Is the hammer hovering over the G.fast roll-out?
G.fast has had a difficult birthing in the UK but from our figures it is an ultrafast service available to around 1.885 million premises and provides a speed upgrade path for people where the top VDSL2 package is not enough speed and no other options exist or they just don't want to change provider.
So we learn that Openreach is now re-assessing its investment in the G.fast roll-out and of course the alarm bells ring with chants of the G.fast roll-out ending. Unfortunately like so many topics things are more complex and our understanding is that the work for 2020 will continue which is aiming to get G.fast to be available to some 2.73 million premises by end of March 2020 (at end of last quarter the Openreach G.fast footprint was listed as 2.166 million premises).
What is happening, we are told that the issue is about getting better at avoiding overbuilding G.fast pods with FTTP via the Fibre First roll-outs and with a number of Fibre First exchanges having an extensive G.fast footprint this makes sense, i.e. why roll-out an ultrafast technology twice. Surely much better to sit down and re-plan the G.fast roll-out so that based on the next few years of known FTTP plans new pods will only appear where FTTP will not. That was the original stated aim of the G.fast roll-out i.e. G.fast where FTTP is not but we are in a situation where 27,891 premises have both G.fast and Openreach FTTP as an option today.
In terms of the take-up of G.fast in the last financial results the 36,000 live lines from a footprint of 2.1 million looks tiny and its very easy to say that this pause is due to low take-up. As we said earlier things are more complex, ahead of the last Openreach financials we thought the take-up of G.fast would be around 38,000 to 48,000 premises so not massively wrong and using the same source data in the Fibre First areas (i.e. where VDSL2 is also superfast) we believe that Openreach FTTP take-up is in the 32,000 to 42,000 lines region.
On the topic of the the broadband speeds people want and will more importantly pay for the buying choices for those in Openreach FTTP new build areas should be a good indicator of what people are happy with:
At the end of Q2 2019 in Openreach FTTP new builds speed tests indicated:
- 36% are on a 40/10 FTTP service
- 62% are on a 55/10 FTTP or slower service
- 80% are on an 80/20 FTTP or slower service
- Just 20% are clearly buying the ultrafast speed options
Of course if the price of the ultrafast options was to be slashed then take-up would probably increase and the beauty of FTTP is that in a few years when we need 500 Mbps to just check our email and whatever social network is in fashion will scale to handle that.
This also illustrates why new entrants are so aggressive with their full fibre pricing, i.e. to get high take-up you either need to roll-out in an area where there is no superfast/ultrafast service already available or you need to undercut the existing service even if you are a lot faster.
So G.fast roll-outs have not ended just yet, our money is on the Openreach FTTP roll-outs now taking the position of priority at both planning and resource levels giving Openreach an ultrafast footprint of 18 to 19 million premises once it has finished its 15 million premises FTTP roll-out. A reminder on the FTTP plans, target of 4 million by the end of March 2021 and ambition for deliver 15 million for 2025.