This is an article posted in News
Connecting Everyone
The issue surrounding broadband connectivity for everyone continues to plague policy makers and governments all round the world, but no consistent way forward has emerged. The inescapable truth is that it costs money to provide any broadband service, and that money has to be recovered, with a profit, or the provider makes a loss and stops providing the service. We need to remember that this problem is global, although countries with a lack of fixed infrastructure does make things more urgent and technically challenging. Even in the UK, the government has just announced a further $500m to address wider delivery of broadband on top of the $1bn it has already put into the scheme. So at the moment, where it doesn’t make commercial sense for an operator to provide a service, intervention is the only option.
We need to look at the problem by breaking it down into its constituent parts to see if there is a better intervention method:
Technical:
There are a multitude of ways to deliver broadband, most are tried and tested and require fibre or wireless technologies, all that differs is the way they are deployed, be it satellite or drones, ducts or towers. It all comes down to an access network, then the backhaul, to a node or data centre, then connected to the internet. The critical consideration for operators is always the Capex and the Opex needed to deliver the services, which they simply balance against the revenue that can be generated to see if there is a sustainable business case. In say a dense urban area with the majority population being middle-class, then the revenues are high and the technology utilised can be the best possible giving the highest service quality, say fibre to the home or LTE-A. In a rural area where the population density is low, then almost regardless of the affluence of the people covered, revenues will be low, so operators either ignore the area, or deploy the cheapest network possible to lower the breakeven point. Bottom line, a mixed technology and deployment strategy is now more than ever must be the goal of all operators, Capex and Opex must reduce as the revenue opportunity decreases. You can make a general rule that the further the access network gets from the major urban areas the cheaper it must get to deliver the service, both in Capex AND Opex terms.
Regulatory:
The regulatory involvement goes back a long way, with things like the Universal Service Obligations written into many operators license. Simply put, this is an annual levy on each operator to contribute to a central fund for expanding access networks to under-served locations, or put another way, to where there is no commercial reason for putting in a broadband service because it will lose money. Unfortunately this is normally a one-off contribution that often does not reflect the ongoing costs of delivering a service. In the majority of cases the regulator also imposes obligations on operators to provide services “Nationally”, but such obligations are seldom enforced, often due to the tacit recognition that by doing so the operator will lose money and not be able to invest in the core services that fuel its business. The question of what “Nationally” means is also problematic, as EE has just shown in the UK, they have 95+% population coverage but only 75% geographic coverage. Regulators normally go with the population as the measure, but this further condemns the rural population who all sit in the 25% or more where there is no geographic coverage.
Commercial:
At this point, it might be worth looking at a typical operator metric that shows the revenue generated by, in this case, each base station. This chart is idealized and showing only 40% of all sites lose money, often there is a much higher percentage, but of course the fewer sites losing money the better for the operator.
All wireless access operators have charts like this and they show that the majority of sites just don’t make money. In a voice sense, it is obvious this will be the case, you need to have contiguous coverage on roads between cities and towns, (although often we don’t have), it is highly likely these sites do not generate enough revenue to break even. The customer base is transient by its nature and is whoever is driving down the road over any given month, and happens to use their mobile, or at least the passenger uses their mobile so no laws are broken…..
The cold hard fact here is that around half of the sites in any network will not make enough money to cover their costs, and that’s without a strong presence in rural locations. So in Semi-Urban and Rural, there is rarely a commercial reason for offering a service for voice, and even less to offer data. Same for fibre, the cost of trenching and installing fibre over long distances for small populations or widely spread populations has no business case.
Socioeconomic:
Digital inclusion is a core policy for all governments now, which means everyone must have access to broadband of some sort. All economies need to move into the digital world, delivering government services and programs, improving economic outputs or moving further into a service based economy. But there are several questions we seem to miss, should everyone be able to access broadband from anywhere, what devices are in locations to access the services, how do users pay for the service and does the service really translate into real tangible benefits. It is clearly not the case that Semi-Urban or Rural populations will be needing broadband to enable their service based economy to grow, some will exist, but the vast majority of their outputs will be agriculture or commodities, so the need for broadband has to be balanced with reality. The current drive by governments is “build it and they will come” which is somewhat of a failed approach, as far as I know no one has been brave enough to really examine broadband access on a commercial needs-based approach across an entire population spread.
What to do, as some of the above seems to be pulling in different directions….
Taking the broad topics in the same order:
- Under Technical, as the density of customers/revenue thins out, then the choice of technology needs to be progressively cheaper, adjusting the breakeven point downwards. We could assume, Fibre for dense urban, LTE for the same – plus Urban, maybe a mix of LTE and WiFi for semi-urban, Wifi for Rural. Basically, minimise Capex and Opex in balance to the revenue that can be generated. Backhaul also needs to come into this equation, but the same rule applies, so the choice of satellite, microwave or meshed WiFi is driven by cost. The good news is that all this technology is available, we just need a reference design to become a standard approach, identifying the right horses for the right courses and how they all integrate.
- Under Regulatory, stay within the current requirements but define national geographies, down to SqKm, where operators may compete directly with each other with the option to duplicate infrastructure as they see fit, but in less dense areas where the commercial imperatives breakdown, force infrastructure and spectrum sharing to lower Capex and Opex. Through the reference design above, develop a target price for service coverage so that the appropriately priced technology is deployed. The definition of coverage in each SqKm also needs to be made, perhaps for rural situations, coverage in the centre of the community is sufficient, so this become the requirement – abandon population and geography and determine a social or community measure. Maybe somewhat controversially, but allow differentiated tariffs in areas to reflect the cost of delivering the service. It may well mean people in rural areas paying significantly more for the service, but this can be alleviated, (see later).
- Under Commercial, it is highly likely that regulators will start to ensure that value for money is being delivered for data services and to have a view on the cost of delivery for each operator against their tariffs, similar to the way water and power companies are regulated in the UK. If this is taken further and broken down into a more granular view, say at each site or node, then it is easier to give some form of support or credit to the operator who finds it is uncommercial to operate a service at a particular location. Of course as the regulator will be able to see across multiple operators, it can decide if any loss making is self-inflicted by poor service or marketing, or due to the location where the service is delivered. Intervention could be in the form of monetary subsidy, forcing infrastructure sharing or working with operators to deploy cheaper solutions that maybe give a good service in a restricted location.
- Under Socioeconomic, There are some good reasons for digital inclusion, it can save governments and citizens a lot of time and money when things are done on-line and it also allows better outreach of services. So for semi-Urban and Rural areas, the government is an anchor customer for broadband services, or should be, which may in itself be enough to allow the pricing in these areas to be considered somewhat marginal, allowing prices to be similar or even below that of the urban areas. In addition, government may choose to subsidize the services to allow a lower pricing regime to be in place to stimulate uptake, it could even be differentiated to what customers in those areas do with the broadband service. If you watch Netflix/YouTube, it is a choice and you pay the true price of the delivery of the service, if you access government registered sites it is much cheaper….all well within the control of the operator’s billing engines. In effect the government is in the end doing some form of wholesale deal that is negotiated and agreed by the regulator on their behalf, with proper commercially driven review points and rules around costs related to maintenance and improvements.
What this all speaks to is a partnership between the actors in this ongoing drama. The leadership necessary for this is not government, or the operators, it is probably the regulator, but regulating with a clear understanding that the objective is to give the most efficient delivery of broadband services, but wherever possible without regulation. But, where regulation is the enabler to make broadband as widely available as possible, to bring all sides together and force a pact to get the best solution. This will require regulators to have much wider skill sets than they currently have, or their ability to truly interact and drive for the best outcomes will be compromised. As with all things, once the widely inclusive services are initially delivered, the branches and twigs of the network tree will grow stronger and the regulator can retreat when the commercial equations are balanced. In a strange way, you may see the principles of network sharing travel further back into the urban areas as operators see the strength of working together and sharing resources, focusing on services rather than using infrastructure and spectrum as competitive advantages.