Blog Post

PE Firm Plans Open LTE Network to Challenge AT&T and Verizon

A New York private equity firm plans to build a multibillion-dollar 4G wireless network that will cover most of the country by 2015. The ambitious plan by Harbinger Capital Partners relies on deploying a Long Term Evolution network over spectrum owned by a few satellite companies — and would create an open wholesale wireless network available to retail companies, PC manufacturers or anyone who wants to offer mobile broadband.

Last November I wrote that certain satellite companies were visiting Washington hoping to somehow cash in on the 100 MHz of spectrum they collectively have. After reading the FCC order and Harbinger’s plans filed with the FCC on Friday night, it looks like those satellite firms may have found a way thanks to the FCC’s faith in mobile broadband as a means of promoting innovation and competition. [digg=]

A Competitive Threat

The new network will rely initially on 23 MHz spectrum owned by SkyTerra, which is owned by Harbinger, and could later include spectrum from Terrestar Networks, another satellite firm in which Harbinger holds a stake. The Harbinger network could help ensure competition among the major wireless carriers thanks to the conditions the FCC has placed on the spectrum that the private equity firm plans to use as part an agreement to let Harbinger take control of SkyTerra  — namely that SkyTerra has to be a wholesaler, and that traffic from the largest and second-largest wireless carriers in the U.S. cannot comprise more than 25 percent of the traffic over the SkyTerra/Harbinger network. This means AT&T (s T) and Verizon (s vz) could not buy up huge chunks of the network or spectrum to keep others off of it.

Harbinger’s plans to build out a mobile broadband network stretch all the way back to a 2003 order allowing satellite companies owning spectrum in the L and S bands to build out terrestrial networks in conjunction with their satellite networks. The idea was to offer an alternative to the cellular carriers, but the stringent satellite requirement (that the cell companies lobbied hard for) has proved tough. It’s expensive to launch and build satellites (plus build out a terrestrial network), and any phones working on such a system are pretty clunky. Another strike is unless the satellite companies found a willing terrestrial partner, the initial satellite mobile broadband speeds were too slow.

However, the FCC has given appears ready to give the satellite spectrum holders a break by relaxing the satellite aspect of the network buildout allowing Harbinger to consolidate so much spectrum in exchange for guarantees about how fast the terrestrial network will be built out. Plus the FCC will enforce the conditions outlined above to ensure that the SkyTerra/Harbinger network is a real alternative to the cellular carriers.

Harbinger’s LTE Buildout Plans

The planned network would launch before the third quarter of 2011 and cover 9 million people, with trials set initially for Denver and Phoenix. The next milestone is that 100 million people have to be covered by the end of 2012, 145 million by the end of 2013 and at least 260 million people in the United States by the end of 2015. Harbinger said in its statements to the FCC that all major markets will be installed by the end of the second quarter of 2013.

However, the 36,000 base stations that Harbinger plans to use, along with the tower sites, backhaul and other gear associated with a terrestrial network will require billions of dollars. Analyst Chris King at Stifel Nicolaus estimates that Verizon’s LTE network will cost about $5 billion to deploy, and Verizon already has some of its network elements in place. Clearwire (s CLWR) has also spent billions on its network, with analyst estimates ranging from $3 billion to about $6 billion.

Harbinger didn’t talk about it’s capital-raising plans, but a spokesman emailed me the following statement:

Harbinger is in discussions with a number of companies, which we cannot disclose, that provide a variety of services and solutions, but have not finalized anything with any potential partners.

SkyTerra and Terrestar have been searching without success for partners to help build their terrestrial networks for years, with Intel (s intc) once being floated as a potential candidate.

For those spectrum nerds out there, Harbinger says it has access to the following hertz:

At the outset, the network will have no less than 23 MHz of spectrum, consisting of 8 MHz of 1.4 GHz terrestrial spectrum, access to 5 MHz of 1.6 GHz terrestrial spectrum and 10 MHz of MSS/ATC L-band spectrum. Through a cooperation agreement with Inmarsat and associated waivers of the Commission’s ATC rules, by 2013 Harbinger will have access to an additional 30 MHz of ATC spectrum.

By 2011, when the SkyTerra network would launch, the nation will have 4G network coverage from Verizon, Clearwire (s clwr), MetroPCS (s pcs) and AT&T. Clearwire’s cable partners and Sprint seem to be getting some real traction with their customers as resellers of the Clearwire service, an indication that consumers don’t care what mobile broadband provider they buy from, as long as they can get access to the mobile web.

So in looking at that list and realizing the AT&T and Verizon can’t play, and that Clearwire-Sprint has plenty of spectrum to deploy, the most logical partners for Harbinger’s expansion are likely T-Mobile or Leap. My friend Tim Farrar, a satellite industry analyst who saw the Harbinger action coming weeks ago, thinks Harbinger will hit up T-Mobile. I think he’s right.

Images courtesy of NASA

29 Responses to “PE Firm Plans Open LTE Network to Challenge AT&T and Verizon”

  1. Great coverage guys, thanks for all the info. Our blog focuses on hedge fund portfolio tracking and we’ve followed Harbinger’s movements for some time now. Something that also caught our eye (as another commenter mentioned) was their large position in Sprint (S) that they started in the fourth quarter 2009:

    This spectrum play has been in the works for quite a long time and we detailed their Terrestar pick-up back in April of 09

    Will be drafting up a quick article that links to here and to Tim’s for more info. Great stuff & keep us posted with anything else you might find.


  2. The Harbinger network could help ensure competition among the major wireless carriers thanks to the conditions the FCC has placed on the spectrum that the private equity firm plans to use as part an agreement to let Harbinger take control of SkyTerra — namely that SkyTerra has to be a wholesaler, and that traffic from the largest and second-largest wireless carriers in the U.S. cannot comprise more than 25 percent of the traffic over the SkyTerra/Harbinger network.

    I had to sit down to finish that run on sentence. Not only has Om’s site turned into a stream of consciousness diary, the writing has increasingly deteriorated. Do schools teach composition any longer?

  3. Why do you all assume they don’t know what they are doing? This has been put together over the past three years and you’ll be surprised when you hear who’s involved …

  4. Can someone tell me why new spectrum bands can’t be developed? For example, couldn’t someone develop a new spectrum called 10000mhz and build new devices to support this spectrum?
    And isn’t there a way to have cellphones without using spectrum? Fibre optics uses light.

    • Ryan, the issue here are many. A few at a really simplified level are: higher frequency spectrum doesn’t hold information well over long distances or penetrate walls. So there’s less usable spectrum for transferring bits at reasonable power levels in places that people want to do it. Folks are of course studying ways to improve on this, but that gets us to the next point, which is once you have a way to use a different band of spectrum at a higher frequency level, you also need to build out silicon for the radios and get device makers to put it in their gear. Those radios are usually hard to build and expensive at first, so no one wants to buy them until mass market adoption is achieved or ensured.

      As for fiber optics there’s a challenge in terms of line of sight. If you break the light with a wall or a body part, you’ve cut the flow of info. Again, this is super simplified, but many of the commenters here could tell you far more.

  5. Some Guy

    $5 Billion… ha ha… that is less than $150K per site. If they are lucky that would pay for the equipment. Operating cost of the network alone would be $5-$10 Billion per year. They also forgot to figure in the cost of running the company and acquiring customers. It is at least $200 per customer/sub for sales/marketing. I am not even including an equipment subsidy (i.e. VZW claims $350 subsidy on smartphones). So, that would be another $5 Billion to grow the customer base to 25 million subs. You would need at least 25 million subs to support the $10 Billion in annual operating cost. It would take at least 3 years to get the company to build out and get to 25 million subs. Here are the real numbers:

    -$5 Billion for 36,000 cell sites (this way conservative in both cell site count and cost to build)
    -$5 Billion to acquire 25 million subs
    -$30 Billion to operate the network for the first 3 years.
    (I am assuming that any revenue will cover the non-network costs of operating the company)
    5+5+30 = $40 Billion (absolute best case). More realistic it would $60-$80 Billion.

    Of course, the article doesn’t address the fact that none of their spectrum is in LTE or cellular frequency bands. This means more cost and less device selection.

    Given the fact that both AT&T and Verizon spend around $15 Billion per year in capex on their cellular network, I think my numbers are more realistic then what the article presents. Both AT&T and VZW (and all the other US carriers) have a huge advantage in that they already own and/or lease tower space. So, adding 4G is an incremental cost whereas a new entrant would be starting from scratch.

  6. FAIL. Your using a picture of a GPS satellite, which is for navigation and not telecommunication. Do a search for “commercial communications satellite” next time.

  7. Capital investment in wireless will require experts during the pre-deployment phase. I’d like to hear HOW they plan to deploy this 36K base station network. Bunch of consultants and money people think they can simply roll out a nationwide LTE network? My bet is that they miss the “how much it’s gonna cost” mark by 80%.

    • Anonymous

      Agree with Anon. They have no idea what they are doing. Sounds like an awesome idea, but it’s basically a copy of Clearwire’s plan. Except instead of OFDMA/OFDMA, it will be OFDMA/SC-FDMA. They will underestimate cost.

  8. Somewhere in a closet is my old blue Irridium cap. Time to break it out!

    Few things better than when other people with lots of money actually use that money to build a (doomed?) business that we can all benefit from.

    And I say again: The smartphone is the computer! There is no edge. If they build a 4G nationwide network, it will be used.

    • What local providers are going to partner with satellite companies?

      Cripes, a significant chunk of PBS local TV providers can’t even agree on whether or not they should provide a HiDef signal to DirecTV or Dish.

      You expect what local providers to build into 4G?