March 2021

When the UK’s 700 MHz & 3.6-3.8 GHz auction wrapped up its principal stage last week, one of the most interesting aspects were the attitudes of different operators to the low-band 700 MHz spectrum on offer. Specifically, EE invested heavily in this lower band frequency, while Vodafone avoided the 700 MHz spectrum altogether. 

A big part of the decision can be explained by the existing makeup of the operators’ low-band holdings. Vodafone and O2 both have significant low-band assets already, across both 800 and 900 MHz which are used for 4G and 3G respectively. Vodafone announced that it plans to convert this 900 MHz spectrum into 5G capacity, and as such, did not need to invest further in the 700 MHz on offer.

EE, on the other hand, had just 10 MHz (5MHz up, 5MHz down) of 800 MHz spectrum available before the auction. Its acquisition of 2x10 MHz of paired 700 MHz frequency brings it closer to Vodafone and O2’s pre-auction holdings, but what is striking is the impact of the additional 20 MHz of supplementary downlink spectrum that the operator acquired. With this, EE will now have more low-band downlink capacity than Vodafone, even though its uplink capacity remains smaller.

Why does low-band spectrum matter?

EE has long been established as the dominant provider in the UK in Tutela’s reporting: in September’s State of Mobile Experience report it demonstrated a speed advantage over the next closest competitor of over 10 Mbps, and also took the top spot for Excellent Consistent Quality, Tutela’s metric of how often a connection is good enough for demanding applications like 1080p video streaming, HD group video calls and real-time mobile gaming. With this in mind, it would seem EE’s existing assets are performing well – so why would more low-band spectrum be required?

The same report also revealed close competition with Vodafone, particularly in Wales and Scotland. A significant contributing factor here is the greater proportion of rural area in both of these nations, which is where EE has traditionally been more reliant on 800 MHz due to the better coverage traditionally achieved with lower frequency spectrum.

Heavy utilization here is defined as more than 50% of LTE data traffic using 800 Mhz as the primary band

Why wasn’t the existing 800 MHz holding enough?

Examining EE’s 4G performance by the primary LTE band also tells an interesting story. For devices that report being connected to 800 MHz as the primary band, not only does performance fall significantly lower than other primary bands, but there is a noticeable decline in performance as usage increases throughout the day. This suggests that the spectrum is overloaded – hence the need for supplemental resources.

In practical terms, this translates to a significant decline in performance during congested hours in 800 MHz-dependent areas of up to 41.9%. This dip is over twice the size of that in regions that are not as dependent on 800 Mhz.

Heavy utilization here is defined as more than 50% of LTE data traffic using 800 Mhz as the primary band. Best hour and worst hour represent the hour with the highest and lowest Excellent Consistent Quality  across 180 days of data.

What supplemental downlink means for EE’s long-term network planning

As it currently stands, EE’s strategy for providing connectivity in hard-to-reach areas relies on one solitary, heavily-used block of spectrum. By acquiring 30 MHz of low-band downlink capacity — more than all the other operators combined — EE has shown that it believes low-band spectrum can provide sufficient network capacity for rural areas long-term. 

It’s a different strategy to the one being adopted in other markets, where mid-band spectrum is being hailed as the future — a future that will require dramatic site densification to provide coverage. By acquiring a significant portion of the low-band spectrum to deploy 5G, EE is presumably hoping that it can limit the expensive process of building tens of thousands of new sites, and make its low-band 5G more than just a “coverage layer”. 

Of course, time (and the growth of network traffic) will determine if EE’s new low-band spectrum holdings are sufficiently wide to avoid major rural site densification. But at worst, EE has addressed an existing low-band capacity issue; at best, it’s laid the foundation for a less capital-intensive 5G buildout in the near future. Either outcome justifies EE’s spend in this auction as it works to provide a consistent and reliable mobile experience to its subscribers.

Tutela’s data can help operators understand how their current spectrum holdings are performing, as well as identify areas where demand and capacity are misaligned. With up-to-date subscriber-focused insight into the current state of network performance and user experience, operators can make smarter decisions about where, when and how to invest in their networks – from spectrum acquisition to infrastructure deployments – and then monitor the results of that investment on an ongoing basis.