Operator Self-Reporting Regulators Can’t Independently Verify

An uncomfortable question

Here is a question worth putting to most telecom authorities, plainly: how do you know the numbers the operators send you are true?

Not “do you suspect they’re false”, that’s a more paranoid question, and a different one. Just this: what is your actual mechanism for checking? For most regulators, there isn’t one. The operator declares its international termination volumes, its interconnect traffic, its revenue, its coverage. The authority receives the declaration, files it, and acts on it. The declaration is the data. There is no second source to hold it against.

The evidence lives on the other side of the table

This persists not because of a lack of will or a lack of analysts, but because of access. The evidence that would confirm or contradict a declaration ( the call records, the signalling, the actual traffic on the network) sits inside the operators’ own systems. The party being measured holds the only copy of the measurement.

Think about what that means in practice. If an operator declares a volume of inbound international minutes, the thing that would verify it is the CDRs. The operator has them. You don’t. So your options narrow to three, and all three are weak: trust the number, sample a sliver of it through a periodic audit, or commission a study that arrives six months later and settles nothing, because the operator can contest the methodology.

Every one of those shares the same flaw. The finding, if there is one, comes long after the reporting period closed, and it’s arguable by the time it arrives.

“We’ll catch it in the audit” doesn’t close the gap

Periodic audits feel like the answer, and they’re better than nothing. But look at the timing. An audit happens annually, or less. It covers a sample, not the whole. And by the time it surfaces a discrepancy, the period in question is a year gone, the traffic is historical, the revenue is booked, the licence fees are assessed and paid.

So even a good audit is a rear-view mirror. It can tell you something was wrong last year. It cannot tell you something is wrong this quarter, while you could still act on it. And because it’s a sample, an operator can always argue the sample wasn’t representative. The discrepancy becomes a negotiation rather than a fact.

The gap here isn’t analytical capacity. Regulators have capable people who know exactly what they’d look for. The gap is access to independently collected evidence, a control point that doesn’t depend on the operator handing you the proof of their own compliance.

Where the declared number quietly shrinks

This matters most where money is attached to the declaration. Licence fees, spectrum royalties and sector taxes are often assessed on declared revenue or declared traffic. So anything that shrinks the declared base shrinks public receipts, invisibly, and without ever looking like wrongdoing.

Bypass is the clearest example. International traffic reclassified as local termination disappears from the international-revenue line. The minutes still happen, the subscriber still pays, but the base the fee is calculated on is smaller. Misclassified services do the same. So does undeclared partner revenue. None of it looks like fraud in the declaration, because the declaration simply doesn’t contain it.

An authority working only from declarations cannot see any of this, by definition. You cannot audit a number that was never reported to you.

The honest version of the problem

To be fair about it: operator-reported data is genuinely valuable. It’s rich, it’s detailed, and most operators report in good faith most of the time. The point is not that operators are lying. The point is that you have no way to know either way, and a regulatory regime that cannot independently verify is one that works only as long as everyone chooses to cooperate.

That is a fragile basis for enforcement. It means your authority rests on the goodwill of the parties you are meant to be holding to account. The moment one of them decides not to cooperate, you find out late, from a sample, in a form they can argue with.

Why this is the problem worth naming

Step back and the shape of it is clear. A regulator’s job is to hold operators to their obligations, on revenue, on traffic, on coverage, on the fees owed to the state. But you cannot hold anyone to an obligation you can only measure through their own account of it. Every other weakness people point to in telecom oversight (slow enforcement, contested findings, bypass that never gets caught, QoS complaints that go nowhere) traces back to this single root: the regulator does not hold independent evidence. It holds a declaration.

Until that changes, more frequent reporting just produces more declarations. Sterner letters ask for better cooperation from parties who already decide how much to cooperate. Bigger audit teams sample a little more of a picture the operator still controls. None of it addresses the actual gap, because the gap isn’t effort or intent or headcount. It’s that the evidence and the authority sit on opposite sides of the table.

The fix is not more trust, or more declarations, or more enforcement letters. It is an independent control point, evidence the authority collects or compels itself, continuously, and can set against what it was told. That is a genuinely different way of working, and it is within reach in a way it wasn’t a few years ago. We’ll make that case next.