Transshipment is ordinary logistics. Most container traffic in the world passes through a hub port that is neither origin nor destination, because moving cargo through hubs is cheaper than moving it directly. The practice is universal and unremarkable.
It becomes a control problem when the intermediate stop is used to break the chain: goods arrive under one description, destination or consignee, and leave under another, so that the exporter’s records show a lawful sale to an unremarkable customer and the goods end up somewhere the exporter could not have shipped them.
How it works
The exporter sells to a buyer in a country with no relevant restrictions. That sale is genuine: the buyer pays, the goods ship, the paperwork is accurate, and as far as the exporter’s compliance file is concerned the transaction is closed.
What happens next is outside the exporter’s view. The goods are re-exported, or moved through a free zone, or sold on to a second buyer, and reach the restricted destination on a second set of documents naming different parties.
The re-export leg is where the legal question sits. Items subject to the US Export Administration Regulations remain subject to them after they leave the United States, so the second leg may itself require a licence that nobody applied for. The equivalent question arises under other regimes through different mechanisms.
Why free zones concentrate the risk
Goods inside a free trade zone are frequently treated, for customs purposes, as not having entered the country. The ordinary import declaration that would fix their description, value and origin is not made, and records of what entered and what left may be held by the zone operator rather than by customs.
FATF’s trade-based money laundering work treats free trade zones as a recurring subject, noting the prevalence of trade-based laundering as it relates to the exploitation of free trade zones1 and dealing with them in a dedicated 2010 report rather than in the 2020 study itself. Its indicator set flags cash deposits forwarded to persons or entities in free trade zones or offshore jurisdictions with no business relationship to the account holder2 , and commodities imported under a temporary importation or inward processing regime that are subsequently exported with falsified documents3 . The documentary chain has a gap in it, by design, and the two ends are not necessarily reconciled by anyone.
The intermediary’s economics
The most robust indicator in this whole area is volumetric. An intermediary in a third country that begins importing quantities of a controlled item far beyond any plausible domestic demand is not consuming them, and there are only so many places they can be going.
This is why so much analytical attention now goes to public trade statistics rather than to individual shipments: a country’s imports under a monitored commodity code, compared against its own industrial base, is a question that can be asked of open data.
How it is caught
Trade statistics. Sudden growth in a third country’s imports of a controlled code, uncorrelated with anything in its economy and correlated with the closure of a direct route, is the aggregate signature. Where that growth is matched by growth in the same country’s exports to the restricted destination, the inference is close to unavoidable.
Post-shipment verification. Export control authorities can and do check that goods reached the end user they were licensed for. An end-use check that finds the goods gone, or the end user unable to produce them, converts an administrative question into an enforcement one.
Route implausibility. FATF and Egmont flag shipments of commodities routed through a number of jurisdictions without economic or commercial justification4 , and the Export Administration Regulations flag a shipping route abnormal for the product and destination5 . A commercially illogical route has to be explained by something, and cost is usually not it.
Consignee reconciliation. The party the exporter dealt with and the party that took delivery are both documented, in different places. Matching them is what most enforcement actions in this area come down to.
Why the intermediary is the constraint
The technique’s weak point is not the goods, the documents or the route. It is that somebody in the third country has to take delivery, hold the goods, and send them on, and that party has to exist, transact and appear in records.
That intermediary faces a genuine commercial problem: it must be plausible enough that a Western exporter will sell to it, which means having a real business, a bank account and a trading history, and it must be willing to do something that will eventually be visible. Those two requirements pull in opposite directions, and the tension is what makes intermediaries findable.
It also explains why the same names recur. Building a usable intermediary is expensive, so networks reuse them, and reuse is a pattern.
The volumetric argument, stated carefully
The strongest signal in this area is the mismatch between what a country imports and what it can plausibly consume, and it is worth stating precisely because it is frequently overstated.
The argument is not “imports rose, therefore diversion”. It is a conjunction: imports of a specific monitored commodity code rose sharply; the rise began when a direct route closed; the importing country has no domestic industry that consumes that code at that scale; and the country’s own reported exports of the same code to the restricted destination rose in step.
Each element alone is weak. Together they exclude most innocent explanations, and the analysis can be run entirely on published data by anyone. This site’s own version of that analysis, with the exact query and its limitations, is published as [a data note](/blog/machinery- exports-after-2022/) and a methodology note.
What it still cannot do is identify a shipment or a company. It identifies a flow.
Free zones: the specific mechanism
The reason free zones recur in this material is worth stating exactly, because “free zones are risky” is not an explanation.
Goods in a zone are frequently treated for customs purposes as not having entered the country. The consequence is that the import declaration which would ordinarily fix the goods’ description, value, origin and consignee is not made. Records of what entered and what left may be held by the zone operator rather than by customs, and the two ends of the movement are not necessarily reconciled by anyone.
That is a documentary gap by design, created for legitimate reasons of trade facilitation. Its consequence is that cargo can enter under one description and leave under another without any single authority having seen both.
Zones that require declarations on entry and exit, and reconcile them, do not present this problem, and the difference between zones on exactly that point is the useful distinction.
Where the exporter’s obligation actually lands
Enforcement in this area consistently rejects the position that an exporter’s duty ends with its immediate customer.
The Epsilon litigation is the clearest statement of it: the standard applied was whether the exporter knew or had reason to know where the goods were going, and a pattern of sales to a distributor understood to send its products onward met it, notwithstanding that the direct customer was in an unrestricted country.
That is the doctrinal reason the behavioural red flags published by export control authorities carry weight. They describe what an exporter has reason to know, and an exporter who observes them and proceeds has answered the question against itself.
Post-shipment verification, and why it is the decisive test
Of everything on this page, the check that resolves the question outright is the one that happens after the goods have gone.
Export control authorities conduct end-use checks: an officer, or a foreign service post, visits the declared end user and asks to see the item. The possible outcomes are narrow. The item is there and in use, in which case the file closes. The item is not there and the end user can account for it. Or the item is not there and nobody can say where it went, which converts an administrative question into an enforcement one immediately.
The reason this matters analytically is that it is the only check in the whole chain that tests the representation rather than the paperwork. Everything upstream — screening, licensing, documentary review — examines assertions. The end-use check examines the world.
Its limitation is capacity. Checks are conducted on a small fraction of licensed exports, they require cooperation from the host country, and they take months. That is why the volumetric analysis described above matters so much: it is the only method that scales to the whole flow, and it is used to decide where the scarce physical checks should go.