How to establish what a company’s spectrum position actually consists of, what it is exposed to, and what it is worth — using sources that are, unusually, largely public.
8.1 The asset that is not in the data room
Consider the typical experience of an investment team examining a satellite business. The data room is thorough by the standards of any other sector: audited accounts, customer contracts, a capitalisation table, employment agreements, insurance policies, a technical folder heavy with system specifications, a launch services agreement, an intellectual property schedule. Somewhere in the corporate section, under regulatory matters, sits a short document. It lists two or three satellite network names, the administration through which they were filed, and a sentence confirming that the company “holds ITU filings” for its planned system. Perhaps a licence certificate is attached.
That document may describe the majority of the enterprise value. It is also, almost always, the least examined page in the room.
The asymmetry has an obvious cause. Investment professionals possess robust instincts for the diligence of things they have priced before — revenue quality, customer concentration, technical risk, intellectual property. Spectrum rights resemble none of these. They are not owned, cannot be inspected, do not appear on the balance sheet at anything resembling their value, and are described in a technical vocabulary that neither the financial nor the engineering side of a deal team commands with confidence. The default response is to treat them as a legal formality — an item for counsel to confirm exists — rather than as an asset requiring valuation and stress-testing in its own right.
This article argues that the default is a serious and expensive error, and sets out the alternative: a structured diligence protocol for spectrum positions, the red flags that recur across transactions, and the translation of findings into price, structure and conditions. It assumes the mechanics — the queue, the clocks, the national layer — that feeds directly into the valuation framework and the deal structures.
One feature of spectrum diligence deserves emphasis at the outset, because it is genuinely unusual and confers real advantage. Most of the primary record is public. The ITU publishes the filing queue fortnightly. National regulators publish licences, applications, dockets and decisions. An analyst with the relevant skills can reconstruct the substantial majority of a target’s international position before asking the target a single question — and can therefore enter management meetings already knowing what the company holds, which turns the conversation from disclosure into verification. In an asset class where information asymmetry usually favours the seller, this reverses the usual polarity of diligence. Few other categories of asset can be examined so extensively from the outside.
8.2 What is being diligenced: the stack
We introduced the idea that a spectrum position is never a single right but a stack, and the diligence protocol follows that structure directly. Three layers must each be established, and — critically — each must be established in relation to the others, since the failure modes tend to live at the joins.
The international layer comprises the filings and recorded assignments held at the ITU through one or more notifying administrations: their priority dates, procedural stages, coordination status and the clocks running against them. This layer determines the company’s rank relative to every other system in the world, and therefore the fundamental quality of what it holds.
The operating layer comprises the national authorisations to operate the system: the space-station or constellation licence from the licensing state, together with any earth-station licences for the ground segment. This layer determines whether the company may lawfully switch anything on, and — under regimes such as the American one, with its bonds and deployment milestones — imposes performance obligations frequently more demanding than the ITU’s own.
The market layer comprises the authorisations to serve customers in each territory: landing rights, service licences, terminal approvals, and the local arrangements that make revenue lawful. This layer determines where the business can actually earn.
The classic diligence failure is to establish one layer and infer the others. A target with impeccable international paper may have market access nowhere; a target with an operating licence and customers may sit on an international position so junior that coordination will constrain it for a decade; a target with everything may hold it through a single administration whose relationship with the company rests on a founder’s personal connection. The protocol below is designed to prevent any layer from being taken on trust.
8.3 The protocol
Spectrum diligence proceeds in five phases. The sequence matters: the reconstruction phase should precede the document request, and the document request should precede management interrogation, because each phase makes the next one sharper and harder to deflect.
Phase 0 — Triage: is spectrum material here?
Not every space transaction requires deep spectrum diligence, and proportionality is a professional virtue. The question is whether the target’s economics depend on rights it holds rather than on rights someone else holds on its behalf. A component manufacturer, a launch provider, an analytics business processing third-party imagery: spectrum is generally immaterial. A satellite operator, a constellation venture, a teleport or gateway business, a direct-to-device play, a company whose pitch includes the words “we have secured spectrum”: material, invariably.
A useful triage test is the counterfactual: if the company’s regulatory position were withdrawn tomorrow, could it buy an equivalent one? Where the answer is “yes, from several suppliers”, spectrum is a cost. Where it is “no, or only after years of queueing”, spectrum is the asset.
Phase 1 — Independent reconstruction
Before the target is asked for anything, build the position from public sources. This phase typically takes an experienced analyst a few days and delivers three things: a draft inventory, a list of discrepancies to test, and the standing that comes from knowing the answers before asking the questions.
Begin at the international layer. The ITU’s space network databases and the fortnightly circulars allow the identification of networks filed by any administration on behalf of any operator, with their filing types, dates and procedural stages. Search by network name, by administration, and — because operators frequently file under names bearing no resemblance to their trading identity, sometimes deliberately — by orbital parameters and frequency bands corresponding to the system described in the target’s own marketing. Where a company’s public materials describe a constellation of a certain size in certain bands, the corresponding filing should be findable; when it is not, that discrepancy is the single most valuable finding the reconstruction can produce.
Then the national layers. Licensing databases and proceeding dockets in the principal jurisdictions are public and searchable: applications, grants, conditions, extensions, waivers, oppositions filed by competitors, and — often most revealing — the correspondence in which a regulator asked the company something awkward. Competitors’ filings in the same dockets are an underused intelligence source, since a rival’s objection to your target’s application will articulate, in technical detail and at no cost to you, the strongest available case against the target’s position.
Finally, assemble the timeline. Every filing and licence discovered carries dates; plot them on one chart against the company’s own programme milestones and funding runway. Much of what matters in spectrum diligence is visible the moment those two timelines are laid on top of each other, and valuation work depends on precisely this chart.
Phase 2 — The document request
With the draft inventory built, the request list can be specific rather than generic — which both improves the response and signals to the target that the diligence will be competent. The core request comprises: the complete schedule of filings with the notifying administration for each; all correspondence with notifying administrations regarding those filings; every national licence and authorisation with its conditions; all coordination agreements, whether concluded, draft or abandoned; correspondence with coordination counterparties on unresolved cases; bringing-into-use and milestone evidence submitted or planned; market-access authorisations by country with their status; any bonds, guarantees or financial assurances lodged with regulators; the company’s own regulatory risk register, if one exists; and the identity and contract status of the individuals who manage all of the above.
Two items on that list are diagnostic in themselves. Coordination agreements are the true content of the international position, and a target that declines to produce them — citing counterparty confidentiality, as targets frequently and sometimes legitimately do — has placed a limit on what any valuation can responsibly conclude, which should itself be priced. The regulatory risk register tells you whether the company thinks about its own position systematically or improvises; its absence in a company whose principal asset is regulatory is a governance finding, not merely an administrative one.
Phase 3 — Management interrogation
The meeting should test the joins, not the facts, since the facts are already reconstructed. Ten questions cover most of what matters.
Box 8.1 — Ten questions for management
- Walk us through your filings by priority date. Which of your bands are you senior in, and to whom are you junior?
- Which coordination agreements are concluded, which are in progress, and which counterparties have refused to engage?
- What operating constraints have you accepted in those agreements — power, geometry, service area, time?
- What are your bringing-into-use dates, and what is the funded plan to meet each of them?
- If you are flying a constellation, where do you stand against your deployment milestones, and what happens to the filing if you miss the next one?
- Which counterparties in your coordination queue do you regard as networks that will never be built, and what is that assessment based on?
- Does the system you are building match the envelope you filed? Where has the design moved, and what refiling or modification is required?
- Which markets are you authorised to serve today, which applications are pending, and what is the longest lead time among your target markets?
- Describe your relationship with your notifying administration. Who holds it, what does it cost, and what would happen if that relationship changed?
- Which proceedings, national or international, could change the rules your business depends on in the next three years, and what are you doing about them?
The quality of the answers is as informative as their content. A management team that answers question one fluently, with dates, is a team that understands its own asset. A team that must consult external counsel to answer question four has outsourced custody of the thing that determines its value — a real risk, and one that becomes acute at precisely the moment the company changes hands.
Phase 4 — Technical verification
Two verifications repay the cost of an engineering opinion, and both concern the join between paper and hardware.
The first is envelope conformity: does the system under construction actually fit inside the filed technical parameters? Filings are drafted early, often from preliminary designs, and systems evolve — higher power, different antenna patterns, altered orbits, additional satellites. Where the built system exceeds the filed envelope, the company faces modification, refiling, or operation outside its authorised parameters, each with consequences ranging from expense to invalidity. This mismatch is common, rarely disclosed, and occasionally fatal.
The second is coordination realism: an independent view of whether the coordination the company describes as “in progress” can plausibly conclude on the terms and timetable assumed in its business plan. In congested bands, with unresponsive counterparties, the honest answer is frequently that it cannot — and that the company’s operating case implicitly assumes concessions it has no mechanism to compel.
Phase 5 — Synthesis
The output is not a list of findings but a set of adjustments: to valuation , to structure, to conditions precedent, and to the risk register the buyer will inherit. Section 8.7 addresses the translation.
8.4 The inventory
The central artefact of the exercise is a complete inventory of the position, at a level of granularity most targets have never assembled for themselves. Producing it is frequently the most valuable thing a diligence exercise delivers — buyers have been known to hand it to management after closing, because it is the first comprehensive picture the company has ever had of its own asset.
| Field | What it establishes |
|---|---|
| Network name and filing type | The identity of the asset in the international register |
| Notifying administration | The sovereign counterparty on which the right depends |
| Priority date | Rank in the queue — the fundamental quality determinant |
| Frequency bands and orbital configuration | The pairing that defines what the right is for |
| Procedural stage | Where in the lifecycle the filing sits, and what remains to be done |
| Bringing-into-use status and deadline | The nearest existential clock |
| Milestone status (constellations) | Deployment obligations and the trimming exposure |
| Coordination status by counterparty | The encumbrance map — the true operational content |
| Operating licence and conditions | Authority to switch on, and the obligations attached |
| Bonds or financial assurances | Capital at risk on performance |
| Market access by country | Where revenue is lawful |
| Responsible individual | Key-person exposure on the asset itself |
Two disciplines make the inventory useful rather than decorative. Every line should carry a source reference — the circular, the docket number, the agreement — so that the buyer can audit and, later, maintain it. And every line should carry a confidence marking, distinguishing what is established from the public record, what rests on management representation, and what remains unverified. The distinction matters enormously when the memo is read months later by someone deciding whether to rely on it.
8.5 Reading the primary sources
Three source families carry most of the weight, and each has characteristic strengths and blind spots.
The ITU record — the fortnightly circulars and the space network databases — is authoritative on identity, dates and procedural stage, and it is comprehensive in a way no commercial database matches. It is silent, however, on precisely the things that determine operational value: what coordination agreements say, whether a filing is backed by a funded programme, and what the operator intends. The circular tells you that a network exists, when it was filed, and where it sits in the process. It does not tell you whether anyone will ever build it. That judgement — the load-bearing-versus-paper distinction that recurs — must be constructed by combining the record with programme evidence, financing evidence and the analyst’s accumulated pattern recognition.
National dockets are richer in narrative and poorer in coverage. Where a jurisdiction runs adversarial proceedings, the docket contains the regulator’s questions, the applicant’s answers, competitors’ oppositions and the reasoning behind conditions imposed — a level of insight into the strengths and weaknesses of a position that is available nowhere else. The blind spot is jurisdictional: many regulators publish little, and market-access status in a dozen countries frequently cannot be established from public sources at all, making it one of the areas where management representation must be relied upon and warranted accordingly.
Commercial databases and consultancies aggregate and sometimes enrich the above. They save time and can be excellent, but they are secondary sources: for any finding that will move price, the underlying primary document should be opened. Diligence memoranda that cite a commercial database for a priority date have not, in a meaningful sense, verified the priority date.
8.6 The red flags
Certain patterns recur across transactions with enough regularity to be worth naming. None is automatically disqualifying; each demands explanation, and the quality of the explanation is usually decisive.
Uncoordinated optimism. The company presents filings as though they were rights, when coordination has not begun, has stalled, or has concluded only with counterparties who did not matter. This is the most common flag by a wide margin, and it reflects a genuine conceptual confusion rather than deliberate misrepresentation: founders frequently believe that a filing is the spectrum. The diagnostic question is simple — how many of the networks identified in your coordination request have you actually reached agreement with, and how many of those matter?
Deadline exposure without a funded programme. A bringing-into-use date inside the investment horizon, with no financed satellite scheduled to meet it, is an asset with an expiry date the business plan does not acknowledge. The same applies with greater force to constellation milestones, where partial deployment means the filing is trimmed to what actually flew. Plot the clocks against the funding runway; if the runway ends first, the position is worth what a rescue financing will pay for it, not what the model says.
Convenience-flag dependency. The filing is held through an administration with which the company has no substantive connection, selected for speed or laxity. The exposure is threefold: the administration may lack the capacity to supervise or defend the network in coordination; the relationship may be personal to a founder or intermediary rather than institutional; and — as we argue — the market is beginning to discount such positions as scrutiny of the practice intensifies. Ask what the arrangement costs, who owns the relationship, what it is documented in, and what happens on a change of control. That last question surprises a remarkable number of targets.
Encumbrance opacity. Coordination agreements are withheld, or exist only as informal understandings between engineers. The value of a position is determined by what its holder has conceded, and a position whose concessions cannot be examined cannot be valued with confidence — only bounded. Where the target genuinely cannot produce agreements for confidentiality reasons, the response is a specific warranty and indemnity, not an assumption of benignity.
The market-access gap. Global coverage in the marketing, authorisations in three countries. This is less a deception than a sequencing problem — market access is slow, expensive and easy to defer — but it directly determines when revenue can begin, and business plans routinely assume authorisation timelines that no regulator has ever delivered. Test the longest lead time in the plan against the actual practice of the relevant regulator.
Filed-versus-built divergence. The engineering has moved on from the filing. Common, correctable, but expensive and slow to correct, and occasionally requiring a new filing with a new priority date — which is to say, requiring the company to surrender the very rank that made it valuable.
Concentration. The value rests on a single administration, a single coordination agreement, a single anchor market, or a single individual who understands the position. Concentration is a familiar diligence concept, and its spectrum-specific forms are simply less visible than the customer concentration a deal team would spot immediately.
Regulatory-weather blindness. Management cannot name the proceedings that could alter the rules governing their bands. In a period when national regulators are actively diverging from the treaty framework and a World Radiocommunication Conference is redistributing allocations, this is no longer an exotic risk. A company whose principal asset is regulatory, and whose management does not follow regulatory developments, is holding an asset it is not managing.
8.7 From findings to price and structure
Diligence that does not change the deal has not earned its fee. Four channels carry findings into the transaction.
Valuation adjustment. The valuation framework is calibrated to precisely the variables this article establishes: rank, encumbrance, coverage, clocks, counterparty durability. Findings feed in as adjustments to those inputs rather than as a separate discount — a junior position in a congested band with unresolved coordination is not a good position with a haircut; it is a different asset.
Conditions precedent. Where a finding is remediable before closing, it belongs in the conditions: coordination agreements produced or concluded, a modification filed, a market-access application lodged, an administration’s consent to change of control obtained. That last is not optional in most jurisdictions — spectrum rights do not transfer silently, and regulatory consent is itself a deal risk requiring its own timetable.
Warranties and indemnities. Spectrum-specific warranties are worth drafting properly rather than adapting from an intellectual-property precedent. The useful set covers: the completeness of the disclosed filing schedule; the accuracy of stated priority dates and procedural stages; disclosure of all coordination agreements and of any constraints accepted; compliance with all licence conditions and the absence of enforcement correspondence; the accuracy of bringing-into-use and milestone representations; and the absence of undisclosed arrangements with the notifying administration. Indemnities should reach the identified specific exposures rather than sitting at a general level.
Structure. Where the principal risk is a clock — a milestone to be met, a coordination to be concluded, a market to be opened — the natural response is to align consideration with the event: earn-outs on milestone achievement, escrow released on coordination conclusion, price adjustment on market-access grant. This is standard practice in other asset classes and unusually well suited to spectrum, because the events are objectively verifiable in the public record.
8.8 Four special cases
When spectrum is the whole asset. In transactions of the EchoStar type — where the operating business is incidental and the licences are the purchase — diligence and valuation collapse into a single exercise, and the analysis must extend to the regulatory-political dimension: the granting authority’s posture towards the holder, any live proceedings questioning use, and the transferability of each licence to this particular buyer. In such deals the regulator is effectively a counterparty, and its disposition is a diligence subject in its own right.
The venture round. At seed and Series A the position is usually a filing and an intention. Diligence proportionate to the stage asks a narrower question: has the company made the irreversible decisions well? Band selection, orbital configuration, choice of administration and timing of filing are difficult to undo and set the ceiling on everything later. A venture investor who verifies nothing else should verify that the priority date exists, that it is early relative to comparable systems, and that the band chosen is one in which coordination is achievable by a company of this size.
The lapse play. A distinct and growing category: acquiring positions cheaply from holders approaching deadlines they cannot meet. Here the diligence inverts — the buyer’s interest is in exactly how much time remains, what would be required to bring into use, and whether the administration would support a transfer and a revised programme. The analysis of the approaching pruning of the speculative filing class suggests this category will be busy in the second half of this decade.
Diligence for the administration. A state assessing an applicant seeking sponsorship runs a mirror-image of this protocol, asking not “what does this company hold?” but “what would we be responsible for?” — the applicant’s financing, technical competence, deployment credibility, and its record with other administrations, including filings it has allowed to lapse elsewhere. We develop this as an instrument of national portfolio management.
8.9 The memorandum
The deliverable should be short, decision-oriented and honest about its own limits. A structure that has served well: a one-page conclusion stating what the target holds, what it is worth in ranges, and the three findings that most affect price; the inventory table; the clock chart plotting regulatory deadlines against the company’s programme and funding; a findings section organised by red flag with each finding’s evidence and confidence level; the recommended adjustments to price, conditions, warranties and structure; and an appendix of source documents.
Two conventions raise the quality of such memoranda markedly. State confidence explicitly for every material finding — established, represented, or unverified — because readers otherwise assume uniform reliability. And state what could not be determined, and why. A memorandum that reports “coordination status with the three networks senior to the target in the 12 GHz band could not be established because agreements were withheld” is far more useful than one that quietly omits the topic, and it is the sentence that protects both the analyst and the investment committee.
A complete worked example, on a composite target, appears as Appendix A.
8.10 The limits of diligence
Three things this protocol cannot deliver, and honesty about them is part of professional competence.
It cannot reveal the content of confidential coordination beyond what the target discloses. Where agreements are withheld, the position can be bounded but not determined, and the memorandum should say so rather than substituting an assumption.
It cannot forecast administration politics. A notifying administration’s future disposition — towards the company, towards the sector, towards a buyer of a different nationality — is a sovereign matter, and the events that change it are rarely visible in advance. The treatment of losing rights, and of spectrum under geopolitical stress, describe what this exposure looks like when it materialises.
And it cannot price rule change. The framework itself is in motion: power limits being rewritten jurisdiction by jurisdiction, allocations under negotiation at WRC-27, national regimes diverging from the treaty baseline. Diligence can establish exposure to these processes — which bands, which proceedings, which outcomes would hurt — but the outcomes themselves belong to the analysis of Part III rather than to the certainties of a data room.
What the protocol does deliver is the thing the opening section of this article found missing: a target’s spectrum position converted from a paragraph of assertion into an inventory of established facts, dated clocks, mapped encumbrances and named exposures. In an asset class where the public record is unusually rich and the professional habit of reading it unusually rare, that conversion is available to any buyer willing to do the work — and, on the evidence of the transactions of recent years, it is worth considerably more than it costs.