On 1 September 2026, Switzerland's Federal Communications Commission (ComCom) launched the tender for mobile frequencies in the 800, 900, 1800, 2100 and 2600 MHz bands, which become available from 1 January 2029. BAKOM's announcement says the award will be made by auction, expected in the second quarter of 2027. Its stated aims are to maintain and further develop the three mobile networks and to foster competition.
The strongest case for an active regulator
The argument for tight regulatory design is serious. Spectrum is a scarce public resource. Without safeguards, one operator could buy enough of the best low-band spectrum to foreclose its rivals. Switzerland has only three national networks (Salt, Sunrise and Swisscom), so losing one would leave a duopoly. Coverage obligations, bidding restrictions and new-entrant provisions are the tools regulators use to prevent that outcome. ComCom's release says bidding restrictions are meant to ensure that no bidder can acquire an excessive number of frequencies in a way that would undermine the current level of competition. That is a legitimate reason to intervene.
What the tender actually fixes
The mechanics are fairly clear from the published notice:
- Interested companies that want to provide mobile services over their own infrastructure may apply by 3 December 2026.
- Licences run for 15 years from 1 January 2029, so operators get long-term planning certainty.
- Licences are technology-neutral. Operators choose whether to run 4G, 5G or later technologies, provided they meet the terms of use.
- The auction uses the two-stage procedure from 2019, which Swiss regulators already know.
The tender documents also cover coverage obligations, new-entrant provisions and reserve prices. I could not confirm those specific terms from the announcement pages, so this analysis does not assess them. They matter, and readers should check the documents themselves.
Why the 2019 precedent is a reasonable guide
Reusing the 2019 format is a sound choice. That auction ran from 29 January to 7 February 2019 and, according to BAKOM, raised around CHF 380 million for the Confederation. All three incumbents won spectrum. Euronews reported that the bidding had been expected to raise at least CHF 220 million, so competitive bidding produced a much higher figure than the floor implied.
That result cuts two ways. It shows the auction mechanism works and that all three networks remained viable. It also shows that the money comes out of operator budgets. Every franc paid for spectrum is a franc not spent on antennas, fibre backhaul or rural coverage. A regulator that wants better networks should treat auction proceeds as a cost to the goal, not as revenue.
Where the design risks lie
The first risk is reserve prices. Reserves set well below expected value leave the market to find the price. Reserves set high can leave spectrum unsold or push operators to shift costs onto subscribers. The 2019 outcome, far above expectations, suggests bidders value this spectrum highly, so a modest reserve is enough to protect the public interest.
The second risk is coverage obligations. Mandates that improve service in areas the market would skip can be justified. But obligations written around a particular technology date quickly. The technology-neutral principle in this tender is the right instinct, and coverage rules should be defined by outcomes, such as speeds and geography, not by which generation of equipment operators must deploy.
The third risk is the new-entrant provisions. Switzerland's market is small, and the tender's stated aim is to maintain the three existing networks. A new entrant needs a real chance of building a network, and the entry rules should not become a subsidy for a fourth player that cannot sustain its own infrastructure. Designing them well means being honest about the economics of a country of roughly nine million people, a figure the notice does not address. Rules that give a credible but not guaranteed path, such as spectrum blocks available on a fair basis, are better than reserved lots that go unused.
A pro-innovation reading
The long licence term is the best feature of this tender. A 15-year horizon lets operators plan network investment without fearing an early reassignment, and it ends in 2043 with a clean renewal. Frequent regulatory interference during the term would erode that value. The regulator's job after the auction is to enforce the stated terms and stay out of the way.
The process is also transparent. Publishing tender documents, an application deadline and an auction window months ahead lets operators, vendors and investors plan. Other countries have delayed spectrum awards or changed the rules late, which raises risk premiums. Switzerland is sending a stable signal, and it should keep sending it through the auction in 2027.
What to watch
Between now and the December application deadline, the key questions are whether any non-incumbent applies, how the bidding restrictions are calibrated for the low bands, and what the final reserve prices look like relative to the CHF 380 million raised in 2019. If reserves are modest, obligations are outcome-based and the rules stay stable for 15 years, the tender will meet its stated aim of three healthy competing networks. If they drift towards revenue maximisation or technology-specific mandates, the cost will be paid later in slower rollouts and higher prices.