Extensions to Wi-Fi platforms (see separate item) will enable the US cablecos to take on the telcos in mobile and quad play markets, and at the same time solve the video download problem. As usual it is Comcast with the big vision and somewhat in the lead, and executives this week are finally spelling out their plans clearly to the media. ‘W-Fi gives us a chance to change the competitive equation,’ is the clear message.
In Europe we have the advantage of seeing first hand just how transformational contiguous Wi-Fi has been in places like France, where broadband and cable operators use a combination of hotspots and homespots (using home gateways to deliver a commercial hotspot) alongside little more than an MVNO presence. But in the US it is just starting to become apparent how the same strategy can be used to challenge the established pecking order.
If we go back in time Cablevision was the first to try to build out its territory with outdoor Wi-Fi hotspots, and those Cablevision locations hold the key in the New York area, to giving Comcast and others an opportunity to offer a collective Wi-Fi presence that dwarfs anything else ever seen globally.
Cisco home gateways are being pushed as homespots in the US now, and come ready-made with two SSIDS. We know that in Europe in territories such as France, the Netherlands and Belgium, cable and other operators have fitted this retrospectively to existing home gateways and routers, triggering 3m homespots in the Netherlands, where there are only 7.2m homes, and a further 2m in Belgium, where there are just 4.4m.
That indicates the massive potential scale in a country the size of the US. But, the big question in the end is how to use this Wi-Fi potential. There is still limited clarity on the broader strategy, and that’s because it can go either way right now.
First off it is far harder to give the impression that a Wi-Fi signal is all you need, when you are talking about the US ‘ houses are larger, (although less often built with hard-to-penetrate concrete), and gaps between Wi-Fi coverage areas are far greater.
Then there is the issue of how to allow authentication. In the Wi-Fi world there is Hotspot 2.0, which allows someone to transfer attachment between two access points, as long as they are both managed by the same authentication servers. In cellular the ANDSF (Advanced Network Detection and Selection Function) does just what is says, including linking to existing SIM authentication servers in the mobile core. But cable customers which will merely be asked to use their unique cable subscriber number and an allocated password, to log into access points, whether they are hotspots, or homespots, effectively all using the same authentication, in the private cloud of cable companies.
Of course if Comcast and co want to partner existing cellular operators and allow their hotspots to be used by non-customers, for mobile offload, then there needs to be a link between their own authentication centers and an authentication system inside a cellular network ‘ in terms of overall system cost and technical decision making, we think that’s a trivial problem to have to solve.
But that’s the first decision ‘ whether to use the distributed access point capability to increase the number of customers that cable has, or whether to make as much money as possible out of its Wi-Fi capability.
We think that this is a matter of time. First US cable will try to give their own customers a major advantage, and then later, if they find that they cannot harm the telcos sufficiently by stealing customers, or when the US telcos have recovered and caught up, they will offer the capability for sale to the highest bidder and make as much money out of it as they can.
The end point of that attack is now starting to be seen. Cablevision made the mistake of trying to build out a system which replaced the entire cellular service with Wi-Fi and that was never going to work.
But over the past week it has become clear that the cable companies that were due to partner with Verizon over multiscreen delivery and cellular, have pulled away, signified by the closure of the joint innovation lab between Verizon and the cable companies. Verizon is still supposed to have a joint marketing deal with Comcast, Time Warner Cable and Cox Communications, but the two groups are now far enough apart that they don’t want to share R&D.
At the heart of this separation will be a key technology decision. Verizon, if it knew that it was expected to carry multiscreen content on behalf of its cable partners, would want to use LTE and LTE-Advanced and use the guaranteed traffic and therefore revenue, to light up the spectrum it bought from cable and use conventional 3GPP technologies ‘ small cells, HetNets and LTE Broadcast – to deliver the content. This way, with guaranteed video traffic, Verizon could afford to build the most advanced network in the world, reclaiming its edge over AT&T, and building a case to replicate this network excellence in other countries, either as consultant or partner to overseas businesses. It would also allow it to bring FiOS to market over cellular, in a way that AT&T could not match.
Initially the cost of doing this with Wi-Fi would be almost as great as going down the 3GPP route, in a scenario like Verizon’s. There have been some estimates that Wi-Fi is about half the cost of 3G to install, but a business case analysis by research firm Wireless 2020 has found that the savings were more likely to be in the order of 7% to 8%.
But once homespots are introduced to the mix, with existing arrangements for peering and international internet traffic and backhaul already in place and the equipment paid for, then even if the cableco still has to build out some hotspots to fill in gaps, there is still an irresistible proposition, perhaps 35% cheaper than cellular. That will be reflected in the continuing cost of what needs to be charged in order to make it profitable, ie a lower cost per Gbyte.
Cable would have seen that an MVNO relationships with Verizon, with their own home gateway assets doing the heavy data lifting, was far preferable to going down the 3GPP route, and if they did not need the backhaul excellence of Verizon, or its existing long haul fiber, then they would not need to give away the crown jewels by sharing R&D.
The delivery of the spectrum to Verizon was always going to result in an MVNO proposal and that has now been brought forward and can begin in 2014, courtesy of the intervention in the terms of the deal by the FCC, who thought that keeping cable out of MVNOs was not a good thing for the consumer ‘ quite right.
Now with the rabid build-out of hotspots in locations that support the shift to Wi-Fi offload, we see these three cable companies, with perhaps others to join in, especially Cablevision, pointing to a US future somewhat similar to that in France.
There, Iliad’s Free is being investigated to see whether or not it is building out its cellular spectrum rapidly enough to meet its commitments to its regulator. It has a tiny cellular presence, but began with an agreement to use roaming cellular capacity from Orange plus its own licence. By not building out more infrastructure, it pays a large fee to Orange, but collects even more through its new cellular customers.
It totally disrupted the French cellular market, savaging valuations, and putting operators like SFR into a weak position, where it has been put up for sale by its owner Vivendi, which has since proposed a merger with cable player Numericable. Free pays huge fees, but has reaped great customer acquisition benefits.
Free offloads some 70% plus of all cellular data delivery to its own homespots, paying nothing for any traffic that goes over this network. It has 5m homespots and has built out just 2,200 base stations.
The profitability of the entire cellular sector was radically altered by a move from a single cellular entrant, which doesn’t even have a network, but which has widespread Wi-Fi. It could afford to sign customers for a small payment, something closer to collectively paying for handsets, plus the Orange roaming charge, which turned out to be about 30% of prior market pricing, which is why it picked up 2m customers overnight.
Imagine Comcast, Time Warner Cable and Bright House Networks offering cellular at a collective price that was just enough to pay for the MVNO minutes that US customers used, plus the handsets they bought. It could come in with unlimited data deals that were restricted to Wi-Fi, for just a few dollars a month, plus say $10 per customer per month to pay for the voice minutes and the set up costs and the financing on handsets. The stock price of Verizon and AT&T would crash overnight. It would also make it incredibly hard for someone like Dish Network to get anyone to take seriously the idea of a wholesale cellular network (although if Verizon found a way to wriggle out of the MVNO deals, this might be the next stop for cable).
This would mean that AT&T and Verizon would have to pursue the same Wi-Fi strategy to avoid being unable to use the same offload economies. AT&T has already invested heavily in hotspots and a few weeks ago tied up with FON, an organization which offers software and a consumer contract to download and create two SSIDs on an existing router, and which has built up a network of some 12m end user customers across Europe, via telcos.
At the time AT&T only talked about giving customers a better Wi-Fi experience, but clearly this is all about AT&T offloading Wi-Fi from cellular phones. We would suspect that most U-Verse broadband routers could be turned over to this system, so perhaps 9m of those, plus AT&T now claims to have 30,000 hotspots in the US. It inherited more like 80,000 when it bought Wayport in 2008, but it looks like these have been rationalized, and it has also lost the Starbucks account to Google, which will have also diluted its Wi-Fi holdings.
So far, AT&T has never openly discussed its FON partnership in the same breath as cellular Wi-Fi offload, but FON has told us privately that it is in the midst of helping some operators to negotiate these types of contracts. Now we have to wonder about where this leaves Verizon Wireless in the coming war for Wi-Fi offload capability. It will be able to negotiate reasonable pricing with the cable operators, but its own home Wi-Fi base and its hotspot holdings are certainly not enough on their own to provide Wi-Fi offload, and anyway both AT&T and Verizon’s physical fixed line networks are regionally based, with the exception that AT&T acquired the original AT&T long lines business, so has huge backhaul capability, if not actual Wi-Fi holdings, outside of its fixed line territories. Verizon does not.
If cable holds its act together and comes to market with a nationwide Wi-Fi offload network, it will be set to act as kingmaker to the two large cellular operators. The two major US cellular operators would be faced with co-operating over Wi-Fi offload, with one another. So either way expect Verizon to join FON in short order. Many of the Verizon broadband lines are old ADSL and in need of updating, and probably only its FiOS lines are worth using for Wi-Fi homespots, which is a number under 5m once you pick out any unsuitable devices.
Between them AT&T and Verizon offer broadband to 25m homes, but around 11m of these are slower ADSL based devices, and may not have the spare capacity or the Wi-Fi reach to offer any service outside the homes they are attached to.
It is now down to the individual cable companies to see how they use this new Wi-Fi advantage. Will they use a national Wi-Fi chain to force customers to sign up for their broadband, accelerating the drain on the major telcos’ broadband customers, which have been static for years? Yes at the very least. Will they offer the capacity up for cellular offload? Almost certainly? But if they are brave enough they will begin their own MVNOs and offer automatic Wi-Fi offload across the entire country, and use that to slash cellular pricing. In the process, this will crash the share price of both of the US cellco majors and send reverberations around the entire global cellular marketplace. Exciting times.
This analysis was written by Peter White, principal analyst of Wireless Watch’s sister research service Faultline, which provides weekly analysis of the digital video and content markets and the strategies of quad play carriers.