Broadband in the US is in a mess. There are two bodies of opinion on what will change it – repeal the Net Neutrality rules embodied by placing broadband under the Title II legislation, or throw brand new wireless technology at it.
The truth is that neither are guaranteed, and the reason is straight forward – this is due to lack of competition in broadband in the US, where in many parts of the country there is just one, usually rather slow, broadband provider.
Double up on competition and prices would almost certainly fall, resulting in less money being available to spend on broadband capex. That’s the modern equivalent of Catch 22. Eliminate Net Neutrality and cash would surely flow into the coffers of all the existing broadband providers, in the process putting a damper on internet innovation – as pure internet companies are forced to pay for fast track broadband delivery. It would slow the likes of Google and Netflix and take money destined for their services and deliver it to broadband suppliers, and at the same time it would put a Venture Capital block on investing in any company that relied on the internet. Perverse.
Take for instance the 650 US start-ups in and around AI. Right now every one of them has to rely on broadband delivery, as for the most part their AI smarts are in the cloud, and getting data to it, and decisions from it, especially in real time, rely on broadband. Start-ups would effectively be taxed to death or certainly into being less successful.
Broadband companies like AT&T and Verizon in particular, as well as cable firms Comcast and Charter, would all be “guaranteed” extra revenues whether or not they innovate. We are supposed to believe that this would in turn make them “invest” in a way that they have not since the beginning of the internet. We are expected to believe they would invest in green field broadband territory to take on one another. In other words, create greater competition in broadband, with less territories having a single ISP choice. AT&T is at least trying, but Verizon has no such plans and cable firms the US over have never stopped investing.
This week consulting firm IHS Markit came out with a statement paper on Net Neutrality, and it published a map of the US based on how many suppliers there are who can deliver 10 Mbps and above broadband speeds in each territory.
It’s not pretty reading. 38% of the country has just one broadband provider. Another 34 have only 2 providers, while 10% of the US is simply not covered at all. That means that just 18% of the US has stiff price competition between ISPs.
That is not going to change through altering legislation. Ironically the only operator who has talked about taking broadband outside of its existing territories is AT&T, one of the most vociferous about shutting down Title II.
Investors don’t care a jot. If everyone in the world can get faster broadband than US citizens, and pay less for it, they don’t care. What they care about are the profits of the large US ISPs and guaranteeing them. Only the US citizens care, and they are deprived of making an informed choice on broadband providers, because for the most part there is no such choice available to them.
It takes a very weird mentality to allow rivals from cable to eat your lunch in broadband, while investing like crazy in cellular, all the while complaining that failure in broadband is the government’s fault.
In 2007 AT&T had 16.3 million broadband lines and Verizon had 9 million, and between them they were dominant. Today Comcast has over 25 million, and Charter has 21.8 million, both more than double the 2007 numbers, up well over 1 million each year over that 10 years. Meanwhile AT&T now cites 14 million, although when you add up enterprise broadband and such it is just about the same number at 16 million, and Verizon has fallen back to just over 7 million with 5.7 million of them over fiber. Verizon has stopped installing fiber and has no broadband strategy except untried 5G Fixed Wireless.
Any normally commercial organization, which isn’t guaranteed money for nothing through sympathetic legislation, would simply pick a strategy and back it with their own cash. AT&T and Verizon want water-tight government guarantees of their profitability. Had they treated their cellular interests in the same way, they would be out of business by now.
But during 2018 new strategies will emerge. AT&T said this week that it is making Fixed Wireless Broadband available in more markets, much of it in and around South Carolina, but also in parts of Alabama, Arkansas, California, Florida, Georgia, Illinois, Indiana, Kansas, Kentucky, Louisiana, Michigan, Mississippi, North Carolina Ohio, Tennessee, Texas, and Wisconsin.
This is an impoverished strategy that customers will have to pay through the nose for, to receive a poor service. The service is just over 10 Mbps, with 1 Mbps upstream, it is not the 5G wireless broadband that both AT&T and Verizon have committed to – that will be delivered elsewhere. The company describes it as an efficient way to deliver high-speed internet to customers, many of them for the first time. In other words AT&T is using traditional Wireless Broadband to serve that underserved 10% of the community, and it will likely price it just below satellite delivered broadband, because that’s all it’s up against and the price for that is exceptionally high.
The last company to adopt a strategy based on this type of technology was Clearwire in the 2006 to 2015 timeframe, during which time it managed to capture 11 million customers, a considerable number, but not profitably because it was using expensive WiMAX devices. AT&T will not want to repeat that, but instead offer it at a high enough price to make money, and in a state where it can upgrade it to “5G” wireless broadband, without replacing too much hardware, if any, at yet another, higher price level.
Another item in the news this week is that AT&T has another trial of its AirGig technology, with a rural electric utility Georgia Power. It also said for the first time that one of its recent trials was outside of the US – we suspect Mexico.
AT&T said that AirGig makes sense to deploy in areas that don’t have any broadband service, whether rural, suburban or urban. So again it is harping on about serving the underserved community – that 10% and some others which have a single ISP – with this technology.
We also see AirGig as a great way to deploy cost effectively. The company continues to talk about it as if it were powerline delivered, when in actual fact it simply piggy-backs on powerline poles and is a millimeter wave wireless technology, which has a robust encoding scheme that is not interfered with by proximity to power lines. It relies on cheap plastic antennas, and is patented by AT&T. And planning is easy, because every building that needs broadband is already served by power poles.
We suspect that this is a great way to bring a 1 Gbps or faster line into an MDU as well, shooting it from the top of nearby power poles up to the roof and dropping broadband down using G.fast over coax, another technology that AT&T is known to be investigating. It could come down the same wires used by DirecTV for a channel stacking switch on the roof, so nothing new would need to be installed except a single rooftop antenna.
But the business end of AT&T’s broadband thrust is likely to come from G.fast over twisted pair, in fact over multiple twisted pair so that over 1 Gbps can be delivered, and by building out more and more fiber to the home. In its last quarterly figures fiber now passes 6 million homes, and this number is rising at the rate of 500,000 homes each quarter.
This week Verizon confessed that Ericsson will be the network provider that will trigger its “pre-standard” commercial fixed-wireless 5G and it will come in the second half of 2018.
We do not believe for a second that the radios used in this, although they may adhere to the core technology approach in 5G New Radio, will be the same radios that will end up inside mobile phones. It is likely they will be prototype, oversized, power hungry, early market monsters. They are likely to be supplied by Qualcomm, and Ericsson will build out the rest of the infrastructure, which will include backhaul. At this point the backhaul is likely to be a considerable order for Ericsson, relying on its existing Mini-Link series of microwave backhaul point to multipoint services which operate up and down the spectrum be that 28 MHz or anything from 4 GHz to 80 GHz. This will cost Verizon an arm and a leg, and its capex for 2018 will rise over its 2017 numbers.
The gamble that both AT&T and Verizon are making here is that other wireless operators around the world will also have the right commercial conditions to support wireless broadband, and that global volumes will grow on the components, driving down last mile connection pricing. Right now there is little evidence that anyone outside of the US is under the same pressure – having trod water for ten years in fixed line broadband instead of moving their connection points closer and closer to the home.
Verizon and Ericsson have also conducted fixed-wireless 5G trials using millimeter wave spectrum in several cities in the US, so the deal is no surprise.
Both AT&T and Verizon can classify their widely varying fixed broadband investments any way they like, and we are sure to see references to investment plans being laid from the moment Title II was knocked on the head. But these are vastly differing approaches. Verizon is placing its entire fixed broadband strategy on an untried Fixed Wireless plan, that is likely to be about as successful financially as Clearwire’s, while AT&T has hedged its bets four different ways – fiber, MDUs, Fixed Wireless and G.fast to the home – and the effect of using pre-standard – read expensive – 5G radios will be diluted at AT&T by the other activities, whereas at Verizon it will not.
Once all of this shifts into the price for Verizon, and its strategy plays out, in what can only be at best either a “slightly” successful adventure or an abject failure, Verizon will look more and more like a takeover target, while AT&T is likely to make its first in-roads into cable broadband for over a decade.