Some observers expressed surprise at Reliance Jio’s acquisition of 10 MHz of spectrum in the 700 MHz band, in the recently concluded Indian 5G auction. Yet Jio’s arch-rival Bharti Airtel felt moved to explain why it had spurned that band. The distinction between the two approaches reflects partly the spectrum holdings the two operators currently hold, but also differing views on the value of moving quickly to support 5G Standalone (SA), which is mainly needed to support the most demanding applications, especially those requiring ultra-low latency as in the industrial IoT and advanced gaming.
Airtel regards such use cases as lower priority for India than establishing more widespread high speed broadband connectivity, given the paucity of fixed line options for the majority of the country’s population. Airtel came into the auction with the strongest midband holdings for 4G, which still has a lot of mileage in India. This made it more attractive to deploy 5G Non-Standalone (NSA) since that runs with the existing 4G core, which is going to be around for a long time.
Airtel CEO Gopal Vittal also explained just after the auction that with 5G NSA, the 3.5 GHz midband spectrum would support greater range because it could be dedicated to the downlink, while the 4G band between 1.8 GHz and 2.1 GHz would still carry uplink traffic at sufficient speed. Given that midband spectrum costs less per-MHz than low band spectrum, this was also more cost effective, Vittal insisted.
Meanwhile, Jio had been drawing attention to its acquisition of 700 MHz spectrum, which offers lower capacity and bit-rate than midband, because not a lot of spectrum is available, but is more valuable per-MHz because range and coverage are greater at the lower frequencies. This led Jio to claim that, equipped with this low band spectrum to complement the midband holdings, it would be the only Indian operator able to deploy a true 5G service in the foreseeable future, by combining the midband for capacity and performance with the low band for coverage.
We have seen such arguments over the nature of ‘true 5G’ elsewhere where deployments have already occurred, including the USA, where first T-Mobile and then AT&T, made hay by claiming wider coverage than its rivals, by using 600 MHz low band spectrum for wide coverage. This led rivals to argue that the performance achieved at 600 MHz was often merely comparable with 4G, irrespective of whether the network uses 5G Standalone or not.
This argument has now surfaced in India ahead of deployments, with Airtel claiming that those areas falling back to 5G SA at 700 MHz would not in fact enjoy true 5G service, citing tests conducted internally. “In fact, we have seen in our trials that the sub-GHz layer on 5G SA gives only 8-10Mbps speeds, no different from 4G,” said Vittal. That was one reason Airtel decided not to bid for 700 MHz spectrum, he indicated.
Vittal went on to cite other advantages of 5G NSA at this stage, including superior coverage in those urban areas where the midband is already providing the uplink over 4G. That equates to an average extra 100-meter range, Vittal argued. He also pointed to the undeniable fact that almost all current devices can support connectivity over that mid band, even if that benefit will be eroded over the next few years as more support 5G SA. For now, in countries where 5G SA has already been deployed, most of the traffic still runs over NSA networks, as Vittal pointed out.
The final advantage of NSA by Vittal’s book follows from Airtel’s large midband spectrum holdings, which allows faster uplinks and therefore lower voice call connect time.
All these other benefits, though, are rather secondary to coverage, and none would be decisive on their own. Jio’s argument is that although the low band spectrum will not provide the same performance as the midband, it provides a perfectly acceptable fallback while many users in urban areas will then be able to enjoy full 5G SA performance. Jio will then be better placed to target enterprises with private 5G using network slicing for demanding use cases.
It remains to be seen which of these two strategies will work out best, but at least it gives Indian consumers and business users a choice. It is certainly true that Jio has spent a lot for its 700 MHz spectrum in terms of bandwidth. It is also true that the high reserve price deterred any bids for 700 MHz at the previous two auctions it was available for 4G, in 2016 and 2021. This time the reserve price had been cut by 40% to INR3,927 crore/MHz ($493.2m), which is still a lot, reflecting the size of the Indian market. The key change though in Jio’s eyes at least was the big prize of early mover advantage in 5G SA, and with more 700 MHz spectrum to come at a future auction, Airtel may well be forced to come in then and pay an even higher price.
Jio paid INR39,270 crore ($4.92 bn) for that 10 MHz of 700 MHz spectrum, out of its total auction spend of INR88,078 crore ($11.1 bn) for 24,470 MHz. Airtel spent INR43,084 crore ($5.4 bn) to acquire 19,867.8 MHz of spectrum.