Jump to content

Contributors to this blog

NewCo needs to drop some of the PCS from MetroPCS

WiWavelength

12,306 views

blog-0806643001349388290.jpg

by Andrew J. Shepherd
Sprint 4G Rollout Updates
Friday, October 5, 2012 - 8:00 AM MDT

 

Unless you have been under a telecom rock the past 48 hours -- or stuck in the boonies with only a GSM device (I kid, I kid) -- you have read that T-Mobile USA and MetroPCS have agreed in principle to a complicated reverse merger arrangement that would create a combined carrier, at least provisionally called NewCo. Now, Sprint has jumped back into the fray, this after Sprint's executive leadership had readied a bid for MetroPCS earlier this year but was vetoed by the board of directors. Sprint's motivations for pursuing a counter bid could be multifold.

  • Sprint could actually be trying to acquire MetroPCS, feeling a sense of urgency that it did not this spring. Plus, Sprint's perception on Wall Street has improved dramatically during the past few months, making a merger a more financially palatable prospect.
  • Sprint could be attempting to force T-Mobile to sweeten its offer for MetroPCS, potentially costing competitor T-Mobile additional financial resources.
  • Sprint could be trying to gain some concessions in order to allow the merger to proceed.

That last possibility is what this article will explore, namely, that NewCo would agglomerate an egregious amount of PCS 1900 MHz spectrum in several markets in which Sprint also happens to be a bit PCS spectrum shy. By throwing its own hat into the ring, Sprint should pressure NewCo to divest excess PCS spectrum to Sprint voluntarily. Alternatively, Sprint could lobby the FCC, oppose the merger and its transfer of spectrum licenses, and try to get some mandated divestitures that way.

To illustrate, MetroPCS currently operates in at least some PCS spectrum in 10 major markets. The linked spreadsheet below compares NewCo's potential PCS A-F block spectrum holdings to Sprint's current PCS A-F block spectrum holdings in those 10 markets.

 

akxx8y.png

 

https://docs.google.com/spreadsheet/ccc?key=0ArY31Mr219-ydE1tRVdJS19ocjBZXzVibk01Wm5wLWc&usp=sharing

In those 10 markets, Sprint holds 20-30 MHz of PCS A-F block spectrum, while NewCo would have 35-60 MHz of PCS A-F block spectrum, including 50-60 MHz in four of the markets. Considering that 60 MHz represents fully half of the total 120 MHz bandwidth of the traditional PCS band, that is an outrageous amount of PCS spectrum -- especially for a carrier that is hitching its LTE wagon to AWS, not PCS. Even AT&T would blush at acquiring that much spectrum within a given band. Keep in mind, too, that this analysis does not take into account the 40-60 MHz of AWS 2100+1700 MHz spectrum that NewCo would hold in those same 10 markets, including 50-60 MHz in all but Atlanta. And that 50-60 MHz would be even more than half of the total 90 MHz bandwidth of the AWS band.

Furthermore, T-Mobile has made it known that it intends to pare down its exclusively PCS GSM/GPRS/EDGE spectrum utilization to 10 MHz per market, refarming its remaining PCS spectrum to W-CDMA/HSPA+ in a desperate attempt to attract unsubsidized iPhone users. The Dallas Region Case Study graphic from the NewCo investor presentation corroborates this plan. Moreover, the graphic shows how NewCo plans to operate DC-HSPA+ (20 MHz) for at least the next three years in parallel on both PCS and AWS plus 15-20 MHz FDD LTE on AWS -- an unnecessarily redundant, inefficient strategy.

25z0d1c.png

In short, NewCo does not need as much PCS spectrum as it is set to acquire. Otherwise, it is just as much a spectrum glutton as are VZW and AT&T. So, here is the solution. In Atlanta, Jacksonville, Miami, Sacramento, and San Francisco, NewCo should preemptively choose to or be required to divest 10 MHz of its accumulated PCS spectrum. Sprint would be the obvious buyer, as that would increase its PCS A-F block assets to 30 MHz in those markets. Meanwhile, NewCo would still retain 35-50 MHz of PCS in those same markets, plenty of spectrum for 10 MHz of GSM, 10 MHz of HSPA+ or even 20 MHz of DC-HSPA+, and 10 MHz of CDMA1X/EV-DO for MetroPCS legacy.

 

Sources: FCC, MetroPCS

  • Like 12


25 Comments


Recommended Comments

I think T-Mobile would try to hold on to 40MHz, so giving up 10 in Sacramento might be a tough sell.

Share this comment


Link to comment

Awesome article, I'd love to see a potential one of Sprint + Metro PCS, would be interesting speculation.

Share this comment


Link to comment

Wouldn't the accumulated 1900 MHz PCS spectrum be a huge win for Sprint since that's what they're deploying their LTE on?

  • Like 1

Share this comment


Link to comment

I'm under the impression that LTE Advanced capable devices will be able to use multiple bands at the same time as well as HSPA/LTE at the same time, imagine using all that bandwidth.

Share this comment


Link to comment

Wouldn't the accumulated 1900 MHz PCS spectrum be a huge win for Sprint since that's what they're deploying their LTE on?

 

Some of MetroPCS's PCS 1900 MHz spectrum would be a "huge win for Sprint. But I would not want Sprint to acquire all of it because then Sprint would have, for example, 50 MHz of PCS A-F block spectrum in Atlanta and Miami. And as I chide NewCo a little bit in this article, the shoe would then be on the other foot. Sprint would have far more PCS spectrum than it honestly needs.

 

AJ

Share this comment


Link to comment

Is there an article somewhere that expalins spectrum in layman terms in regards to the carriers, what they are holding, and the uses of that spectrum.

Share this comment


Link to comment

AJ does this article include the PCS G block.

 

No, I tried to be clear in the article and spreadsheet that they reflect only PCS A-F block spectrum holdings. I specifically excluded the PCS G block for several reasons. It is compensatory spectrum that Sprint received for its SMR 800 MHz spectrum given up in public safety rebanding. It is effectively proprietary to Sprint and requires a separate band/band class, so it is of lesser value to other carriers. And it contains no 2G/3G operations, while this discussion is largely about how much PCS A-F block spectrum is needed to manage a transition from CDMA1X/EV-DO or GSM/W-CDMA to LTE.

 

AJ

  • Like 1

Share this comment


Link to comment

AJ does this article include the PCS G block.

 

It omits the G block. Sprint has PCS G and SMR in addition to PCS A-F and T-Mobile has AWS. Guess which has more spectrum?

Share this comment


Link to comment

It omits the G block. Sprint has PCS G and SMR in addition to PCS A-F and T-Mobile has AWS. Guess which has more spectrum?

 

In most markets, T-Mobile has more spectrum, especially following its AWS transactions this year with AT&T, Leap, VZW-SpectrumCo-Cox, and C Spire. MetroPCS, in my opinion, would push T-Mobile over the top to AT&T like excess, particularly in PCS, in many of the affected markets. The only way that Sprint has greater access to spectrum would be to count Clearwire and its holdings under the Sprint umbrella.

 

AJ

Share this comment


Link to comment

Why not just have a three way merger, and commit to GSM/LTE as the long term winner in Sprint's technology path?

Share this comment


Link to comment

Why not just have a three way merger, and commit to GSM/LTE as the long term winner in Sprint's technology path?

 

GSM certainly is not the winner. It is just living out the last of its days in the old folks home.

 

But, yeah, LTE is the long term winner for data. For voice? In the long term, yes, probably. In the medium term, that remains to be seen.

 

Because the W-CDMA operators bought into a Eurasian-centric standard that requires an ungainly amount of spectrum per carrier, they have greater incentive to move to VoLTE. They cannot really justify operating a fully 10 MHz W-CDMA carrier or two just for voice alongside LTE.

 

But the CDMA2000 carriers have options, since a 2.5 MHz CDMA1X carrier requires a quarter of the spectrum of a W-CDMA carrier. And for those like you, Ryan, who live in more rural areas, you should hope that CDMA1X sticks around for a good long while. VoLTE may be a boon to operators but will likely be a step down for rural subs, who may come to miss the gold standard consistency and reliability (soft handoff for the win) of CDMA1X voice.

 

AJ

  • Like 2

Share this comment


Link to comment

I was speaking of the GSM evolutionary path, not GPRS/EDGE, which I agree has outlived its useful life. As far as the rural carriers, I don't see a rush for backing off CDMA for voice. It's useful as a backup where VoLTE won't easily penetrate. I don't see that being many areas though.

 

Also, there's no need for soft handover with the network infrastructure of LTE. Not that much of a loss by this point.

 

http://parklands-wir...r-exist-in.html

 

Now if you deploy like Sprint and have a RRU, I would think that coverage on VoLTE would equal CDMA 2000 on the same spectrum without a RRU.

Share this comment


Link to comment

Also, there's no need for soft handover with the network infrastructure of LTE. Not that much of a loss by this point.http://parklands-wir...r-exist-in.html

 

Now if you deploy like Sprint and have a RRU, I would think that coverage on VoLTE would equal CDMA 2000 on the same spectrum without a RRU.

 

I am not a degreed engineer, but I disagree with Parklands' "Engrish" explanation. Propagation/penetration is not the issue as much as is the very nature of the airlink.

 

CDMA1X is a great airlink for transmitting a small data rate spread across a much wider bandwidth. That makes it ideal for voice, which is always a small data rate. Furthermore, that spreading ratio allows CDMA1X to operate to very low signal levels. And that, along with soft handoff, helps greatly in rural areas with low site density.

 

On the other hand, much empirical evidence, thus far, has shown that LTE is an airlink best suited for providing high data rates with high site density and strong resistance to multipath. In other words, it is a great urban area airlink.

 

And be assured, providing service to urban areas is what is driving the transition to LTE. Rural areas are peripheral. If they experience some collateral damage, so be it. At least, that seems to be the LTE attitude.

 

AJ

  • Like 2

Share this comment


Link to comment

Now if you deploy like Sprint and have a RRU, I would think that coverage on VoLTE would equal CDMA 2000 on the same spectrum without a RRU.

 

Based on our observations, both Sprint LTE and Verizon LTE is unusable with a weaker signal than -95dBm RSSI. That's a full 10dBm less than 1x will do from the same site. That is a lot of lost square mileage of coverage in rural areas.

 

Robert

  • Like 1

Share this comment


Link to comment

I am not a degreed engineer, but I disagree with Parklands' "Engrish" explanation. Propagation/penetration is not the issue as much as is the very nature of the airlink.CDMA1X is a great airlink for transmitting a small data rate spread across a much wider bandwidth. That makes it ideal for voice, which is always a small data rate. Furthermore, that spreading ratio allows CDMA1X to operate to very low signal levels. And that, along with soft handoff, helps greatly in rural areas with low site density.On the other hand, much empirical evidence, thus far, has shown that LTE is an airlink best suited for providing high data rates with high site density and strong resistance to multipath. In other words, it is a great urban area airlink. And be assured, providing service to urban areas is what is driving the transition to LTE. Rural areas are peripheral. If they experience some collateral damage, so be it. At least, that seems to be the LTE attitude.AJ

 

Release 10 and CoMP add a lot of the soft handover features anyway. By the time LTE gets to the sticks where I live in 2013, I'm not sweating it much. CDMA 2000 will still be important for a while but by the time it gets shut down, you'll have a lot of the issues with soft handoff on LTE solved.

 

I'll probably be gone from here by then, but I'm not worried about it by the time CDMA sunsets, which VZW won't even do until 2020.

Share this comment


Link to comment

At least nobody is merging with ATT! I dont like them.

 

Funny that not as many people complained with the Verizon SpectrumCo AWS. Verizon passed AT&T for the national spectrum lead with that purchase. People have a hard time grasping the nebulous, and the SpectrumCo deal was nebulous to say the least. AT&T can get 700 B spectrum out of the deal however. That's still not going to get them up to Verizon's total.

Share this comment


Link to comment

Funny that not as many people complained with the Verizon SpectrumCo AWS.

 

The wireless element of the VZW-SpectrumCo-Cox deal was less controversial because, unlike the AT&T-T-Mobile merger, it did not remove an actual competitor from the market. The wired side of the deal that basically formed a cabal between VZ and Big Cable got plenty of opposition, rightly so.

 

AJ

Share this comment


Link to comment

I like the proposal, but lets not forget about PCS constrained T-Mobile markets with the huge amount of POPs like NYC. I'm pretty sure T-Mobile would much rather agree to swap than sell, unless its forced to.

Share this comment


Link to comment

I like the proposal, but lets not forget about PCS constrained T-Mobile markets with the huge amount of POPs like NYC. I'm pretty sure T-Mobile would much rather agree to swap than sell, unless its forced to.

 

I can appreciate that T-Mobile *feels* PCS 1900 MHz spectrum constrained in some markets. But I can offer a fairly persuasive counterargument.

 

First, in NYC, the big four carriers control all 120 MHz of the traditional PCS A-F block band. Here is the PCS spectrum holdings hierarchy:

 

VZW: 40 MHz

AT&T: 30 MHz

Sprint: 30 MHz

T-Mobile: 20 MHz

 

Sprint has no spare PCS spectrum to swap T-Mobile in NYC. If Sprint did, then Sprint would be the one left with 20 MHz. Most likely, T-Mobile would have to look to VZW.

 

However, the PCS spectrum holdings look reasonably proportional as they stand now. By significant margin, T-Mobile is the little guy among the big four. It should not require as much as spectrum as the larger carriers do. So, the 40:30:30:20 split right now is not out of line.

 

Furthermore, keep in mind that T-Mobile (via NewCo) would control 50 MHz of AWS 2100+1700 MHz spectrum in NYC. That is over half the AWS-1 band. So, T-Mobile would not be spectrum shy overall in NYC.

 

That T-Mobile might desire but lack specifically more PCS spectrum is, in many ways, a problem of its own doing. To go into detail about those many ways is beyond the scope of this short response, but a brief listing of those reasons would include sticking with the Eurasian centric 3GPP ecosystem, transferring PCS spectrum in NYC to Cingular to break up their GSM joint venture, and realigning its network in a desperate attempt to attract unsubsidized iPhones.

 

Lastly, the most compelling point of my counterargument is this parallel. If one person is hungry, overfeeding another person does not counteract the first person's hunger. T-Mobile may wish for more than 20 MHz of PCS spectrum in NYC, etc. But amassing 60 MHz of PCS spectrum in Atlanta, 50 MHz in Miami, 50 MHz in San Francisco, etc., does not help T-Mobile in NYC. Instead, T-Mobile (via NewCo), still the little guy, just ends up with an excess of PCS spectrum in those markets. And that does not benefit anyone.

 

AJ

Share this comment


Link to comment

Nothing a spectrum swap can't solve, AJ. T-Mobile USA has a proven record of swapping spectrum to better their efficiency. I'm sure they'll look at way to deal with any discrepancies you or anyone else may raise. They can trade excess spectrum in any of those markets for spectrum in NYC. Boom, done.

 

Also, I don't know about the validity of shutting down AWS DC-HSPA until you have a while to replace those phones that only support DC-HSPA on AWS with LTE handsets. That's going to take a while.

Share this comment


Link to comment

Nothing a spectrum swap can't solve, AJ. T-Mobile USA has a proven record of swapping spectrum to better their efficiency. I'm sure they'll look at way to deal with any discrepancies you or anyone else may raise. They can trade excess spectrum in any of those markets for spectrum in NYC. Boom, done.

 

I would not count on that, Ryan. I know that we are focusing solely on NYC to the exclusion of other affected markets, but NYC really is a poster child. And the way that the spectrum holdings lay out in NYC, I am not at all confident that a voluntary spectrum swap could be worked out.

 

To explain, T-Mobile in NYC holds 20 MHz of contiguous PCS spectrum, split between the PCS A and D blocks. The only two carriers that could offer T-Mobile an additional 10 MHz contiguous to its current 20 MHz are AT&T (PCS A block) and Sprint (PCS B block). As I listed in my response to Milan, though, both AT&T and Sprint have 30 MHz of PCS A-F block spectrum in NYC, and neither would be likely to reduce its holdings to only 20 MHz.

 

So, as before, that would leave only VZW as a potential spectrum swap partner and would raise a few issues, namely that the rich kid, VZW, would get even richer in another market, such as Atlanta, where it holds only 10 MHz of PCS spectrum but 25 MHz of Cellular 850 MHz, 22 MHz of Upper 700 MHz, 12 MHz of Lower 700 MHz, and 40 MHz of AWS 2100+1700 MHz spectrum.

 

Also, I don't know about the validity of shutting down AWS DC-HSPA until you have a while to replace those phones that only support DC-HSPA on AWS with LTE handsets. That's going to take a while.

 

Yes, that would be a problem if the problem were to actually exist. It does not. Even though T-Mobile started offering DC-HSPA+ devices before it publicly announced plans to refarm PCS spectrum for W-CDMA, all of its DC-HSPA+ devices are already dual band AWS/PCS.

 

AJ

Share this comment


Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×