Taiwan's Ministry of Digital Affairs (MODA) confirmed in mid-May that its Department of Resources Management has begun formal planning for 6G spectrum, with a full strategy due by the end of 2026. The ministry is evaluating three candidate ranges: the lower 600MHz band, prized for coverage and building penetration; a mid-band slice between 4.7GHz and 5GHz; and the upper 6GHz band, spanning 6425MHz to 7125MHz, which offers the wide channels next-generation networks will need for capacity. Officials have explicitly tied the timeline to the 2027 World Radiocommunication Conference (WRC-27), where the International Telecommunication Union's member states will formally decide which bands get harmonized for mobile use worldwide (Taipei Times).
The Physics Behind the Band Choices
The three-band approach is not arbitrary. Low-band spectrum like 600MHz travels farther and penetrates walls better, making it the backbone for rural and mountainous coverage — a real constraint in a country where terrain limits line-of-sight deployment outside the western corridor. Mid-band and upper-6GHz spectrum trade that reach for bandwidth, the ingredient that will actually differentiate 6G from 5G in throughput terms. MODA official Tseng Wen-fang described the underlying challenge as spectrum refarming — reclaiming and reallocating frequencies currently used for other purposes — comparing it to "an invisible urban renewal," where existing allocations function like properties that must be cleared and rebuilt to meet rising demand from AI and IoT applications (Taipei Times). That is a fair characterization: spectrum is a finite, already-occupied resource, and reallocating it always displaces some existing user, whether a broadcaster, a military system, or a fixed-satellite link.
Why WRC-27 Is the Real Deadline
The emphasis on international alignment deserves credit. Taiwan is not a member of the ITU, so it cannot vote at WRC-27 — but its telecoms and chipmakers still depend on whatever bands the conference harmonizes, because device and base-station supply chains are built around globally standardized spectrum, not national ones. A country that goes its own way on band selection risks paying a premium for bespoke equipment and complicating international roaming, a mistake several markets made with fragmented mid-band 5G rollouts in the early 2020s. By calibrating its own roadmap to land just before WRC-27's decisions crystallize, MODA is positioning Taiwan's telecom sector, and adjacent hardware makers like the ones behind ITRI's newly demonstrated 6G base-station chipset, to build for the bands the rest of the world will actually adopt rather than for a smaller, costlier domestic niche (RCR Wireless).
The Shadow of 2020
The planning also carries an implicit lesson from Taiwan's own 5G history. The National Communications Commission's 2020 5G auction raised roughly NT$138 billion (about US$4.6 billion) as five operators competed for 3.5GHz and 28GHz spectrum — a per-megahertz price in the 3.5GHz band that made it, by most industry accounts, the most expensive 5G bandwidth sold anywhere in the world (IEEE ComSoc). There is a legitimate case for aggressive auction design: spectrum is public property, competitive bidding reveals its true market value, and treasuries have every right to capture that value rather than gift it to incumbents. But the 2020 result also loaded five already-thin carriers with debt that had to be recovered somewhere, and it left the underused 28GHz band — where equipment still isn't mature — sitting on operators' books years later. Regulators elsewhere have since moved toward hybrid models that mix competitive bidding with coverage obligations and lighter fees for immature bands, precisely to avoid that outcome.
MODA's decision to fold 6G planning into the six-year, roughly NT$27 billion (about US$880 million) Next-Generation Communication Technology Development Program approved by the Executive Yuan for 2025-2030 suggests the government understands this (Executive Yuan). Public money is going toward simulation networks, testing platforms, and regulatory groundwork before a single frequency is auctioned — the opposite sequencing from 2020, when spectrum went to market before the ecosystem around it was ready.
What Proportionate Regulation Looks Like Here
The test will come around 2028, when spectrum cleanup and pre-auction preparation are slated to begin. If MODA and the NCC use the intervening two years to clear incumbent users, publish base-station and licensing rules early, and calibrate reserve prices to reflect genuine early-stage demand rather than 2020-style scarcity psychology, Taiwan could turn a structurally difficult reallocation into a template for orderly 6G rollout. If the auction reverts to revenue maximization once treasury officials see what carriers are willing to pay, the same debt overhang that slowed 5G capital expenditure will repeat itself, just with a bigger price tag.
- Coverage equity: low-band 600MHz allocation should carry rural build-out obligations, not just population thresholds, given Taiwan's terrain.
- Auction design: reserve prices should reflect immature-technology bands like upper 6GHz differently than proven mid-band spectrum, echoing carrier objections to 2020's 28GHz pricing.
- Regulatory sequencing: NCC base-station and licensing rules should be published well before any auction date, not concurrently with it.
The underlying instinct — plan early, align internationally, avoid ad hoc reallocation — is sound policy. Whether Taiwan actually improves on 2020 will depend on choices the NCC hasn't yet made public.