After all the waiting, the 5G era is now with us. EE switched on its 5G service in six cities at the end of May. Vodafone launched at the beginning of July, and the other big players are sure to follow close behind.
But as they do this, operators will be carrying out a cost/benefit analysis, at the same time, to calculate as accurately as they can, how much 5G will cost them and how much they can charge customers for it.
It is important to distinguish the two potential phases of 5G New Radio (NR) that we will see, as they have markedly different costs attached to them.
The first wave, starting this year – NSA 5G NR – is expected to make use of existing base stations with new 5G antenna placed on them. SA 5G NR, which would require a larger number of (predominantly) new base stations, may not hit the UK or Europe for three or four years, other than in isolated high-use urban scenarios.
The cost of creating NSA 5G NR will be much lower than for SA 5G NR. However, it will still be significant. There are perhaps 50-60,000 base stations in total around the UK, and the vast majority will (in time) need 5G antenna. That represents a major investment. On top of that, the installation costs will be considerable. Many of the base station towers will also need reinforcing as the new massive MIMO antenna required for 5G services can weigh in excess of 40kg.
Some of these costs will be offset by a rise in network sharing. We will see an increasing move from the passive sharing of today (where operators share the same tower but each have their own radio antenna) to active sharing (where operators share not only the tower but also the antenna).
SA 5G NR will be even more expensive. New base stations will be needed – and a lot more of them. We will see a process of network densification with smaller base stations (or micro cells) being installed in urban areas to boost capacity.
So, how much is 5G likely to cost and how much will it return? Working this out is not as simple as it sounds.
Operators have already sunk significant sums in buying spectrum from Ofcom, with the first auction in April 2018 raising around £1.4 billion. Ofcom is working on making more spectrum available for future wireless broadband services by December 2022.
The cost of building out an actual 5G network is complex to calculate. Some costs are sunk already in creating the existing 2G, 3G and 4G networks, with incremental costs on top for the new 5G antenna, increased backhaul capacity and other equipment, such as transmission equipment. Estimating the incremental or fully allocated cost of a 5G service is not straightforward. But, operators need to do this in order to understand the real costs and returns.
What really matters is whether they believe they can justify the network upgrade costs required even for NSA 5G NR: over what time period and at what take-up assumptions will a return be achievable? This is a fundamental question for investors in telecoms stocks, at a time when mobile operator revenue, profitability and cashflow are all under pressure, leading to share price weakness.
Much depends on customers’ willingness to pay more for 5G than for other services. Vodafone has taken a novel approach to pricing, based on unlimited data packages, but with tiered speed categories so that subscribers pay more for a faster connection, similar to many fixed broadband packages. 5G handsets, however, will be expensive. So levels of take-up, at the moment, simply remain unknown.
Another factor that will impact on the speed, success and therefore the profitability of 5G NR rollout is the availability of 5G antenna. At present, in Europe, there are only three credible suppliers of these – Huawei, Nokia and Ericsson.
Of these, operators say that Huawei is the leader in producing antenna in volume and to the required specifications. The political issues surrounding this will need to be resolved. Mobile operators have stated that the speed and coverage of 5G rollout is highly dependent on the availability of suitable radio antenna.
The widespread availability of 5G handsets is also likely to be a significant factor in the speed of 5G take-up. We note that there are relatively few handset providers with the capacity and technological capability to produce 5G handsets at scale in the short term. We expect at least Apple, Samsung, Huawei and ZTE to be capable of this, but there may be others too.
However profitable 5G may or may not be, operators are committed to launching it, albeit in a limited coverage area initially. The returns achieved on the first wave of 5G are likely to dictate further coverage plans, both for NSA 5G NR and for potential roll-outs of SA 5G NR.
For now, it is a commercial and competitive imperative to launch a service. Operators also need to do so for the simple reason that they don’t have enough capacity on parts of their 4G networks, affecting download speeds in dense areas.
Turning these imperatives into accurately costed and profitable lines of business is the challenge that faces the telecoms industry today.
Read the next blog post: