An effective UK response to Covid 19 will require over-ruling current Ofcom policy

1) The Covid19 pandemic accelerates changes that have already overtaken Ofcom’s ability to plan ahead

The effect of corporate and individual responses to the Covid-19 pandemic is to expedite changes in demand that make a nonsense of Ofcom’s recent market reviews. They also overtake most of the responses to the consultation currently under way. Effective and timely responses to some of the challenges posed by Covid 19 require immediate change to regulatory policies Ofcom agreed with its peers in BEREC, for rubber stamping by the Commission. The changes needed appear unlikely to require primary legislation. They may, however, require the Secretary of State to set new priorities for Ofcom, in accordance with the 2003 Communications Act .

2) Ministers should set policy. Regulators should implement it.

The Labour Government appointed its Telecoms advisors to run Ofcom and abdicated from setting policy. I described the consequences back in in 2015. They included the near bankruptcy of BT (to protect the Cable companies from being sold to Sky for £1) and the collapse of UK investment in full fibre. We are still near bottom of global league tables. DCMS, under the coalition government, started to take back control in 2011. It is time to complete the process and revert to division of labour under the Conservatives when Sir Bryan Carsberg regulated price and behaviour within policies set by Ministers.

The mantra: “Ministers set policy, Permanent Secretaries implement it”, was rarely true, even under Mrs Thatcher. But delegating policy to a regulator, (who felt more answerable to their peers in an EU “Regulatory College”, BEREC , than to Ministers or Parliament), has been shown to be a step too far. Brexit gives the opportunity to correct matters. The need for an effective response to Covid 19 and the switch from personal to on-line contact provides the urgency.

3) The impact of  Covid 19 on local, national and global communications networks

Conferences around the world are being cancelled. There are growing calls to work from home to reduce the risk of becoming infected with Covid 19. The delivered speed of many broadband networks was already falling – because investment in local fibre connectivity has not been matched by investment in backhaul and switching. We can expect those falls to accelerate.

4) UK backhaul and switching is already overloaded

The headline figure that 1 in 7 UK premises has theoretical access to Gigabit Broadband conceals growing problems and delays in delivering that access in practice. Whether the service is called Superfast, Hyperfast or Gigabit the delivered speed of response depends on a lot more than the capacity of the final cable or local mast.

The broadband cover (both fixed and mobile) for the Ross of Mull (the road to Iona) stops dead in summer, when thousands of tourists arrive.  The reason is less the need to upgrade the equipment on the local masts than capacity problems between Oban and Glasgow.

In South London, in the shadows of the Crystal Palace and Beulah Hill masts there are similar problems as traffic grows and network capacity stagnates. BT may have “enabled” its exchanges and Virgin may have wound up the nominal speed of its networks but neither promotes full fibre business connectivity except when threatened by competition from third parties. They cannot handle the extra demand without upgrades that will not cover their cost unless they are allowed to charge up front.

5) The reason is simple economics.

In Scandinavia the investment in upgrading local connectivity is commonly covered by up-front payments by the customers (both business and residential). Such contracts are barred by Ofcom for consumers. Even worse, Ofcom is planning to bar them for many business customers An upgrade not underpinned by payment upfront or by a long term contract that can be used to obtain leasing investment, is a “risk” investment. It cannibalises existing revenues and may never make a return unless the investment is calculated against the loss of business to a competitor. This is one of the consequences of current UK regulatory policy.

Local loop unbundling assumes that a single “wholesale” network from a single supplier sold by multiple “retailers” is adequate “competition”. Policy is now to encourage the sharing of poles, ducts and wayleaves. The communications of those critically reliant on their on-line connections (whether for life support or business continuity) should, however, be  multi-sourced, with hot-standby contracts for resilience, over networks which do not share the same single points of failure. Thus I have long had contracts with BT, Virgin, Sky, Vodafone and O2. It is less expensive and inconvenient than being off air when any one or two of them go down, as they all do, from time to time.

6)  Speeds will fall as millions more seek to work from home over the next few weeks or months

From Crystal Palace to Iona, response times will fall.

The speed of network construction will also fall.

The failure to train UK youngsters in the skills to build new network infrastructure (on which I have blogged regularly over the past 18 months) was going to delay network upgrades, even before we faced the problems of reliance on imports of equipment from China. The consequences of those delays now become very much more serious

7)  The need for reliability and resilience will rise.

I recently used the Huawei controversy as the hook for a blog on the need to multi-source telecoms at all levels (from suppliers and routes to equipment and technologies). This morning I began receiving press releases from would-be suppliers hoping to exploit the spike in demand for robust, reliable corporate networking.

BT may be a financial basket case, overdue for break-up (e.g. the high risk content business sold to fund investment in critical utility strength converged, fixed and mobile, infrastructure) but it will be part of any UK multi-source strategy (alongside Virgin and Vodafone) for the foreseeable future. The network components may be increasingly be built (and initially operated) by players like City Fibre, Gigaclear, Hyperoptic but BT (like Telia in Sweden) will amost certainly remain the main national operator. The spike in demand for corporate networking facilities is likely to give a rise in revenues for BT wholesale which will make it easier to organise a management-led break-up.

8) Will DCMS and Ofcom will give BT the freedom to respond.

The current Ofcom regulatory model is a major obstacle to allowing BT and its competitors to use up-front contract payments to fund not only local fibre connections but the removal of backhaul and switching bottlenecks.

DMCS and BEIS should help industry implement the recommendation in the 2017 Conservative Manifesto for the third-party mapping of utility (including telecoms) networks. A neutral (perhaps hosted by Ordnance Survey) service would facilitate an open market in wayleaves, duct, fibre and switching access network and co-operation between suppliers in funding alternative routings (including hot standby across compatible but independent networks) to the many “single points of failure” in our current networks. Such standby should also, of course, be available to BT on similar terms. It would also help reduce the risk of “accidental excavations” removing on-line access for days or even weeks.

9)  Until they do UK networks are likely to remain than most to Covid-19 related overload and failure. 

The OECD data indicates that the UK is more reliant than most on crapband (copper rust and assorted pollutants) in the local loop. This rations the load on the back haul and switches. Meanwhile Ofcom prevents BT from using upfront contracts to fund new capacity.

The time has come for Government to join up the dots.

That should not mean abandoning competition. It should, however, mean changing the goal posts. The target should be “at least three inter-operable networks, each operating to critical infrastructure standards, which do not share the same single points of failure” to as much of the UK as practical.

Data Center
Data Management