Hellish year for 5G could wipe $6B off RAN sales

Dublin seemed an apt location for the Telecom Infra Project's FYUZ event, largely dedicated, as it is, to chattering about the future of the radio access network (RAN) – perhaps the biggest capital expenditure item for the average telco. The Irish air has an icy touch at this time of year. By dusk, the River Liffey looks glassily static and uninviting. Ireland's favorite tipple is the color of night. Today's RAN market, with its similarly gloomy appearance, fits well into the narrative.

The revenue situation has deteriorated further, according to the two analyst firms best known for tracking it – Omdia, a sister company to Light Reading, and Dell'Oro. Last year, revenues generated from the sale of RAN products plummeted 11%, to about $40 billion, according to Omdia's data, as telco chief financial officers put the kibosh on expenditure. Until August, Omdia had been expecting another decline of between 7% and 9% in 2024. But it is now forecasting a drop of between 10% and 15%.

Given the number for 2023, this would imply an actual revenue decline of between $4 billion and $6 billion, reducing sales across the entire market to as little as $34 billion in the worst-case scenario. The revision is attributed mainly to a worse-than-expected slowdown in Asia and Oceania, and especially China and India.

Dell'Oro is broadly in agreement. In a LinkedIn post this week, Stefan Pongratz, who leads the work on RAN data gathering and forecasts, said revenues for the first three quarters of 2023 were down year-over-year by at least 10% and possibly as much as 20%. This, he said, is "paving the way for one of the worst years since we started tracking the market in 2000." The last double-digit decline occurred way back in 2016, according to Dell'Oro's data.

No money in 5G

Vendors have blamed the slump partly on a long-running inventory correction in North America, where big telcos behaved like those pandemic-era shoppers who overloaded on toilet roll, filling their larders with all manner of componentry to avoid trips to the Ericsson and Nokia shops for several quarters. There have been signs the likes of AT&T and Verizon have emerged from hibernation and started spending once again. AT&T's guidance, for example, implies it will invest between $6 billion and $7 billion this quarter, up from $5.6 billion in the same period of 2023.

But other factors are in play, too. Mobile data traffic is not growing as fast as it previously did, or as vendors would like, based on year-over-year percentage increases. "Helping to explain this output deceleration is more challenging comps, excess capacity, and monetization challenges," said Pongratz in his LinkedIn post. Excess capacity sounds problematic for suppliers because of the widespread industry assumption that capacity upgrades are more profitable than initial buildouts.

Meanwhile, rollouts to expand coverage have clearly stalled in some regions. Outside China, 5G coverage supported by midband spectrum – the range vendors champion for use with the standard – stood at just 35% in the second quarter of 2024, nearly five years after the initial launch of 5G services, according to Ericsson's latest mobility report. The 25% drop in AT&T's share price over this period tells you everything. Despite some wild expectations, and talk of 5G powering all sorts of new and profitable services, the technology has done zilch to improve the fortunes of the telco crowd.

In North America, where only three big telcos battle for customers and average revenue per user is high, companies have actually made decent progress on midband rollout, according to Ericsson's data. The real laggard is Europe, with its jigsaw puzzle of national fiefdoms where sheep outnumber people and gazillions of operators vie for business. In the absence of customers demanding a 5G connection, rather than just a mobile service that works well, telcos with budgetary constraints and poor returns on investments have limited incentives to spend.

The other problem, which a Trump administration might attempt to rectify, is the disinclination of telcos and national governments to replace Huawei, a Chinese vendor deemed a security threat by various agencies in Europe and North America. Huawei-banning governments have given telcos years to replace it (in the UK, for instance, operators still have more than three years left to complete a 5G swap). Others have resisted bans altogether.

They include Germany, which fears a ban would hurt sales of cars and machine tools to China, its biggest trading partner. Its only requirements are that operators remove Huawei from the "core" network control center (which they had already done) and replace a small part of its management software for RAN and transport infrastructure. Antennas, radios, server equipment, baseband software and other gubbins can stay.

True open RAN in retreat

The big question is whether the industry has now reached a low point from which it can only improve this quarter and in 2025. Remy Pascal of Omdia says the year-over-year comparison for the third quarter was an improvement on what happened in the first two. But this does not guarantee the current quarter will be much better, he told Light Reading.

"Market conditions are slowly improving, but regional imbalances remain significant," Pongratz said on LinkedIn. He anticipates growth next year in North America and the Asia Pacific but expects slower 5G activity in China. There were no ruminations about Europe in his latest post.

Nor did either analyst firm call out open RAN. The concept, which supports the use of open interfaces in the RAN, was the focus of this year's FYUZ and was conceived largely to spur competition in the supplier market. Yet there has been little evidence of progress by smaller vendors. Samsung, the fifth-biggest RAN vendor worldwide, saw its market share drop from 7.6% in 2022 to 6.1% last year, according to Omdia, and Dell'Oro thinks its market share was on the slide in the first nine months of this year, too.

Both analyst firms also cite gains by Huawei and Ericsson. US hawks will be disappointed to see the Chinese vendor picked up market share in several countries, according to Omdia. Ericsson, meanwhile, benefited from a bigger share of business in North America, where it is replacing Nokia across the one-third of AT&T's footprint it did not already serve. The only small vendor to have made significant gains is India's Tejas Networks, said Pascal. Thanks to a major deal with state-backed telco BSNL, it is currently one of India's biggest three suppliers.

Ericsson's sprawl at this year's FYUZ event is symbolic of the open RAN sector, where big vendors have squeezed out smaller ones after pledging allegiance to new specifications. Staff from Mavenir, the only sizable US software vendor, were prominent exhibitors, panelists and speakers at previous shows. This year they barely had a presence. Whatever open RAN is about, it is not vendor plurality.