Chinese exporters pivot to DDP as 'chain owner' model reshapes supply chains

The Analysis

China Water Transport News carried an analysis on 18 July arguing that Chinese manufacturers are moving from FOB and CIF to DDP, building warehouses, fleets and distribution networks abroad, and calling the result the chain-owner model. It names four problems — trade compliance, overseas resources, capital occupation and end-to-end risk — and proposes digital identity, resource coding and standardised multimodal data as the answer.

Strictly speaking, DDP is a delivery rule, not a status. It allocates cost, risk and the clearance task between two contracting parties. It does not say who the importer is in the eyes of the destination authority, and it cannot, because that is not a question a contract between two companies is competent to answer.

The warehouses, the fleets and the data architecture all sit on top of that gap. The gap is where the model is won or lost, and the analysis, careful as it otherwise is, treats it as an implementation detail.

Under DDP the seller delivers to the named place, clears for import and bears the duty. Between the parties that is a complete allocation: the buyer's landed cost becomes one number and the buyer's compliance workload becomes zero. Two parties are free to agree that, and the commercial logic in the analysis is real.

What the agreement cannot do is decide who is liable to the authority. Importer of record, liability for a misdeclaration, the product-safety role in the destination, tax registration, record retention, the answer to an audit: all statutory. They attach by operation of the destination's law, not by the terms of a sale. Two companies can agree who pays; they cannot agree who answers. Every DDP model I have seen fail, failed at that sentence.

So the applicable premise for the whole model is this: in each destination there must be a party that can lawfully stand as importer. Whether a non-resident seller can do that without local establishment, a fiscal representative or a registered entity is a per-jurisdiction question with a per-jurisdiction answer, and the Incoterm does not supply it. The sentence we are going DDP is incomplete until it names that entity, country by country.

Here is the exception that changes the analysis. Where the seller has no establishment, the model in practice runs through a local entity — a subsidiary, an agent, a distributor — inserted between the factory and the customer. Insert it and you are no longer in a clean DDP structure: you have a sale to a related party, and the customs value question changes character, because related-party transaction value has its own test and an invoice does not satisfy it. The tax base stops being the export price, and that risk is not on the analysis's list.

Chain owner is an organisational description, not a legal category. No customs code recognises it. Obligations attach to roles: importer, exporter, carrier, declarant, economic operator and, in product regimes, responsible person. A manufacturer that becomes a chain owner accumulates roles. It does not acquire a status that supersedes them, and it cannot shed a role by naming a counterparty in a contract.

By size, the split is harsh. A manufacturer with volume can staff the importer role in its main markets and absorb an audit. A mid-sized one cannot, so it rents the role, usually from a distributor or a service provider, and the liability then sits with a party whose incentives are not aligned with its own while the reputational exposure stays with the brand. By direction, the change is a transfer of the destination file: an importer that was importer of record under FOB hands that file, and its visibility, to its supplier. What it may not hand over is its own product-law obligations, which in many regimes attach to the person placing goods on the market and are not extinguished by buying DDP. Duty allocation and product liability are two regimes, and a contract fixing the first does not touch the second.

The timing here is an audit cycle, not a shipment. Nothing goes wrong at the port: the invoices look correct, the goods arrive, the customer is happy. The exposure surfaces at the first valuation query, the first product-safety incident, the first registration gap a tax authority notices. That horizon runs in quarters to years, which is exactly why it gets discounted in the business case.

Here is the part the analysis leaves out, and I would put it in front of a board. Its remedy is data: digital identity, resource coding, standardised multimodal data. That solves a visibility problem. The binding constraint is a standing problem, and standing is not produced by data. A perfect trail does not give an entity the legal capacity to be importer, and it does not discharge an obligation that is not delegable. There is a second-order effect running the other way, too: precise records make an exposure legible. A thin record set gives an authority nothing to reconcile against; a complete one gives it a map. The architecture that lets you see your chain lets someone else see it in detail, and in a misdeclaration case the quality of your records is not the defence, it is the evidence.

Run a capital number with my assumptions marked. Assume 100 FEU a month, an average cargo value of 50,000 dollars per FEU, and 45 days of combined transit and overseas warehouse stock. That is about 150 FEU in the pipeline and about 7.5 million dollars of goods, both resting on my assumptions, not on the analysis. Assume a combined duty and import tax burden of 20% — my assumption again, to be replaced with the actual rate from the executing text for each destination — and the cash carried is about 1.5 million dollars, revolving. Under FOB that cash was the buyer's. Under DDP it is yours, it recurs every cycle, and 20% is not the point; the point is that it sits on your balance sheet for 45 days.

Three ways to get this wrong, and they are not symmetric. Going DDP without an entity that can be importer gives you the liability without the standing to discharge it. Staying on FOB keeps the buyer as importer and keeps the buyer's ability to blame your documents. Going DDP through a related local entity gives you control and hands you a related-party valuation question you did not have. Each is defensible; choosing one without knowing which is not.

The conclusion holds only if the destination permits your entity to act as importer, the product regime permits the same entity to be responsible person, and the valuation basis survives a related-party test. Where any one fails, DDP is a price clause rather than a business model, and the warehouses and the platform are assets looking for a legal structure to sit in. That is the boundary I would want written down before the first container moves, not after the first audit.

↑ Back to the key points

— By Dr. Ingrid Voss