FTTP, FTTP on Demand, G.fast, Ultrafast

Apologies to those few who actually still read this blog, and being reasonable why would you, for the lack of updates.

Going to cover a few different things.

Openreach have started a trial in Haydon Wick, Swindon in the hope of deploying FTTP more quickly and cheaply. This is a combination of using the connectorised fibre solution tested in Huntingdon and placing rope into ducting to pull the fibre in advance. This has allowed delivery apparently considerably more quickly and at a far lower cost per premises passed.

Something really strange is why Openreach didn't use connectorised fibre in their original rollout but instead used far more expensive and time consuming fusion splicing resulting in installs as Think Broadband noted taking an entire day. This isn't new technology. Verizon were using it for FiOS installs in 2005.

Placing rope into ducts to pull cables in advance isn't a new thing, either. Virgin Media place rope into their swept tees ready to pull coaxial cable through from taps and this, too, has been used for FTTP in underground plant.

Apart from using an active fibre solution, which wouldn't have happened as it would've left the possibility that other operators could unbundle the fibre, I struggle to see how Openreach could've done their original tens of thousands of commercial FTTP premises more expensively. They ended up having to do a ton of work to build the initial networks due to interesting deployment area choices, have ended up asking for thousands in construction charges to connect some properties which begs the question of why they'd deployed there, and chose to use fusion splicing instead of cheaper connectorised throughout.

In many ways Openreach treated FTTP less like something that they could take mass market, and planned to, and more like each and every property was receiving a leased line. Strangely enough this worked out to be really time consuming and really expensive. Trying to avoid putting the tin foil hat on you find yourself wondering if they ever had any intention of deploying widely initially, instead wanting to prove to themselves it was infeasible, and now they've more commercial motivation they are increasingly using appropriate construction techniques.

FTTP on demand. I've given up on this one. My best laid plan has fallen through because the operator doesn't want to involve themselves in the product any more. The informed that it was awkward to order, the surveying process caused problems, the provisioning caused them problems, the way the product was built at a technical level caused them problems, it took way longer to provision than leased lines, and demand was so low due to the cost being so high. A 100Mb leased line, guaranteed bandwidth, guaranteed availability, would actually cost them and, in turn, me, less to provision than a best effort FTTPoD service with no SLA right up until about month 33, and wouldn't involve a 36 month contract so would be cheaper across the period I required it.

The ISP mentioned that initially the only thing about the FoD product that wasn't a nightmare was the price, and this was swiftly changed in early 2014 when the install price was increased by over 50% and the monthly rental charge to them considerably more than doubled from £38+VAT per month to £99+VAT per month.

With considerable irritation and a couple of rather upset comments to people in places of power I've given up any chance of a resilient solution through Openreach and made Starbucks and Costa my backups for working from home.

This leads on to G.fast and the Openreach ultrafast deployment plans.

Openreach have informed they plan to reach about 10 million premises in the UK with ultrafast broadband, likely heavily G.fast, by 2020. They could do this relatively simply by putting new G.fast kit next to their existing cabinets. No need to push fibre deeper into the network, no need to spend on FTTP. It would, however, mean that if you're further than 2-300 metres from the cabinet you're out of luck.

They have made noises about increasing the FTTP in their network and the trials they're doing seem to imply this, however it should be noted that much the same noises were heard a few years ago, and to date Openreach have built to less than 0.4% of the UK without outside funding. Less than 100,000 premises when excluding co-funded builds in Cornwall and around the country via BDUK.

We'll see. I hope to be pleasantly surprised but, based on past experiences, will keep an open if possibly cynical mind.

An area like this one is interesting. FTTC/superfast broadband uptake is very high here - of about 550 premises covered by our cabinet 500 have taken it. Which leads to the question of how Openreach decide where to deploy G.fast and/or FTTP. I'm interested to see if the very high FTTC uptake actually deters Openreach from further investment. From a cold, commercial perspective why spend money, even if it's a relatively small amount colocating G.fast next to the existing equipment, building ultrafast when superfast is selling so well?

I've been told one won't influence the other but, for obvious reasons, it's tricky to imagine it won't form part of the business case. Surely where Openreach are having their backsides kicked by Virgin Media or an alternative network they'll be far more inclined to spend to win customers back? Likewise why make even the relatively modest investment in FTTC redundant when it's bringing in a good amount of money and will pay for itself then turn a profit relatively soon?

Talking of Virgin Media I noted on a forum thread that I can't remember when there has been such a large difference between the top widely offered speeds between the two. Openreach only go above 76Mb in a tiny fraction of the UK while Virgin Media are offering 300Mb, which performs at around 320Mb, to nearly half the country.

The explanation is relatively simple - Openreach's customers have no interest in funding a speed war with Virgin Media. G.fast won't win it, with relatively modest expenditure Virgin could release services faster than the 500Mb G.fast is 'planned' to achieve by 2025, only FTTP would allow Openreach's customers to compete and their customers have no interest in paying for it. Two of the three largest Openreach customers don't offer FTTP via Openreach's existing services and constantly and loudly state their desire to pay less for the existing FTTC services.

There are a number of divides between areas in the UK and here, as in many other countries, a big one is between those who can get cable and those who cannot. The incumbent neither here or anywhere else has a pressing business case to change this with any kind of urgency.

A monopoly of mediocrity

The announcement of FTTP trials in Bradford alongside 12,500 premises of G.fast in Gillingham had me thinking a little.

One thought was who in Gillingham was influential enough to persuade Openreach to invest some serious money there; it's fair to say that it wasn't in any way related to the benefits it might bring given the entire constituency is 92.2% ultrafast, 98.6% superfast, however I suspect the extensive Virgin Media coverage wouldn't have hurt.



Another, partly inspired by Virgin Media digging up pavements nearby, was to what extent Openreach invested in ultrafast technology and indeed the extent of their investment in urban areas generally during their initial rollout.

I've seen the numbers, however they are put into focus when trawling the Think Broadband data.

Leeds is the 3rd largest city in the UK. It has a population of over 3/4 of a million. It has the largest legal and financial centres in England outside of London. It's home to the Bank of England's office outside of London.

According to Think Broadband it has no native Openreach GEA FTTP. BT's contribution to ultrafast broadband in this entire metropolitan area is, as of right now, zero. Every single FTTP/H premises is Hyperoptic enabled apartments, every ultrafast premises is either them or Virgin Media.



This is something I've felt keenly as I've been looking into home office options to get around a wobbly superfast broadband service. The only business broadband option from BT Group potentially delivering both performance and reliability would cost over £15,000 over the mandatory 3 year contract term, and this wouldn't carry guaranteed performance or availability targets.

A leased line supplier informs that the install quote I received for FTTPoD, the shared fibre service from BT, is 'rather high' and they can install dedicated fibre to my property for 1/3rd that.

AQL can probably offer me a leased line, guaranteed performance and availability, for considerably less than what is basically a business broadband service from BT, which is absurd.

Alternatively I can rent office space. Also cheaper than the BT option.

We should, by now, have an urban core of FTTP/H in the major settlements with FTTC/VDSL on the outskirts, in some market towns and where appropriate rural areas. Instead there is far more FTTP/H per head in rural areas and the amount of it in urban areas, apartment blocks and a few areas built out to for political and other reasons excepted, is vanishingly small.

The UK has one hell of a long way to go and a hell of a lot of work to do moving broadband to the next level in the 'just in time' approach espoused by BT Group. The approach that has allowed Virgin Media to tread water and others to sit back and do nothing to one being upgraded to one that's fit for purpose going forward. Openreach deploying ultrafast in an urban area should not be newsworthy. It's depressing that what should be the status quo, 300Mb being available in a couple of urban areas and gigabit to very limited areas of another, hits the news.

Superfast is the new normal. Ultrafast is the new 'superfast'. Both should be available in our large cities to those who wish to pay.

The market as a whole has, with Ofcom's encouragement and indeed thanks largely to their obsessive over-regulation, become a monopoly of mediocrity.

How much bandwidth does a busy household need?

At the moment our line is running way slower than normal.

So how are things going? Are we managing?

Mmm kinda. Here's the Broadband Quality Monitor from yesterday.

Usage on the line is pretty high during peak periods, the end result being higher latency and jitter. Here's some of the usage over the 'peak' period: this is 6pm - 11pm.

Video isn't quite buffering on YouTube, it just about keeps up but we couldn't sustain 1080p SuperHD Netflix if the line were busy. The latency requires use of some technical methods to manage the line.

Could be worse; could be our line pre-FTTC, though that was also before teenagers were using it!

In summary we can manage but I'm very much looking forward to getting this fixed sooner rather than later as it has the potential to be problematic. We are high-end users, which is why we pay for a business service.

I'm looking into alternative options however, as previously mentioned, there are none besides spending about the price of a car tying ourselves into a 3 year contract... which we may have to do if things can't get sorted.

Virgin Media are trolling me - home office broadband and moving

Walk out of my street and onto the road that serves ours to find this:

Virgin Media are going right past the entrance to our street on their way to properties across the road.

It'll be interesting to see what impact VM passing an area served by cabinet 82, albeit maybe 20-25% of the properties served by cabinet 82, has on the extremely high FTTC uptake here.

Predictably this work has caused complaints. This is, perhaps, something that Virgin Media are getting quite used to by now. Digging pavements is slow, expensive and, albeit briefly, disruptive.

I have been having some issues with my home office service's reliability and performance that make my desire for a backup that uses a different network all the stronger.

Less than half usual performance

A quote for FTTP on Demand not only gave a huge price but I was also told they weren't willing to progress the order due to my road's status.

For many reasons we've been considering a possible house move. Unavailability of alternative broadband options for my home office goes onto the list. It's nowhere near the top but has to go on there.

Over half of the UK has the choice of BT and another fixed line operator.

The vast majority have a choice of BT FTTC and either a >8Mb Openreach ADSL line, cable, wireless or 3G/4G. Sadly we aren't, and have no prospect of being, in that cohort.

FTTC is good enough for most purposes, is far cheaper and faster to deploy so made economic and business sense - I have no complaints about that solution being here. It'd be fine as a primary line for us if we had a backup solution however no cable, no worthwhile ADSL, no wireless and no 3G or 4G puts us almost back to the position we were in during the Fibre for Middleton campaign that saw BT deliver superfast broadband to us.

Maybe we should move to an FTTP-enabled rural village or hamlet; more FTTP serving houses in those than Leeds Central. In common with many other urban areas the fibre to premises in Leeds Central is all delivered to apartment buildings by Hyperoptic. FTTP is by far the most reliable broadband out there so if you can get it, do.

We're between rock and hard place, and while it seems as far as broadband goes the in thing is to complain relentlessly on forums and the comments section of ISP Review, the usual and more constructive solution is to get out from between that rock and hard place.

Update: I just had a local resident ask me when he could expect Virgin Media to cover his street. Had to tell him that if he was lucky it would be 2019. If unlucky never, and that it was extremely unlikely he would have better broadband options than the ones he has now, which are 1Mb if he's lucky or 40-60Mb depending on where in his street he is. 

This estate has a household income over 3 times that of the rest of the ward, and far more home businesses and home workers. Much of the estate will have broadband of at best 1/8th the performance of the rest of the ward and more likely 1/14th to 1/25th by 2019.

There's a digital divide.

Broadbad or just bad?

So let's dive into this work of research that is Broadbad. Quotes from the paper in italics.

Page 8:

Key Findings

• Openreach has so far received £1.7 billion in taxpayer subsidies to connect harder to reach areas of the UK to superfast services, but has repeatedly failed to deliver

Openreach have not received £1.7 billion in taxpayer subsidies and have not repeatedly failed to deliver. They have received approximately £700 million according to BT's accounts and projects are ongoing. Openreach have delivered the coverage agreed almost universally in the timescales agreed with local authorities, and will continue to increase coverage under the Superfast Extension Programme.

• Around 5.7 million people in the UK have internet connections that do not reach Ofcom’s ‘acceptable’ minimum speed of 10Mbit/s. 3.5 million of these people live in rural areas. 

Not the case. The source material data was collected during May and June of 2015 so is out of date, and that same Ofcom Connected Nations Report states:

'Government programmes, such as those administered by Broadband Delivery UK (BDUK), are helping to address the problem of poor broadband coverage, particularly in rural areas. We would expect to see further improvements in rural broadband availability over the coming 2 to 3 years.'

This is more accurate and up to date.

• Poor internet connections are costing the UK economy up to £11 billion per year

Based on research from March 2015. I have no idea how valid that figure is, however it's not the job of private companies to spend money to increase UK GDP. The Conservatives privatised BT. Private companies have a responsibility to operate in the best interests of their shareholders. I'm sure we can find other industries that have cost the UK economy more and don't have MPs having papers written demanding their reorganisation. A number centred around the Canary Wharf and City of London area come to mind immediately.

• 42% of SMEs report experiencing problems with their internet connectivity and 29% also report poor service reliability.

Yep. The exact quote from the source material is the following:

'Our research has found that 42% of SME internet users had experienced issues with their internet connectivity in the preceding 12 months. Poor service reliability was the biggest problem, with 29% citing it as an issue, followed by slow download (16%) and upload (13%) speeds.'

Of course what is ignored is that reliability problems can be caused by wholesale providers and service providers, not just Openreach, as can performance issues, and that these are subjective measures. At what point will an SME complain about reliability? An outage every day? Month? Year? How do we know these problems are all the fault of Openreach?

• Following the announcement that BT will be merging with EE it has been calculated that BT will have a 40% share of the retail telecoms market and a 70% share of the wholesale market.

Separating Openreach from the rest of BT Group, as the paper demands, will have no direct impact at all on BT's retail market share and will potentially have quite the opposite indirect impact from reducing it. Free of the regulatory burdens involved with having Openreach as part of the group BT's retail arm may be free to more aggressively pursue market share. It is unclear what is defined as their share of the wholesale market as the source for these numbers isn't an Ofcom research paper, it's a Guardian article discussing Sky's and especially TalkTalk's advocacy of separating Openreach from BT Group.

• The time has come for BT to be forced to sell off Openreach to encourage more competition and a better service for every internet user and for the benefit of the UK economy. 

Strangely given this is apparently so self-evident there are zero specifics given as to how this would encourage competition and improve services; in fact the expected impacts of separation are given 2 vague paragraphs in a 22 page report.

The introduction of the paper mentions Mr Shapps being alarmed by complaints about broadband in his constituency despite its close proximity to London, because London is the centre of everything and so proximity to it should be used as some kind of measure in these things. Alongside this comes confusion over what a natural monopoly is.

The following merits quotation. Page 9 paragraph 2:

This report contends that Britain, the birth place of the man who invented the World Wide Web, Tim Berners-Lee, should be leading the world in digital investment. The Government and local authorities have now generously granted a total of £1.7 billion in subsidies to BT to deliver broadband to the harder to reach areas of the UK.

A noble aim and with that in mind government and local authorities have provided a 'generous' subsidy. Except that subsidy, actually £332 million from central government matched by local authorities at the time of the numbers in the 'Key Findings' section, though higher now hence my mention of £700 million earlier, looks somewhat less than generous noting France's £17 billion public-private partnership, Germany's E2.7 billion for 50Mb+, or Australia's £14.9 billion to cover a population less than 40% of ours for the NBN project.

Singapore spent £639 per premises on broadband subsidy to deliver fibre to premises in a city-state. Our spending relative to that is hardly generous.

It's extremely disingenuous to quote total subsidies available and criticise based on results when a fraction of that subsidy has been used to date while also ignoring that all the subsidy isn't going to be used as Openreach have already returned some of it.

Page 9, paragraph 3:

Britain should have the most developed digital economy in the world but is instead lagging behind countries such as Japan, South Korea and others. 

I could discuss at length why this is a ridiculous statement but we'll stop with a few simple facts. A far greater proportion of the populations of Japan and South Korea live in apartment blocks. In addition utilities in Japan and South Korea tend to be on poles. Those things that are considered to make places look 'third world' or become 'nineteenth century landscapes with overhead cables and wires.' but are really handy to string fibre optic cabling onto rapidly and relatively cheaply. Lastly there are more people in heavily densely populated urban areas in Japan and South Korea and their population densities overall are higher than ours.

We could, of course, focus on urban areas and delivering ultrafast services to those at the expense of leaving rural areas underserved for sure, delivering 1000Mb to many urban areas costs less per home than delivering 10Mb to some rural ones, but that would make complaints about the digital divide all the louder.

The idea that we should be at parity with these nations given the different factors at play and the level of public subsidy supplied thusfar is delusional.

Page 10 I will not comment on beyond to note that premises, not people, are the usual metric with regards to broadband coverage. Other parts are already covered elsewhere.

Page 11, paragraph 7:

As fig 1 on page 5 shows, vast swathes of the nation suffer from slow or even non-existent broadband speeds. A look at the South West of England, huge areas in Scotland and Wales and also the North of England show a nation plagued by poor broadband service.

These are sparsely or even virtually unpopulated areas. As a general rule operators don't tend to deliver copper and optical lines to places where there are no people. These aren't wireless solutions where people roam into the coverage, they are called fixed line for a reason. There's no fixed line broadband because there are no fixed lines to carry the broadband.

Page 13, paragraph 9:

It famously claimed in 2009 that 2.5 million homes would be connected to ultra-fast Fibre to the Premises (FTTP) services by 2012, which is 25% of the country. Yet by September 2015 they had only managed to reach around 0.7% of homes. 

http://www.ispreview.co.uk/index.php/2013/04/bt-abandons-native-uk-fttp-broadband-rollout-for-fttpod-and-fttc.html

Frustrating as it was that the FTTP was scaled back, and that same ISPR review notes that BT actually spent more than they planned to on greater coverage than originally mooted in 2009 alongside upgrading the FTTC from 40Mb to 80Mb, 2.5 million homes isn't 25% of the country. Believing this would explain much about Mr Shapps' rather unfortunate tenure as housing minister given that there aren't 10 million homes in the UK, but as of 2014 26.7 million.

Page 13, paragraph 10:

BT has also frequently been accused of abusing the natural monopoly it has over the nation’s network and not giving equal access to other internet providers

I can't comment on this as the link cited, http://www.reuters.com/article/us-britain-telecommunicationsidUSKCN0J70W420141123, doesn't work.

I've had what looks like the correct link sent to me. Those making the accusations are Sky and TalkTalk, so obviously zero chance of a vested interest there. The link also makes clear where the 40% consumer and 70% wholesale market claim mentioned in the earlier Guardian article was sourced. In this article it's a direct quote from Dido Harding, TalkTalk Chief Executive.

Page 14, paragraph 11:

Competitors, like Sky and TalkTalk, argue that the relationship between BT and Openreach further exacerbate the problems brought about by the current situation already being a natural monopoly. These other providers have to pay a wholesale price to BT to make use of the network and they then charge their own customers for services.

It should be noted those guys pay Openreach for the most part. This comment muddies the water somewhat by being unclear. I'm unsure what alternative arrangements there could be. Separating Openreach doesn't alter the 'natural monopoly' point, it removes vertical integration which is a completely different concept.

They have consistently argued that Openreach has little incentive to invest in upgrading the network and in fixing faults quickly.

They have little incentive because no-one else is investing. This is a failure of the UK's broadband market and the the business case for building ultrafast broadband networks here. Separating Openreach isn't going to change this, they will still have no incentive to upgrade or fix faults more rapidly as it won't introduce competition to Openreach; they remain a natural monopoly until someone steps up with an open wallet and willingness to wholesale the network they build.

Page 14, paragraph 12:

Logically this means Openreach has little to gain from improving the network. At the moment the infrastructure is largely a copper network, meaning it is outdated. BT has made much about a new technology it is introducing, G.fast, that it claims will enhance the performance of the existing copper network, using better signalling kit to push more data into the wires. The obvious criticism to make of this approach is that BT are merely trying to eke out what life there is left in an outdated network system instead of planning for the future and upgrading to a fully fibre network

It's believed approximately 50% of BT's spend on G.fast will be re-usable for a 'fully fibre network', just as a proportion of their spend on FTTC is reusable on G.fast or a full fibre network. This is just-in-time capital expenditure. Virgin Media in common with other cable companies are doing the same.

Page 15, paragraph 13:

BT has so far received £1.7billion in taxpayer subsidies to roll out superfast broadband to harder to reach rural areas. Ministers have done the right thing by wanting to connect up these areas but were
badly let down by the regulator and BT. The push should be for stronger, more competitive networks. 

As mentioned above BT have not received £1.7 billion from the taxpayer. This was not specified to cover harder to reach rural areas, that was down to local authorities in co-operation with BT to decide. This programme was nothing to do with Ofcom. I have no idea what is meant by 'more competitive networks' that seems like a bit of 'buzzword bingo'.

Page 15, paragraph 14:

Openreach generates 50p in earnings before interest, tax, depreciation and amortisation for every £1 of revenue. No other telecoms companies, such as Virgin or sky, are in receipt of such generous subsidies. Yet 48% of rural homes still don’t have internet that Ofcom would describe as merely acceptable. To make matters worse, there are still large numbers of homes who cannot even reach speeds of 2Mbit/s. This means they are unable to make use of online services like music streaming or catch up TV sites. 

Use of EBITDA is a gross simplicity at best, it also ignores pension deficits and indeed that some of Openreach's pricing is intentionally held high by the regulator in order to try and encourage others to build competing networks. The performance figures are out of date as noted. To be quite frank it looks as though someone ran through the Digital Comms Review and picked out bits that were 'on message'.

Page 15, paragraph 15:

BT benefits hugely from these subsidies. They get to spend the money given to them and will own the newly created infrastructure afterward. This effectively means the UK taxpayer is subsidising BT owned infrastructure through Openreach that they will then profit from. BT have also recently been caught in a scandal where it was found that £1.7 million of Openreach revenues had been used to fund the recently approved merger deal with EE, leading to a censure from Ofcom. This raises renewed questions about the takeover and has caused further worries for competition, compounding the problems that already existed over Broadband. It means that BT now has a 40% share of the retail telecoms market and a 70% share of the wholesale market.

The funding is gap funding. It specifically provides the amount of money in between BT's normal 'commercial' spend and the actual amount required to service the subsidised properties. The above comment ignores that there are clawback clauses in the subsidy contracts, worth £129 million as of July 30th, where sales of the subsidised superfast broadband have meant a smaller subsidy than originally provided was actually needed.

Other than the governent going back into the telecomms industry what was the alternative?

As previously noted I have no idea what the retail numbers have to do with divestiture of Openreach, and no idea what the share of the wholesale market refers to given its source is a Guardian article.

Page 17, paragraph 16:

The UK broadband network is largely made up of incredibly outdated copper wires. This technology may have been cutting edge when it was first installed, but today it sees us lagging behind other leading economies, such as Germany, Japan and France. It is frankly alarming that London’s Tech City has some of the worst broadband speeds in the country.

The UK has higher availability of Superfast broadband than Germany or France.

BT are apparently addressing Tech City, however I would ask why BT are expected to service Tech City and no-one else? If the demand is there and it's a viable investment why hasn't anyone else stepped in to deliver the products? BT aren't the only company capable of delivering broadband. There is nothing beyond the UK's regulations, costs of deployment and the willingness of companies to pay stopping others.

http://www.itpro.co.uk/broadband/25271/bt-answers-tech-city-demand-for-more-fibre-in-london
http://www.v3.co.uk/v3-uk/news/2425505/bt-plans-tech-city-fibre-broadband-rollout-expansion

Page 17, paragraph 17:

BT’s approach to the need for faster broadband is to resort to G.fast. This technology makes use of more intelligent signalling kit to push more data through the existing copper wires allowing for the upgrade to ultrafast internet. This should allow for speeds of between 150Mbit/s and 1Gbit/s. However, it can only achieve speeds at the higher end of this scale in limited circumstances and that pure FTTP cables would be needed to guarantee these top speeds. Rather than acknowledging the need to comprehensively rethink their broadband investment strategy, they are instead effectively postponing the decision and trying to strain every last bit of profit they can from the outdated and struggling copper network. Whilst G.fast will increase speed for some it is a reactive measure, a short term fix that won’t address the long term need. 

Just-in-time CapEx is a legitimate strategy. G.fast as noted earlier will push fibre deeper into the Openreach network with approximately half of the spend being re-usable on deploying fibre to homes.

It is not considered a long term fix any more than FTTC was considered a long term fix. The G.fast range of speeds is considered ample for the foreseeable future. If BT offer the option of 1Gb via FTTP on Demand at more competitive prices as would be expected given G.fast pushes fibre much closer to customers that will serve 'power users'.

Page 18, paragraph 18

What is needed is the acknowledgement by BT that the network needs to be converted to all fibre throughout.

Openreach are aware of this, however they would disagree with the timescales posited.

Demand for broadband is forever increasing throughout the UK as more and more people rely on digital services for work, entertainment and day to day living.

Yes it is. This is actually showing signs of slowing down though as the major driver, video, becomes more efficient and reaches ubiquity. 

By making the jump to an all fibre network we will be following countries like Japan, South Korea and even Spain. Ofcom have shown that as consumers get better download speeds, they consume more data.

Virgin Media's experience indicates that the higher the speed before upgrade the lower this increase. Very much a case of diminishing returns as people do the same things but more quickly.

Incidentally what's the big driver of FTTP in Spain? Competition. Telefonica didn't need separation, just a third party to come in and invest rather than relying on them to deliver everything and complaining about the manner in which they do. That and of course that while Ofcom require BT to allow access to all their products on an equal footing, Telefonica didn't have to wholesale anything over 30Mb. If you wanted to sell >30Mb in Spain you had to build your own network. Perhaps a reason to blame Ofcom and the obsession with retail competition that has allowed ISPs to avoid investing in their own networks as they can compete just fine with BT's own retail operations while having BT bear the risks of investment.

Whilst in the short term most users could make do with speeds that G.fast is capable of providing, this is hardly a compelling argument for the status quo. Demand will keep increasing and it makes sense now to invest in future proofing the network, not papering over the worryingly large gaps.

Neither is it an argument for separation of BT Group, and it does not by default make sense to invest in future proofing the network as this assumes all cash is on-hand and the investment doesn't amortise, which is wrong. BT would have to borrow, have to pay interest on the borrowing, and be writing down the investment.

Paragraph 19 I'm not going into depth on. Business parks tend to be passed by because there are few premises there so harder to make the economics work to deliver broadband to them. Repair and install times are below par, no question, but hopefully improving. There's no reason to think separation will improve these factors as it doesn't address the natural monopoly scenario. I've discussed the dissatisfaction above.

Page 20 paragraph 20 has no real content to discuss.

Page 20 paragraph 21:

This report considers that there is only one real option that would satisfactorily address the Openreach question; the structural separation of BT and Openreach. This means BT would be forced sell off Openreach so it becomes entirely separate company. This would directly address the current reasons BT has to discriminate against competitors. As well as this it would also increase Openreach’s incentives to invest in the network and improve on their issues with performance and customer services. 

I do not see any explanation why such an action would increase Openreach's incentives to invest in the network or improve their issues with performance and customer service, They would remain the 'natural monopoly' discussed above, able to continue to behave in the manner of a monopoly.

BT Wholesale would remain their largest single customer, BT Retail the largest customer of BT Wholesale and hence there would remain a perfectly legitimate chain of influence between supplier and their largest customer. This report has presented no instances where Openreach are discriminating against other customers in favour of BT Wholesale. Openreach do not deal directly with BT Retail. There is no reason to believe Openreach's deficiencies are not affecting all customers equally.

Page 20 paragraph 22:

This report believes that formally separating BT and Openreach into two fully separate companies would be of immense value to the UK digital economy. The current model actually constrains BT. It makes perfect sense for BT to favour the status quo and underinvest across all the broadband services it provides through Openreach. It has a ‘natural monopoly’ and severely restricts proper competition. This hurts all internet users, as well as the wider UK economy. The current arrangement is a hangover from the days of nationalised businesses.

This reiterates paragraph 21 and is equally lacking in detail. As previously noted the report confuses natural monopoly and vertical integration.

I would suggest that, rather than freeing Openreach to invest separating them from the rest of the group leaves a for profit company directly at the behest of shareholders who will be wanting a return on their investments. For them spending as little as possible would likely present the best possibility of returns, as with the status quo. In the absence of commitments from third parties to invest, and I would point out that of the two loudest 'voices' in this discussion one doesn't have the funds to invest and the other has steadfastly avoiding committing to do so, I see no reason why separating Openreach from BT Group would have any short or medium term impact other than to ensure Openreach do not invest in their network.

Page 20 paragraph 23:

Under the proposal in this British infrastructure Group report, Openreach would operate as a totally independent company, no longer tied to BT.
Repetition.

By opening up to competition it would ensure that Openreach could turn to investing in the future and focus on digital innovation.

I am still completely unclear as to how separating Openreach would open anything up to competition. They remain a natural monopoly, they aren't going to be bidding against anyone else for use of their own copper and fibre assets. If suppliers wish to reach customers connected to Openreach's network they will have to pay Openreach to do so.

The only way to compete outside of some business areas with Openreach is to build a new network, and then allow wholesale access to that network. The only alternative network of any scale going to homes and businesses in the UK belongs to Virgin Media and is a closed network, no wholesale access.

BT's labs have a long history of innovation, and Openreach are at the forefront of G.fast, a technology that telecomms operators worldwide are planning to use in a variety of ways.

Rather than having to obey the orders of BT it could search out long term investment and partners for itself. It could seek this from all manner of sources, such as pension funds, and open up the market to greater competition. Meanwhile real competition would force the pace of investment to increase as companies would not be able to rest on their laurels and compete for custom by offering better speeds and cheaper deals.

Rather than having to obey the orders of BT it would have to obey the orders of profit-motivated shareholders.

Again what competition would force the pace of investment to increase? Openreach aren't going to compete against themselves and suppliers will only be able to offer the speeds the Openreach network delivers them at the prices that allow them to make a profit while paying fees to access the Openreach network.

The only way this could change would be for Openreach to allow access to all of their ducting so that others may place their own fibre in it and indeed for Openreach to replace the copper in there with fibre where others don't want to leaving suppliers to add the technology either side.

This would involve Openreach spending tens of billions over several years to make their current copper, FTTC and FTTP networks redundant and lose all of their current revenues in favour of becoming a dark fibre supplier to homes and businesses. Who is going to invest in that on a nationwide basis? The losses in ultra-rural areas would be immense.

The complication of deciding who has to fix what and who pays if there is, say, a fibre break affecting an Openreach owned duct with multiple suppliers' fibre inside it a genuine overhead. If in doubt are Openreach to pay for everything? Is this to be the attractive business case to institutional investors that will ensure money pours into Openreach or is it far more likely the business would end up run down, even more poorly performing, and potentially asset stripped where feasible?

To avoid all of this would require way more regulation than there is currently. A strange thing indeed for Conservative MPs to desire.

Page 21 paragraph 24

Unless BT and Openreach are formally separated to become two entirely independent companies little will change. They will continue to paper over gaping cracks. Whilst rural SMEs and consumers are left with dire speeds, or even no service at all, Openreach makes vast profits and finds little reason to invest in the network, install new lines or even fix faults in a properly timely manner. The time has come for a bold and comprehensive solution, full separation and deregulation will provide that solution. 

A lot has changed, some of it for the worse, some for the better. The only way to prevent extensive unintended consequences as a result of full separation and deregulation is to renationalise Openreach. Anything else leaves that company, and the UK's telecomms infrastructure, beholden directly to investors whose main concern will be profit and loss. Deregulation opens up another level of unintended consequences from preferential treatment for profitable urban areas while rural areas see poorer service as there is no profit in serving them equally through to predatory pricing in those areas where there are competing networks.

Does a Conservative, or any other, government have the appetite to renationalise Openreach, taking on the prodigious pension deficit, in order to spend tens of billion reducing its revenue stream?

Note: This article is copyright - if you want to use it please ask.

Changelog:

25th January 2016 - added new paragraph with apparently correct Reuters link; corrected typos, tidied up formatting issues.