A Routine Renewal With High Stakes
On September 1, 2026, Switzerland's Federal Communications Commission (ComCom) published the invitation to tender for mobile spectrum in the 800 MHz, 900 MHz, 1800 MHz, 2100 MHz and 2600 MHz bands — frequencies currently held by Swisscom, Sunrise and Salt under licenses that expire at the end of 2028 (ComCom press release; BAKOM tender page). Companies wishing to bid have until December 3, 2026 to apply; the auction itself is set for the second quarter of 2027, with new 15-year licenses running from January 1, 2029. On paper this is administrative housekeeping — spectrum comes up for renewal, a new procedure replaces the old one. In practice it is the single most consequential lever Swiss regulators have over the shape of the mobile market for the next decade and a half.
What ComCom Is Actually Trying to Prevent
The stated objective is explicit: "maintain and further develop the three high-quality mobile communications networks and foster competition in the mobile communications market" (ComCom). The mechanism is a two-stage auction — a multi-round allocation phase followed by a sealed-bid stage that assigns specific frequency positions within bands — paired with bidding restrictions that stop any single operator from acquiring an outsized share of the available spectrum. The tender documents describe this directly as a safeguard against bidders acquiring "an excessive number of frequencies, which could undermine the current level of competition."
The concern is not hypothetical. Switzerland's mobile market is already asymmetric: Swisscom holds roughly 54% of postpaid subscribers, against Sunrise's 25% and Salt's 17% (ComCom market share data). A spectrum auction with no concentration limits would let the incumbent with the deepest balance sheet — historically Swisscom, which paid CHF 195.5 million of the CHF 380 million raised in the 2019 5G auction, more than double what Salt or Sunrise paid (Swiss Federal Council) — simply outbid its way to a durable capacity advantage. Over a 15-year license term, that advantage compounds: better coverage, lower per-subscriber network cost, and a widening moat against the two smaller carriers.
Steelmanning the Cap
There is a real case for intervention here, and it's worth taking seriously rather than waving away as reflexive dirigisme. Spectrum is not a normal input that a well-functioning market allocates efficiently through price alone — it's a fixed, non-substitutable public resource, and once it's locked into a 15-year license, competitors can't simply build around the shortage the way they might route around a scarce component. Switzerland's three-network model has delivered genuinely good outcomes: near-universal 5G coverage and consistently high marks in European mobile-network benchmarking. If an uncapped auction let one carrier buy enough headroom to make the other two structurally uncompetitive, the state would have spent a competition-promoting instrument to produce the opposite result — and unwinding a concentrated spectrum position after the fact, through merger review or forced divestiture, is far messier than preventing it at the auction stage. A regulator that has watched telecom markets consolidate to two effective competitors elsewhere in Europe is not wrong to want a design that forecloses that path here.
Where the Design Still Gets It Right
Having stated that case fairly, the Swiss approach deserves credit for solving the concentration problem without reaching for the more heavy-handed tools regulators elsewhere favor. ComCom is not mandating a fourth network entrant, imposing wholesale-access obligations, or setting a floor price disconnected from actual scarcity — all interventions that tend to blunt investment incentives for years afterward. A bidding cap that only binds when one operator tries to buy a disproportionate share of a fixed resource is a narrowly tailored fix for a narrowly defined risk. It preserves price discovery — the state still captures fair value for public spectrum, and the auction format lets relative demand across the five bands set that value rather than a regulator guessing at it. And because the caps apply symmetrically to all three incumbents (and any new entrant), the rule doesn't pick winners; it just stops the highest bidder from converting a wallet advantage into a durable structural one.
The Real Test Is in the Annexes
The risk is calibration, not the principle. ComCom has not yet published the exact numerical thresholds — those sit in Annex II of the tender documents, alongside the auction procedures — and that is where this policy will actually be judged. A cap set too loosely does nothing to constrain Swisscom's existing scale advantage; one set too tightly could leave usable spectrum unassigned or force an operator to under-provision for genuine capacity needs, degrading service quality for the sake of a competition metric. Switzerland's regulators have a solid track record of running clean, well-subscribed spectrum auctions — the 2019 sale cleared well above its CHF 220 million reserve. The Q2 2027 auction will show whether that same discipline can hold three healthy competitors in balance without turning spectrum caps into a de facto ceiling on network investment.