 14th annual report on mobile wireless competition which was published in June 2010. First table shows the spectrum assets in the US.
Second table from the FCC mobile wireless competition report shows the distribution of spectrum (MHz-POP) among wireless operators. This table was prepared before AT&T’s purchase of Qualcomm’s 700 MHz assets and FCC’s waiver for LightSquared to build a terrestrial-only system (provided that they can resolve the concerns about GPS jamming.)
- Over 35% of 700 MHz spectrum (once the Qualcomm purchase is completed)
- Over 42% of cellular spectrum
- Over 45% of PCS spectrum
- Almost 40% of AWS spectrum
Considering Sprint’s SMR spectrum is stuck with legacy IDEN until 2013 and Clearwire’s use of BRS/EBS is not really going anywhere until they revert to LTE, and commercialize TDD version of LTE, there is not much competition left to AT&T and Verizon duopoly.
Following two graphs show the MHz-POP share before and after AT&T / T-Mobile merger based on the figures from FCC 2010 report.
Before the merger:
After the AT&T / T-Mobile merger:
I find it quite hard to believe that this merger will get a blanket approval. Instead it will be reviewed on a license block basis. T-Mobile doesn’t add much to AT&T in terms of coverage. On the other hand it will have a very significant impact in terms of capacity. This is primarily due to the much lighter loading of T-Mobile’s spectrum with respect to its customer base. Following figure shows the customer number per MHz-POP for the big seven operators. I excluded Clearwire from this graph since they are in a build-out phase with very few customers as opposed to their vast holdings of spectrum.
Above chart shows that T-Mobile’s current loading of its spectrum assets (MHz-POP) is about 35% lower than Verizon which is the market leader. Interestingly, T-Mobile’s usage is even lower than MetroPCS and US Cellular. Only Leap has a lower utilization of its spectrum assets compared to T-Mobile. To put it another way, T-Mobile has enough spectrum assets to serve almost 50M subscribers to have a similar utilization as market leaders Verizon and AT&T. Ultimately that deficit of about 15M subscribers forced DT’s hand to sell T-Mobile while making it an equally attractive target for AT&T. Following chart that shows the spectrum utilization after AT&T / T-Mobile merger (assuming all assets are merged) takes place.
This chart shows the lighter loading of spectrum assets for the combined company. This new spectrum position will allow AT&T to acquire another 12-15M subscribers to achieve parity with Verizon in terms of spectrum usage.
Alternatively AT&T might be more focused on the growth of data usage on their network and they simply wanted to get additional spectrum (currently used rather lightly) assets to close that gap until technologies such as heterogenous networks and intelligent WiFi traffic off-loading become having more impact.
Coming back to the title of the article; I strongly believe iPhone is the main culprit behind the merger. Here is why:
- T-Mobile’s growth has come to a halt after the introduction of iPhone. Without iPhone’s impact T-Mobile would have easily surpassed 45M subscribers mark by now.
- AT&T’s network woes have become much bigger after they started offering iPhone. Its popularity has impacted the network loading dramatically, i.e., significantly more than what average subscriber numbers show.
- Lack of AWS spectrum support in AT&T’s HSPA+ devices such as iPhone will limit their ability to use T-Mobile’s 3G network in the short run. However, there are multiple ways to deal with that including the support for AWS spectrum in all AT&T devices including iPhone 5 immediately; starting to re-farm T-Mobile’s PCS spectrum for 3G; accelerating T-Mobile customer’s handset updates to include 850 MHz (cellular) support that will eventually be the 2G spectrum for the combined company.
Some other predictions/questions:
- It’s hard to believe Verizon will sit and silently watch all these. If this deal can be approved, what is there to prevent Verizon’s purchase of US Cellular, Metro PCS and Leap Wireless?
- This deal makes Sprint+Clearwire to be the critical element to competition in the US marketplace. This should give more incentive to those who would want to bet on a third alternative. I am not sure if Google or Intel or cable companies still have the same passion as they did few years ago.
- Sprint has to resolve the IDEN saga, shut it down rapidly and start moving on the TDD-LTE path right away. Obviously the fundamental question is whether it will find the funding it needs and build the device ecosystem, i.e., will Apple be convinced to build TDD-LTE devices at 2.6 GHz?
- Cable companies (Comcast, Time-Warner and Cox) has a sizable spectrum asset in the AWS band. It remains to be seen how they will use that in the new world where number of players is reduced to three. Considering Sprint doesn’t own any AWS spectrum, it is hard to see any synergy. On the other hand, Verizon and AT&T are hardly potential partners for cable companies.
- LightSquared may benefit from this consolidation since there will be desire to create nation-wide competition to big three. However, LightSquared still has some technical hurdles as well as a major deployment challenge to go through. Since both Clearwire and LightSquared have lost T-Mobile or AT&T as potential bidders for their spectrum or services, it is more likely that LightSquared and Clearwire will be competing for wholesale business that must come from others (could Apple, Microsoft or Google be ever a customer of such a network?)
- Smaller players such as EchoStar in 700 MHz band will be folded into either AT&T or Verizon.
These are certainly exciting times. Similar to last 15 or so years (period starting with 1995 PCS auctions), US wireless industry will continue to consolidate while generating a lot of new companies and businesses along the way.