Roughly eight of every ten international flights to and from Iran have stopped operating. The figure circulates as a headline in policy briefings and wire copy. It is not a flight count. It is a measurement of how much of a national aviation network a Western observer can still see.
That distinction is not semantic. Between the tail number and the ticket there is a chain of custody โ lessor, insurer, maintenance provider, fuel supplier, navigation data vendor, correspondent bank โ and the sanctions did not sever that chain in the middle. They severed it at the two points where every aviation chain terminates: liability and parts. What remains is a corridor that still moves people, still moves cargo, and now moves under callsigns and registrations that do not resolve cleanly to any jurisdiction.
I have spent years reconstructing systems that were built so they could not be reconstructed. Sanctions enforcement and post-mortem insolvency work are the same discipline. Both ask you to assemble a timeline out of fragments that somebody had a motive to delete.
Beneath the yield lies the rot. Here the yield is a percentage. The rot is the assumption that the percentage measures what it claims.
The legal architecture is older than most of the aircraft still flying under it. Executive Order 12959 in 1995 closed the door on US trade with Iran, including civil aviation. The Comprehensive Iran Sanctions, Accountability, and Divestment Act of 2010 widened the perimeter to non-US entities that supplied refined petroleum or supported the sector indirectly. The Joint Comprehensive Plan of Action in 2015 opened a window, and the window produced the largest order book in Iranian aviation history: roughly one hundred Airbus aircraft, eighty Boeing aircraft, twenty ATR turboprops. What was actually delivered before the snapback in May 2018 was modest โ a handful of Airbus widebodies and narrowbodies, fewer than a dozen ATRs. Boeing delivered nothing. The order book existed on paper, and paper is not an asset.
The designations that followed did the real work. Mahan Air, designated in 2011 for supporting the Quds Force, was the first carrier to lose its international footprint. Iran Air followed in 2018. Then Caspian Airlines, Meraj Air, Pouya Air โ each designation peeling away a layer of counterparties rather than aircraft. Iran has sat on the Financial Action Task Force blacklist since 2007, with a formal call for countermeasures reaffirmed in 2020. That status does not ban Iranian aviation. It makes every bank that touches Iranian aviation nervous, and nervous banks are more effective than prohibitions.
Escalation through 2024 and 2025 tightened the screws without changing the shape. European carriers withdrew from Iranian routes. Several transit hubs declined to refuel Iranian tail numbers. Reinsurance capacity, already thin, thinned further. Iran's response was not to ground the fleet. It was to re-route it: heavier reliance on the Caspian corridor toward Russia, expanded domestic and regional charters, and a settlement layer that increasingly runs through non-bank rails rather than correspondent banking.
That settlement layer is where the crypto conversation enters, and where most of the commentary gets the causality backwards.
The chokepoint is not the airframe.
Airlines are not primarily constrained by whether they own metal. They are constrained by whether the metal is insurable, maintainable, and payable. Four links matter more than the aircraft itself.
The first is leasing. Most of the world's commercial fleet sits in special purpose vehicles domiciled in Ireland, Singapore, and a handful of other jurisdictions, and those lessors are regulated, dollar-funded entities. A designation converts a lease into a liability overnight. Lessors terminate. Repossession is attempted where possible and abandoned where not. Aircraft strands at foreign airports, accrues parking fees, and eventually re-enters service under a different registration.
Registration laundering is the oldest trick in the corridor and the least discussed. Tail numbers migrate to Armenian, Iraqi, Kazakh, Tanzanian, and Ugandan registries. An aircraft's identity is a piece of paper issued by a civil aviation authority that has no incentive to audit the chain of title behind it. The tail number does not lie; the registry entry can. Any analyst who stops at the registration code is analyzing a filing cabinet, not a flight.
The second link is insurance. This is the real chokepoint, and it is the one that almost never gets named in coverage of the story. Hull and liability cover for commercial aviation is concentrated in a small number of London-market syndicates and reinsurers. An airport will not accept a landing slot, a handler will not push back an aircraft, and a fuel supplier will not open a valve without evidence of cover. Iran's workaround has been domestic insurance pools, fronting arrangements through Gulf intermediaries, and self-insurance through state-owned entities. Each of these transfers risk back onto the operator. When liability sits on the operator, maintenance standards become an accounting decision rather than a regulatory one. The livery is the mask; the certificate of insurance is the bone.
The third link is spares. Airframes tolerate neglect better than engines do. A licensed MRO facility cannot touch an Iranian tail number without exposing its own certificates. The alternative supply chain โ brokers in the UAE, Turkey, and Armenia, working through layered trading companies โ moves parts that are frequently untraceable, sometimes unserviceable, and occasionally counterfeit. The most common failure mode is not a dramatic engine event. It is a component with no paperwork and no provenance.
The fourth link is the one nobody mentions, because it does not look like hardware. It is data. Airworthiness depends on a continuous stream of information: current navigation databases, performance software, engine trend monitoring, airworthiness directive compliance tracking, and revised instrument approach procedures. A perfectly sound aircraft with an obsolete navigation database flies degraded approaches, and degraded approaches into mountainous terrain are how aviation accidents actually happen. The vendors that publish this data are Western, licensed, and sanctions-aware. The flight does not end on a runway. Sometimes it ends because nobody updated a file.
Reconstructing the eighty percent.
The figure is derived from two sources: scheduled-flight databases maintained by Western aviation data firms, and automatic dependent surveillance coverage โ the broadcast signal that makes an aircraft visible on public tracking platforms. Both sources measure observation, not activity. Three distortions follow.
Coverage gaps are the first. Terrestrial receivers do not blanket Iranian airspace uniformly, and satellite-based surveillance is not universal across all operators and aircraft generations. Flights through the Caspian corridor and western Iran can transit with intermittent visibility.
The second distortion is deliberate silence. Operators engaged in sanctions-sensitive activity โ cargo linked to procurement networks, flights carrying personnel or equipment that does not want an identity โ disable broadcast surveillance or operate under call signs that do not match any filed schedule. Silence is the loudest indicator of risk. When a corridor's visible traffic collapses faster than its fuel consumption, the gap is not an improvement in compliance. It is a migration out of the observable layer.
The third distortion is the denominator. The baseline for the eighty-to-ninety percent claim is scheduled commercial service from the 2017-2018 window, immediately before snapback. Comparing the present to that baseline measures the collapse of the scheduled, insurable, bankable segment of Iranian aviation. It does not measure total aircraft movements, because charter, cargo, and state-linked flying never sat inside the scheduled database to begin with. The honest reading is that the figure is a reliable floor for the contraction of commercially normal aviation and a very poor estimate of the residual. My own audit experience says the same thing every time: when a system degrades, the measurement infrastructure degrades first, and the degradation gets reported as the outcome.
Where crypto actually sits.
Three legs of the corridor touch digital assets, and none of them is the headline flight.
The first is procurement. Parts brokers, freight forwarders, and front companies moving avionics components and consumables need to be paid, and they need to be paid by counterparties who cannot open a dollar account. The settlement stack has evolved into a blend: third-country invoicing in dollars where a cooperative bank still exists, hawala where it does not, and increasingly stablecoin transfers for the middle layer โ the broker paying the sub-broker, the freight agent paying the ground crew. This is not a dramatic substitution. It is a grubby, operational use case, and it works precisely because the amounts are small and the disputes are settled socially rather than legally.
The second is ticket settlement. Iranian nationals abroad purchasing domestic-nexus travel, and Iranian travel agencies settling balances with foreign consolidators, increasingly route through non-bank channels. Volume is modest. Persistence is high.
The third is fuel and handling. Some third-country service providers have been willing to service Iranian tail numbers in exchange for settlement outside the dollar system. Stablecoin liquidity has been the medium of choice because it clears in minutes and does not require a correspondent relationship.
Here is the structural limit, and it is absolute. Crypto substitutes for the settlement layer. It does not substitute for the liability layer. You can pay a parts broker in a token. You cannot pay a London reinsurer in a token. Reinsurance is a licensed, know-your-customer, dollar-denominated function governed by a treaty that does not accept bearer instruments. Hype is noise; structure is signal. The reason Iran's aviation network cannot be rebuilt with digital assets is not that the technology is inadequate. It is that the binding constraint sits on a balance sheet in a jurisdiction where the asset class is inadmissible.
There is a second structural limit that evasion commentary consistently ignores: the most liquid non-bank settlement rail is also the most revocable. Dollar-denominated stablecoins on high-throughput public chains dominate corridor settlement because they are liquid and familiar. They are also issued by centralized entities that retain contract-level freeze authority. Enforcement actions have demonstrated that authority repeatedly. The rail that makes evasion possible is, by design, the same rail on which a single administrative decision can seize the balance. In 2020 I spent three weeks dissecting a lending protocol whose price feed aggregation was elegant and trivially manipulable. The elegance of the mechanism told me nothing about who could bend it. Same pattern here, larger consequences.
Designations of specific addresses follow the same logic. The Treasury has listed bitcoin addresses tied to Iranian exchange and ransomware-adjacent activity since 2020, and exchanges operating in the country have been targeted since. Address listing is a naming exercise, not a capability. Once value rotates through mixers, bridges, and freshly generated wallets, the listed address becomes a compliance filter for regulated intermediaries rather than a constraint on the actor. The 2022 designation of a mixing protocol and its later reversal in 2025 illustrates the deeper point: the tactic moves the perimeter, it does not move the center. Regulated entities lose the ability to touch the flow. The flow continues.
What a bear market changes.
Evasion demand is not risk appetite. It does not fall when bitcoin falls. Corridors that exist because a counterparty cannot access dollars will keep operating through a drawdown, because their users are not speculating. What does change is the supply side. In a risk-off environment, market makers retreat, over-the-counter desks tighten onboarding, and liquidity thins for everyone. The compliant edge gets harder to reach, which raises the cost of the settlement layer for sanctioned actors too.
For readers tracking protocol health in this market, the Iranian case is a useful stress test rather than a curiosity. The question it answers is the one that matters in a drawdown: when the conventional rails stop answering, which counterparties can you still reach, and on what terms? Most projects have never asked that question about themselves. Their treasury is denominated in the same assets everyone else's is, held with the same custodians, and reachable only through the same choke points. The corridor is a reminder that access is a structure, not a right.
What the bulls got right.
The evasion advocates and the sanctions maximalists both got something correct, and both got the magnitude wrong.
On evasion: the direction was right. Iran did formalize a framework permitting crypto for import settlement, and corridor participants did adopt non-bank rails where banking failed. But the ceiling is low. No token pays a reinsurer, satisfies an airworthiness directive, or updates a navigation database. Digital assets captured the settlement tail of a liability-heavy system. Anyone modeling that as a full workaround has not read how insurance treaties are written.
On enforcement: the leverage point is correct. Insurance and parts are the right targets, and they are far more effective than blanket prohibitions. But the scorecard is wrong. Using an observability-derived percentage as a policy metric rewards opacity and punishes candor. If the analyst community is going to demand rigor from protocols, it should demand the same rigor from its own headline numbers. I do not follow the wave; I measure its depth.
And there is a blind spot neither camp names. The sanctions did not merely restrict Iranian aviation. They caused the construction of a parallel stack: a Caspian routing corridor, deepened aviation cooperation with Russia, a domestic insurance pool, a re-registration pipeline, and a settlement layer that does not require a New York correspondent. Alternative infrastructure has a habit of outliving the policy that summoned it. In 2017 I wrote a divestment memo on three projects whose proprietary cryptography was a rehash of insecure open-source libraries. The fund ignored it and lost ninety percent. The lesson generalizes: tools built for evasion do not get demolished when pressure lifts. They get repurposed by whoever finds them useful next.
Three numbers worth watching.
Ignore the headline percentage. Track the renewal rate on Iranian-nexus hull and liability cover, because that is what actually determines how much flies. Track the share of corridor settlement that runs through issuer-freezable contracts versus the portion that does not, because that ratio measures how much of the evasion floor is removable by a single administrative action. And track the age distribution of the operating fleet, because that number kills people rather than markets, and it moves slowly enough that nobody notices until it does.
The flights did not stop because a runway closed. They stopped because nobody would underwrite them, and the parts pipeline that kept them alive is a supply chain of paper as much as metal. When the next corridor opens somewhere else โ and it will โ the question to ask will not be how much value moved through it. It will be who was willing to sign the certificate, and what they were promised in return.