Seeking Alpha
About this author:

Interesting data point from CTIA presentation by AT&T's De la Vega-- AT&T's (T) research shows that just 3% of all subscribers use 40% of the data. That's not necessarily the same thing as capacity. These users consumer 13 times the data of the "average smartphone customer" yet are 1% of post paid subs.


He also spent a lot of time arguing that there is a spectrum shortage and even if this spectrum was reallocated quickly, they couldn't put it into use quickly (most probably truthful).

However, what about the impact of the upgrade to LTE? That's going to have a dramatic impact on capacity. Moreover, one has to ask is Verizon's (VZ) subscriber base as skewed as AT&T is? Most probably this is the "IPhone effect". As more phones are launched competitively with the IPhone, will the distribution of data usage spread out?

The big sandbag comment? "We have to manage the network so the few do not crowd out the many". It's the pretext for an argument that AT&T should be able to charge for tiered data.

That's a good point, but if we had a competitive wireless market with 2 more robust entrants it wouldn't be so easy for AT&T to under-invest in its data network. AT&T and Verizon have been slow rolling the capex build on its network for years, most due to retarding the phones it allows on its network. How about diverting a little bit of that massive cash flow to building out LTE?

Print this article with comments

This article has 19 comments:

  •  
    Building out LTE will probably not be a big priority for AT&T in 2010. LTE only increases the bandwidth between the cell tower and the end user. AT&T's existing GSM already does a good job of that and there are no (that is absolutely none) LTE handsets in widespread use so building out this capacity is quite a long ways ahead of the curve. What AT&T needs to do, and is hopefully doing, is improve their GSM coverage in places like NY where everybody seems to be complaining. More importantly they need to increase their backhaul capabilities to clear up the congestion north of their cell towers. That is the real crisis facing companies rolling out web-friendly devices like the iPhone, Pre, etc.

    Verizon has a different but equally daunting challenge. Their CDMA based EVDO network will NEVER handle web-friendly devices so they have no choice but to build out LTE. The stumbling block is fail-over when their subscribers are out of LTE areas. The only possible solution I see in 2010 would be a comprehensive roaming agreement with T-Mobile which would allow the next generation of multi-band (LTE-GSM-EDGE) handsets to fail over to T-Mobile while Verizon's builds out their LTE network.
    Oct 08 06:20 PM | Link | Reply
  •  
    The obvious solution is to charge based on data usage. Imagine if customers paid a flat rate for electricity. Want to imagine what the effect would be on tghe size of system needed?
    Oct 08 06:50 PM | Link | Reply
  •  
    I think Verizon is fine with their present EVDO capacity. Not many high end users for them or AT&T. As long as VoIP calls over 3g are limited LTE really isn't critical.
    Oct 08 08:06 PM | Link | Reply
  •  
    What incentive do any of the Wireless carriers have to build out now? The FCC's stated goals have already diminished the Telecoms value (based on declining market value and analysts reports). The radical net neutrality promoters want unlimited capacity, no traffic governance and to be free of any cost. With no clear path to achieve a return on investment , why make the investment at all?
    Oct 08 09:07 PM | Link | Reply
  •  
    Both AT&T and Verizon have spent billions on spectrum specifically for LTE already (C.I.P.). LTE promises better performance, but we shouldn't forget that it also promises a reduced cost per served MB of data traffic: compared to both legacy technologies: EVDO and HSPA: (efficiency). Next, the anticipated data volumes from open application stores, open access mandates and the addiction to "all you can eat" data plans have got to scare the daylights out of anyone looking to maintain some sort of profit per data subscriber. With this in mind: Capital in Progress, Efficiency and anticipated usage per unit revenue: it's silly to assume that U.S. LTE deployments are somehow throttled by some intention to under-invest in the network. The "massive cash flow" you refer to may quickly become negative if the industry can't figure out how to serve more data with less cost very soon. Leaving spectrum capital on a shelf is not going to help either operator either. The fact is that AT&T and VZ are pushing LTE timelines harder than anyone else in the world. This push isn't about boasting outrageous bitrates, it is to reduce the slope of declining data margins. So from the AT&T and VZ perspective, it's not a stretch to assume that both would begin LTE rollout faster if the technology was ready today. You don't have to take my word for this, observe the mention of "efficiency" in the announcements below.

    www.atis.org/lte/docum...
    news.vzw.com/news/2007...

    Customers care little about efficiency, but investors do......
    Oct 09 07:22 AM | Link | Reply
  •  
    All utilities charge a minimum rate and then additional charges for excessive usage. Why should it be different here. Telecom spectrum usage differs little from water, electricity, gas and other utilities. It is only proper that the more you use, the more you pay.
    Oct 09 09:21 AM | Link | Reply
  •  
    I don't agree with the last post that LTE is more efficient. Higher data rates at the cell site will need to be supported by increased backhaul capacity which is additional cost. The initial cost of deploying a LTE "overlay" will be very expensive while 3g is deployed and capable of handling current and near term data loads. I don't think there is any urgent need for the big service providers to build out 4g network other than maybe bragging rites.
    Oct 09 09:38 AM | Link | Reply
  •  
    Just like the taxes collected to build the high speed fiber optic network the money will be used for bonuses and the company will change hands the new owners will say "What Fiber Optic Network ?".
    Oct 09 10:00 AM | Link | Reply
  •  
    I'm impressed with everyone's jargonese but what is LTE?
    Oct 09 01:04 PM | Link | Reply
  •  
    iggy, i hope you are joking, google it

    LTE delivers far more data per dollar, it is much cheaper from that perspective (~85% cheaper than HSPA+, if i remember correctly). the fiber towers, however, are about 3X as expensive as copper towers, but coming down fast as demand increases. backhaul must be completed first, the pipe running from the tower to the carrier cannot handle multi Tb capacity; the equipment on the edge and most of the metro cores cannot handle the traffic needed either. right now, carriers are band-aiding by bundling Tb cables, strictly temporary

    T, VZ etc have budgeted billions over the next five years to build out LTE. RIGHT NOW the software to run LTE equipment is not even completely ready, that is why neither T nor VZ are agressively building it out; they are working with their networking vendors to sort out the software. STAR and MOT are building out LTE in japan, they def have a head start.

    what percentage of people want a 70-100/mo data bill, esp when they already have BB at home? 30%? i do not know, but long term IMO that is where we are headed. my guess is that the data will be ready before the demand and those stocks will be soft for a few years, but watch the data stats. (long T, MOT, STAR)
    Oct 09 02:09 PM | Link | Reply
  •  
    I love reading the comments here. Anyone ever hear about WiMAX? (Competitive wireless technology to LTE)

    Better look it up because that's going to compete with LTE and before you make any assessments of "how fast" LTE is going to be, you should understand WiMAX and how it is used.

    As to some of your questioning about speed and downloads - it's ALL ABOUT SPEED when it comes to downloading video apps.

    We are now at a point where more and more who have smartphones are downloading not one or two gigabytes of info a month - it's more like 20, 30 even 40 gigabytes a month.

    As to why put in highspeed wireless - that's what other countries are doing and some have much more sophisticated networks than we do. Check out South Korea. It's all about the infrastructure (which includes network infrastructure) supporting economic development.

    This is the "GIgabyte Generation" and if you are thinking about "picking stocks", better read up on the whole industry and not just throw around terms and acronyms.

    Look at wireless going from 100Mbps to 1Gbps to the end-user. If a carrier isn't talking about that type of speed, they don't have a handle on the emerging market.
    Oct 09 06:39 PM | Link | Reply
  •  
    T along with most of the rest of the world is GSM. They have an easy transition to LTE. Vz, with EVDO already deployed opt's out of EVDO REV C(EVDO 4g) and goes with LTE. WIMAX may not be dead, but I don't think it can compete..
    Oct 09 09:00 PM | Link | Reply
  •  
    LTE = Long Term Evolution - which is one technology that the carriers are looking at. It is proprietary standard which is the way most of the traditional carriers like to offer things.

    WiMAX is based on an IEEE standard 802.16e and is actually more ahead in actual implementation than LTE. It's also OPEN architecture because it's a "standard" and NOT proprietary.

    It's already implemented in countries around the world. What that means is that it will be able to roll out faster than LTE here in the US and that is a serious issue with those backing LTE.

    rf44g - you are WAY OFF the mark when you say WiMAX can't compete. It's already up-and-running. It is the threat of WiMAX opening up in the US market that is accelerating work on LTE which is farther behind in actual implementation.

    AND - your comment
    "I don't think there is any urgent need for the big service providers to build out 4g network other than maybe bragging rites."
    is ludicrous - if WiMAX is going into one of their US markets, you can be sure they are trying to rush LTE because as of now. WIMAX is way ahead as far as implementation and a 4G network is SUPERIOR as to downloading video by a factor of at least 4 to 1. (if you read my previous post)

    rf44g - it might be helpful if you went to one of the conferences like 4G World where you could actually see all this technology at work and hear what developers -- for BOTH LTE & WiMAX devices are saying. I don't understand how you formed your opinion.

    ALSO for some of the other commenters - WiMAX was designed originally as a backhaul system so it has more than adequate bandwidth capabilities.

    Where are all of you getting your facts??? If you could call them that.

    And "LTE provides more data per dollar"????

    The big push is video apps and the ability to handle large downloads - I mentioned before we are looking at users downloading 20, 30 and 40 gigabytes per month and that is just going to grow.

    Both approaches are viable but I would NOT be discounting WiMAX out at all especially when it is up-and-running in other countries and starting to roll out in the US. (with CLEARWIRE)

    It is ahead of LTE as being a new choice for consumers (LTE is still on the drawing board in some respects and the threat of WIMAX has actually hastened those pushing LTE to get it done.)

    If WiMAX hits a US market first, it will eat the lunch of an incumbent carrier who can't roll out LTE for another one or two years. Believe me - they are SCRAMBLING to get LTE up.

    Another big plus is that WiMAX is an open standard which means more people can develop applications and therefore create a more feature-rich environment.

    If you have an iPhone you may understand that. Look at all the developers who have created special apps for the iPhone that you can buy for a couple of bucks. Think of that same approach -only on the network side (as well as the device side) With LTE, no offerings are available unless the carrier offers it - which is the way they like it.

    CLEARWIRE's WiMAX build-out plans for 2009 including San Antonio, TX; Austin, TX, Milledgeville, Georgia, Raleigh, NC; Greensborough, NC, and Salem, OR. Let's see what happens to the competition in those markets.

    And before you put a thumbs-down on THIS comment, do a google search on LTE vs. WiMAX comparison 2009 and read some of the articles that pop up or just go to wimax.com to get some insights.

    Hopefully, you will walk away with a much broader perspective which helps everyone.


    Oct 09 11:16 PM | Link | Reply
  •  
    James, thanks for the comment. I will try and attend some conf in the near future, maybe I'll look at it differently after that. But hasn't Sprint already rolled out WIMAX in some major markets? I haven't heard much about it. Anyway, thanks for the comment..
    Oct 10 12:12 AM | Link | Reply
  •  
    WiMax is complimentary to LTE, not competitve; WiMax is even cheaper per bit than LTE, but my engineer friends say it is less flexible. My understanding is that WiMax spectrum is clean now but destined to become dirty as more users fill it up-- precisely the problem with it being open, and precisely the reason it is great for rural and third world applications

    WiMax is widely deployed already, it is not being adopted quickly; Europe, Japan and Korea have widely deployed WiMax, as well as some cities in the US, what has it done for carrier profits? not much.

    for those of you new to wireless investing, remember that there was a boom in 2004-5, investors allocated billions in networking to wire up cities and we all took a deep cold bath a couple years later: the bandwidth was there and people just did not use it, period. those of you expecting a spontaneous revolution (rather than steady gowth) must look at Europe and APAC, it is simply not happening.
    Oct 10 02:42 PM | Link | Reply
  •  
    Using internet bandwidth is not like using electricity at all. It is more like using the roads. Imagine if you were charged for registration fees based on how much you drive.


    On Oct 08 06:50 PM RLLH wrote:

    > The obvious solution is to charge based on data usage. Imagine if
    > customers paid a flat rate for electricity. Want to imagine what
    > the effect would be on tghe size of system needed?
    Oct 10 09:20 PM | Link | Reply
  •  
    I have a Vz wirless card for my laptop. I pay 70/month and they cap me at 5Gbytes/month. Over that I pay a premium. My guess is that T does the same. It's not unlimited usage like ADSL or a cable modem.
    Oct 11 12:48 AM | Link | Reply
  •  
    WISDOM and others.

    You are being misinformed is all I can say. LTE is BEHIND WiMAX as far as a technology that can be turned up right now.

    WiMAX is complimentary??? (I think you mean complementary) That is NOT true. Where are you getting your facts???

    As for WiFI (what you point out to as 2004-2005) that was technology that was NEVER designed to handle major cities. If you talked with people that REALLY knew the technology they would have told you that. The bandwidth was NOT there with that technology.

    If you read all my articles going back to those years, I always said it was never the right fit.
    Oct 12 04:38 PM | Link | Reply
  •  
    VZ is such a rip-off. I know T gets more criticism but my new Iphone hasn't any problem downloading probably gigs of stuff. For only 30 bucks a month. Maybe slow once in awhile. But I have the home wifi too. As a long time VZ user, I did enjoy their wider range of calls being accepted and good people on the service side but their ancient proprietary system and refusal to have anything "open" on their phones..(even ringtones had to paid for through them) and the generally mostly Motorola phones that had planned obsolescent batteries and which never lasted the two year contract. I know T sucks for many but being the Iphone provider with roll-over minutes and unlimited data plans is a huge factor in my switching. And their stock w/ good dividends and a low price looks like a buy right now.

    70 bucks for only 5 gbytes a month....!!! holy bat shit!

    I am not crazy about Apple's proprietary monopolistic practices either but they did the right thing going with a international GSM type of network... and thanks for insight on the hardware/tower/software side of things esp. Carlini!! more open is good imho.
    Oct 12 04:56 PM | Link | Reply