Europe is pouring money into fiber through the Gigabit Infrastructure Act, but nobody plugs a phone into a fiber jack. The last 10 meters of every apartment are wireless, and the fight over who gets the upper 6 GHz band will shape how those 10 meters perform for the next decade.
I have spent a lot of time with the Wi-Fi side of this debate. Its case is familiar: almost everything indoors connects over Wi-Fi, so Wi-Fi needs more spectrum. To balance the picture, I sat down with Luiz Felippe Zoghbi, who leads spectrum advocacy with governments for the GSMA, joined by his colleague Tom Curran. I played devil's advocate. Here is what the mobile industry argues, where I think the argument is strong, and where the multifamily market still has open questions.
This is also the debate behind Riser, our new event in Amsterdam in 2027. Riser is dedicated to connectivity in multi-dwelling units, where fiber, Wi-Fi, and mobile all have to work inside the same building, and conversations like this one are exactly what we want on that stage.
Doing the Math on 6G Spectrum
The GSMA's core argument is arithmetic. According to Zoghbi, 6G is being designed around 200 MHz channels per operator at launch, growing to 400 MHz later. With an average of three operators per market, that means 600 MHz just to start.
That is why 540 MHz, the portion the RSPG recommended for mobile priority, is not enough on its own. "If it's only 540, yes, Europe would be behind," Zoghbi said. The GSMA also projects that by 2030, cities home to more than half the world's urban population will be capacity constrained if mid-band spectrum stays at today's levels.¹
The GSMA's June 2025 report, 6 GHz in Europe: Mobile evolution in 6.425-7.125 GHz, puts numbers on the demand side. It projects average mobile data per connection in Europe to triple, from 32 GB a month in 2024 to 97 GB a month in 2030.² Channel sizes keep growing too: 5G uses up to 100 MHz, while 6G discussions start at 200 MHz and could reach 400 MHz, as one channel or two aggregated 200 MHz blocks. The report itself concedes that the full 700 MHz upper band only covers the bottom of that range in a three-operator market.³
Zoghbi was careful to frame this as a sizing question rather than a turf war. "The idea is not either or," he said. "It is to find the right amount of spectrum that will allow both technologies to flourish."
"Wi-Fi's bottleneck is upgrades, not spectrum"
The most pointed part of the GSMA case is its read on how Wi-Fi uses the spectrum it already has. Zoghbi cited GSMA studies showing that Wi-Fi 4 and Wi-Fi 5 still account for close to 70% of Wi-Fi usage in most countries, and up to 90% in some.
The same studies also found that use of the lower 6 GHz band remains close to zero in most cities. Zoghbi said even the leading cities sit in single digits, with the United States, an early mover, below about 10%. Usage in the upper band is lower still.
The report's European data, based on Ookla Speedtest scans in 11 capitals during 2024 and 2025, tells the same story. Across those cities, Wi-Fi scans split 22% Wi-Fi 4, 46% Wi-Fi 5 and 32% Wi-Fi 6, and Wi-Fi 4 still accounts for 32% of scans in Berlin.⁴ Among Wi-Fi 6 and 6E scans only, the lower 6 GHz share ranged from 0% in Rome, Madrid and Berlin to 7% in Paris and Stockholm.⁵
His conclusion: the jump in spectral efficiency from Wi-Fi 4 to Wi-Fi 7 is so large, roughly 10 to 12 times by his account, that upgrading devices and access points would unlock far more capacity than new spectrum. "The biggest bottleneck for Wi-Fi is the ability to upgrade to a newer technology," he said.
The speed data backs him up. In the same 11 cities, median Wi-Fi 4 download speeds ranged from 31 to 41 Mbps, against 214 to 496 Mbps on Wi-Fi 6, delivered largely over 2.4 GHz and 5 GHz spectrum.⁶
Mobile lives indoors too
One figure surprised me. Zoghbi said 70% to 80% of mobile data is consumed indoors, and most of that indoor traffic rides on mid-band spectrum, not the low bands the industry once expected to carry it. In GSMA testing, he said, 6 GHz behaves much like today's 3.5 GHz band, with 200 MHz delivering around 5 Gbps and 400 MHz around 12 Gbps.
The report's city sample supports the indoor claim: 71% of mobile data use is indoors and 29% outdoors, and mid-bands provide 85% of indoor connectivity, with just 15% coming from low bands. In the European cities studied, 3.5 GHz carries 71% of urban indoor 5G use, and 3.5 GHz 5G download speeds run 7 times higher than low bands and 2.5 times higher than lower mid-bands. Peak speeds in 6 GHz trials reached 12 Gbps.⁷
I pushed back. By volume, most indoor data lands on Wi-Fi, so the Wi-Fi camp can use the same statistic to argue for more Wi-Fi spectrum. Zoghbi conceded the volume point but drew a different conclusion: Wi-Fi can deliver the best experience indoors with the spectrum it already has, thanks to dense access points and spectrum reuse, while mobile cannot reach 6G performance without new spectrum.
He also pointed to a behavioral shift. People still join Wi-Fi at home and at work, but increasingly skip it in restaurants or on the Tube because mobile is good enough.
Riser takeaway: If 71% of mobile data is used indoors, as the GSMA's city data shows,⁷ the building is where the spectrum debate gets settled. Riser puts property owners, managed service providers, fiber builders and mobile operators in one room to compare what actually works behind apartment walls.
Brussels and London are taking different roads
On the EU side, Zoghbi sees the RSPG Opinion as a workable design rather than a defeat. The 540 MHz comes with a path to more: either the 160 MHz below it, depending on WRC-27, or an extension upward toward 7.25 GHz. Either way, he said, Europe keeps the door open to roughly 700 MHz, enough for 200 MHz each across three operators.
The catch is that each EU member state will assign the band on its own timeline, with its own rules and its own incumbents to move. The proposed Digital Networks Act could also change how that process works.
The 7 to 8 GHz range, in the GSMA's view, is a complement and not a substitute. Reaching 400 MHz per operator for later 6G phases depends on going beyond 7.125 GHz, which Zoghbi called one of the last opportunities to secure that much mid-band spectrum.
The UK is where the GSMA sees more risk. Ofcom's split approach gives Wi-Fi first access to the upper band, with mobile priority in the top 540 MHz arriving later and licenses tied to geographically defined high-density areas. "There is no guarantee that Ofcom is going to go beyond 540," Zoghbi said. He described the decision as one that has to be respected, and said the GSMA is watching closely as Ofcom builds its coordination system and safeguards.
He noted that most countries that have decided so far, including Brazil, China and Vietnam, went with 700 MHz for mobile, while the UK and Australia stand out. His warning for Europe: "The mistakes of 5G cannot be passed on to 6G." On harmonization, he was less worried about devices, since equipment is expected to tune across the full range, and more about cross-border coordination, such as between the UK and Ireland.
Riser takeaway: With Brussels, London and each national regulator moving at their own pace, MDU operators across Europe will face a patchwork of rules for years. Hosting Riser in Amsterdam gives the European multifamily market a neutral place to compare notes across borders.
Will operators actually pay for it?
European operators are under financial pressure, so I asked whether they would really buy and build out this spectrum at scale. Zoghbi said spectrum prices in Europe must come down. Operators are not earning more per megahertz or per user as technology generations change, and high prices limit what they can invest.
He pointed to more than 500 licenses due for renewal by 2035. By the GSMA's estimate, automatic low-cost renewals could save operators up to €30 billion over 2025 to 2035, and renewing at no cost could raise network speeds by up to 23% and add up to €75 billion to Europe's GDP by 2035.⁸ For new awards, the GSMA wants low reserve prices, no artificial scarcity, and coverage or other obligations discounted from the price.
On devices, the GSMA is upbeat. Zoghbi said prototypes are already running, including a Qualcomm-based Oppo device tested in China and fixed wireless access live in the UAE. In September, around 65 chipset makers, vendors and operators signed a GSMA statement backing the ecosystem, and vendors expect commercial devices this decade. Adding the band should be as simple and cheap as adding 3.5 GHz was, he said, not a millimeter wave style cost jump.
Why the GSMA rejects sharing
GSMA leadership has said that sharing the upper band between mobile and Wi-Fi will not work, and I wanted to know exactly what fails. Zoghbi said no study so far has shown workable coexistence, short of geographic separation with large distances.
The technical crux is detection. Licensed mobile equipment can recognize Wi-Fi, but Wi-Fi does not recognize mobile. Sharing would require a database layer on top, and nobody has shown how often it would need to update to work.
The GSMA also ran the economics. In its socioeconomic modeling, forcing mobile to lower power, a typical condition of sharing, cut the band's benefits roughly in half, while it estimates the extra benefit of the upper band to Wi-Fi as close to zero. The GSMA's preferred complement is simple: about 500 MHz of lower 6 GHz for Wi-Fi, indoors and in some outdoor uses, and 700 MHz of the upper band for licensed mobile.
That modeling comes from a 2024 GSMA Intelligence study of nine countries: Australia, Azerbaijan, Chile, Côte d'Ivoire, Germany, India, Jordan, Malaysia and the UAE. It compared three options for the upper band: licensed mobile, unlicensed Wi-Fi, and shared use with reduced mobile power. Licensed, full-power mobile came out ahead in every country, with an average economic benefit 7 times greater than unlicensed use.⁹
Each camp is fighting this battle with its own arsenal: dueling reports, competing analyses, rival GDP estimates and an army of consultants on both sides. So who should a regulator believe? I closed by putting that question to Zoghbi: whose evidence should regulators trust, the GSMA's or that of the DSA and the Wi-Fi Alliance? His answer was gracious. Both, he said, because each rests on real evidence, and the regulator's job is to weigh them against what is best for citizens.
The message to multifamily and managed Wi-Fi operators
I asked what the GSMA would tell managed Wi-Fi operators who depend on wide channels to serve dense residential buildings. Zoghbi called their work "incredibly important" and said that, with existing spectrum plus the lower 6 GHz band, they will have the bandwidth they need in most cases.
What the GSMA could not offer was MDU-specific analysis. Roughly half of Europeans live in multi-dwelling units, and many of those buildings struggle to get reliable indoor coverage even for voice calls. Zoghbi said detailed indoor network design sits with another GSMA team; the spectrum team factors densification and investment into its models but does not study MDU propagation directly.
Riser takeaway: Nobody owns the MDU indoor question today. The spectrum teams model the macro network, the Wi-Fi camp models the home, and the building falls in between. Closing that gap, with data from people who deploy in real buildings, is a core goal of Riser.
The Maravedis take
The GSMA makes a coherent case. The per-operator math is hard to argue with if you accept 200 MHz as the 6G entry point, and the data on slow Wi-Fi generation upgrades deserves a direct answer from the Wi-Fi camp, not a dismissal.
Three questions remain open from where we sit in the multifamily market:
• Averages hide the MDU. Managed Wi-Fi operators deploy Wi-Fi 6E and Wi-Fi 7 at scale, often well ahead of the consumer market. The GSMA's usage figures come from Speedtest scans across whole cities,⁵ so they may understate how much wide 6 GHz channels already matter in professionally managed, dense buildings.
• Macro coverage still has to get inside. If 6 GHz behaves like 3.5 GHz, upper 6 GHz macro signals will face the same building penetration limits that already leave many apartments with poor indoor coverage. Without MDU-specific modeling, that gap is assumed, not measured.
• The UK clearing problem. Under Ofcom's model, Wi-Fi gets to use the band first. Once millions of Wi-Fi devices are in homes, moving them out of the mobile priority portion later may prove harder than any study suggests.
A bigger question sits behind all three: who pays for 6G at all? Zoghbi conceded that operators earn no more per megahertz or per user as generations change. Dell'Oro Group projects that operators will invest more than $500 billion in 6G RAN between 2029 and 2034, roughly half of all RAN spending in that window. Yet it expects cumulative 6G RAN revenue in the first six years to come in 10% to 20% below the comparable 5G period, with the overall RAN market growing about 1% a year.¹⁰
The 5G experience explains the caution. Operators spent over $600 billion on 5G infrastructure on the promise of premium enterprise use cases such as autonomous mobility, smart cities, and remote surgery. Those revenues never arrived at scale: 5G became faster mobile broadband with a lower cost per bit, and customers absorbed the extra capacity without paying a premium.¹¹
In our view, 6G risks repeating the pattern. New spectrum such as 7 GHz and more efficient radios protect margins rather than create growth, and traffic alone is a weak thesis when direct generative AI traffic is still only about 0.2% of mobile traffic.¹¹ The credible business models, if they exist, sit above the radio layer: guaranteed AI delivery, edge inference, sensing, and agent identity and security. None depends on 6G specifically, and none has been proven yet.
That shapes the 6 GHz fight. If 6G is a cost of staying in business rather than a growth engine, operators will push hard for cheap spectrum, and regulators will be weighing projected GDP gains against a business case operators have yet to define. For MDU owners, the lesson is simple: indoor coverage will not improve just because a new generation arrives. Someone still has to fund it.
Continue the conversation at Riser
Regulators will make the upper 6 GHz decision, but apartment buildings will live with its consequences. That is why we created Riser, the Maravedis event dedicated to MDU connectivity, taking place in Amsterdam on April 6-8, 2027.
Riser brings together the people who connect multifamily buildings: property owners and developers, managed service providers, fiber builders, Wi-Fi vendors, mobile operators, and policymakers. Fiber will be a big part of the agenda, as will the wireless last 10 meters, including how Wi-Fi and mobile can share the building rather than compete.
If you build, own, operate or connect multifamily properties in Europe, we would like to see you there. Register or request speaking and sponsorship details here.
Notes
1. GSMA, Vision 2040: Spectrum for the future of mobile connectivity, 2026 edition, p. 7.
2. GSMA, 6 GHz in Europe: Mobile evolution in 6.425-7.125 GHz, June 2025, p. 2 and Figure 1, p. 3.
3. GSMA, 6 GHz in Europe, Figure 2, p. 4.
4. GSMA, 6 GHz in Europe, p. 2 and Figure 6, p. 8. GSMA Intelligence analysis of Speedtest Intelligence data provided by Ookla, 11 European capitals, 2024 to 2025.
5. GSMA, 6 GHz in Europe, Figure 5, p. 7. Same Ookla data set; shares are of Wi-Fi 6 and 6E scans only.
6. GSMA, 6 GHz in Europe, Figure 7, p. 9. Same Ookla data set.
7. GSMA, 6 GHz in Europe, p. 2 and Figure 4, p. 6. GSMA Intelligence analysis of Speedtest Intelligence data provided by Ookla.
8. GSMA, Spectrum Pricing and Renewals in Europe, 2025, p. 8.
9. GSMA, 6 GHz in Europe, p. 2 and Figure 3, p. 5. GSMA Intelligence study of nine countries, 2024; benefits measured as net present value over 2023 to 2035.
10. Dell'Oro Group 6G RAN forecast, as cited in Sebastian Barros, The Unclear Path to 6G Business Models.
11. Sebastian Barros, The Unclear Path to 6G Business Models.