An organisation with no power that everybody obeys
Between 23 June and 1 July 2026, in Taipei, the Space Frequency Coordination Group held its forty-fifth annual meeting, hosted by Taiwan’s space agency. Around the table sat delegations from something like thirty-four space agencies — NASA and NOAA, ESA and EUMETSAT, CNES, DLR, JAXA, ISRO, the Canadian, Korean, Brazilian, Australian, Egyptian, Nigerian, South African and Emirati agencies, several Chinese institutions, the UK Space Agency, Sweden’s SSC, Roscosmos, and the National Space Agency of Ukraine. Observers included the radio astronomers of CRAF and IUCAF, the SKA Observatory, the World Meteorological Organization, the interagency operations and Earth-observation bodies, and a delegation from the relevant ITU-R study group.
The SFCG cannot allocate a single hertz. It issues no licences, adjudicates no disputes, and its Recommendations and Resolutions bind nobody. Yet its outputs shape the design of essentially every civil science and Earth-observation mission flown by its members — governing how they use the deep-space bands, how they share the X-band and Ka-band data downlinks, how they avoid one another and protect radio astronomy. It maintains a members’ database of missions and frequencies, and a public repository documenting radio-frequency interference into passive Earth-observation sensors, opened after members found their data products increasingly corrupted. Most consequentially, it publishes an agreed set of objectives for each World Radiocommunication Conference — updated most recently on 1 July 2026 for WRC-27 — which converts a scattered global science community into a single negotiating position.
Why does a voluntary club work? Three reasons that recur across every body in this article. The membership is small and repeatedly interacting, so reputation enforces what law cannot. The output is technical rather than political, which gives it legitimacy that survives disagreement between the members’ governments — note that both Roscosmos and Ukraine’s agency sit on the roster. And the members share an interest no individual state can defend alone: the passive and science bands are commercially worthless and scientifically irreplaceable, which makes them permanently vulnerable to whoever wants the spectrum next.
That vulnerability is live right now. Among the frequency ranges under study for terrestrial mobile identification at WRC-27 is 7125–8400 MHz — a range that runs straight through the 8025–8400 MHz band on which the world’s Earth-observation satellites downlink their data. The science community’s core downlink is, in other words, a candidate for the mobile industry’s next generation. Bodies like the SFCG exist precisely for the moment when the arithmetic of commercial spectrum demand meets a constituency with no revenue to defend itself.
The five families
Step back from the SFCG and the wider architecture resolves into five families of body, each exercising a different kind of power.
The regional blocs: where WRC outcomes are actually decided. Six regional groups prepare common positions for World Radiocommunication Conferences: CEPT in Europe (through its Electronic Communications Committee and Conference Preparatory Group), CITEL in the Americas (PCC.II), the Asia-Pacific Telecommunity’s preparatory group, the African Telecommunications Union, the Arab Spectrum Management Group, and the Regional Commonwealth in the field of Communications for the CIS states. Because a conference of 190-odd sovereign states cannot negotiate line by line, it negotiates bloc by bloc: proposals arrive as European Common Proposals, Inter-American Proposals, African Common Proposals, and the drafting brackets are largely resolved before anyone reaches the plenary. For an operator or an investor, this is the single most important structural fact about WRC influence — by the time an agenda item reaches the conference floor, its outcome is usually determined, and the place to have been was a regional group meeting eighteen months earlier.
The standards setters: where a band becomes a product. An allocation grants permission; a standard creates an ecosystem, and the second is what makes spectrum valuable. 3GPP, which defines the mobile standards the world’s handsets implement, has no regulatory authority whatever — yet a band without a 3GPP band number is commercially dead in the handset economy, and the entire direct-to-device industry became technically feasible only when 3GPP’s releases incorporated non-terrestrial networks into the standard. ETSI performs the parallel function in Europe, where harmonised standards underpin equipment compliance, including the flat-panel terminals that created the earth-stations-in-motion category. In the science and space-operations world, the Consultative Committee for Space Data Systems does something similar for space-link protocols, and it sits as an SFCG observer for exactly that reason. An operator who wins an allocation and ignores standardisation has purchased a permission nobody’s equipment can use.
The industry armies: advocacy at scale. Two associations dominate the field, and they are structurally opposed. The Global Satellite Operators Association — Brussels-based, chief-executive-driven — coordinates the satellite industry’s positions and publishes them agenda item by agenda item as the WRC-27 cycle develops, giving its members a unified voice that individual operators could not achieve. Facing it is the GSMA, representing the mobile industry’s several hundred operators and their broader ecosystem, which pursues terrestrial identification of the very mid-bands the satellite sector occupies, and publishes its own roadmaps for the bands it wants next. On direct-to-device the two are simultaneously adversaries and reluctant partners, since satellite operators need the mobile bands and mobile operators increasingly need satellite coverage. National bodies — the Satellite Industry Association in the United States, and the mobile suppliers’ association with its associate memberships across the regional groups — extend the same logic domestically. For a small operator, an industry association is the most cost-effective influence available: it buys a seat in rooms your company alone would never enter.
The science guardians: defending the bands with no business case. Radio astronomy and passive Earth sensing occupy spectrum whose value cannot be monetised and whose users cannot bid. Their defence is institutional: IUCAF, the scientific committee for frequency allocations sponsored by the international unions of radio science, astronomy and space research; CRAF in Europe; the SKA Observatory as an institution with its own considerable interest; the World Meteorological Organization, whose members’ forecasts depend on protected passive bands; and the Earth-observation coordinating bodies. They work largely by supplying technical evidence into ITU-R study groups and by lending their credibility to administrations willing to carry their case. A newer model has emerged alongside: direct, private coordination agreements between constellation operators and astronomical institutions, negotiated bilaterally and entirely outside the regulatory system — governance by contract, arriving because the formal machinery moved too slowly for the pace of deployment.
The self-help layer: solving problems the regime does not. When interference actually occurs, the Radio Regulations offer procedure but no operations centre. The industry filled the gap itself: interference-reduction groups that drove the adoption of carrier identification so that the source of an errant transmission can be found; the Space Data Association, an operator-run body pooling data on satellite operations and interference; and, for the smallest end of the market, the International Amateur Radio Union’s satellite coordination panel, whose voluntary frequency coordination for amateur-band spacecraft has become a de facto gatekeeping step that many administrations expect to see before licensing a university CubeSat. None of these bodies has legal authority. All of them are consulted before anyone dials a regulator.
| Body / family | Who is in it | What it produces | Binding? | Why you would care |
|---|---|---|---|---|
| SFCG | ~34 civil space agencies; science observers | Recommendations, resolutions, WRC objectives, interference repositories | No | Governs civil science/EO frequency practice; unifies the science bloc at WRCs |
| Regional groups (CEPT, CITEL, APT, ATU, ASMG, RCC) | National administrations | Common proposals and regional positions for WRCs | Politically, largely decisive | Where WRC outcomes are actually settled — months before the conference |
| Standards bodies (3GPP, ETSI, CCSDS) | Manufacturers, operators, research bodies | Technical standards, band profiles, protocols | No — but commercially decisive | A band without a standard has no equipment ecosystem and no market |
| Industry associations (GSOA, GSMA, SIA) | Operators and vendors | Advocacy positions, studies, coordinated lobbying | No | The affordable route to influence for anyone smaller than a giant |
| Science guardians (IUCAF, CRAF, SKAO, WMO) | Scientific institutions and unions | Technical evidence, protection cases, bilateral deals with operators | No | They defend the passive bands — and increasingly negotiate directly with constellations |
| Self-help bodies (interference groups, SDA, IARU) | Operators; amateur community | Interference resolution, carrier ID, voluntary coordination | No — but expected in practice | Where real operational problems get solved, and where CubeSats get their frequencies |
| Legacy intergovernmental organisations (ITSO, Eutelsat IGO, Intersputnik) | Member states | Oversight of filings, orbital assets and public-service obligations | Treaty-based, narrow | They hold or supervise real spectrum and orbital assets — and their memberships are now shifting geopolitically |
The taxonomy of power
Set out this way, the architecture reveals a division of labour that the ITU-plus-regulator mental model completely misses.
Only two categories of actor make rules: the ITU through the Radio Regulations, and national administrations through licences. Everything else in the list makes something more subtle and, frequently, more determinative. The regional groups make positions — and positions become rules by consensus at the conference. The standards bodies make ecosystems — and an ecosystem decides whether a rule is worth anything commercially. The technical clubs like the SFCG make norms — practices that are followed because deviation is embarrassing and technically indefensible among peers. The associations make arguments at scale. The self-help bodies make solutions, filling the space between what the treaty requires and what operations demand.
The practical implication is uncomfortable for anyone who believes influence flows through formal channels. By the time a matter arrives where the authority formally sits, it has usually been decided somewhere with no authority at all — in a regional group’s drafting session, a study group’s interference assumption, a standards release, or a club’s agreed objectives document.
Which door to knock on
For an operator or an administration deciding where to spend scarce attention, the mapping is reasonably clean.
If you want to change a rule, your only formal route is your administration — but the effective route is the regional group, entered through that administration, while the text is still bracketed. If you want your band to have a market, the relevant room is a standards body, and the timeline is measured in release cycles rather than conference cycles. If you are too small to be heard, an industry association is the cheapest voice you will ever buy, and its working groups are also where you learn what everyone else is planning. If you operate science or Earth-observation missions, membership or observership in the SFCG family is not optional — it is where your peers agree the practices your programme will be judged against, and where the defence of your downlink bands is organised. If you have an interference problem today, the self-help bodies will help you faster than any regulator. And if you are acquiring or financing a satellite business, note that all of the above generate public documents — regional group positions, association filings, standards releases, club recommendations — which together constitute the best available forecast of the regulatory environment your asset will live in.
The deeper observation is one this series keeps returning to. The spectrum regime looks, from outside, like a hierarchy with a treaty at the top. It behaves, from inside, like a network of clubs — technical, voluntary, reputational — around a small formal core. Clubs are open to those who show up, contribute competently and stay for years. That is a lower barrier than a treaty, and a higher one than money. It is also the reason the regime’s real experts are not the people who have read the Radio Regulations most carefully, but the people who have been in the rooms where the next edition is being drafted.