Before a single parcel reaches a Chinese shopper, a brand has already made a customs decision. The declaration model chosen at launch decides how duty is collected, how fast stock clears the border, what paperwork follows every order, and how much working capital sits inside the supervision zone. Most brands pick a model by habit and discover the cost only when cash flow tightens. This guide compares the three routes that matter for transfrontalier e-commerce in 2026 — 9610, 1210 and 1239 — and explains when each one fits.
Why the declaration model is a structural choice
A declaration model is not a form you file once. It is the rule set that governs every shipment for the life of the channel. The model fixes where duty is paid, whether goods sit in bond before sale, how returns are treated, and which system of record the customs authority trusts. Changing the model later means re-filing product records, re-mapping the ledger, and sometimes moving physical inventory between zones. The cheap time to choose correctly is before launch, not after a peak-season bottleneck.
9610 — consolidated retail export
Customs code 9610 covers transfrontalier retail exports shipped in consolidated batches and declared against the individual orders after they are placed. It suits brands sending goods out of China to overseas consumers, or selling from an overseas base into a market through a consolidated declaration. The strength of 9610 is speed and low fixed overhead: parcels move in bulk, and the order-level declaration is matched after the fact. The weakness is that it assumes the goods are not sitting in a Chinese bonded zone waiting for a domestic sale, so it is the wrong tool for the classic import-and-sell model most foreign brands use.
1210 — bonded import for transfrontalier e-commerce
Code 1210 is the bonded import route and the one the majority of foreign brands use to sell into China. Stock is imported into a supervised bonded zone before sale, duty is suspended while the goods wait, and duty is collected per parcel only at the moment of sale to a Chinese consumer. The buyer sees domestic delivery speed, while the brand avoids paying duty on unsold inventory. The trade-off is discipline: every movement must be booked to a product ledger, and the three-document match at the point of sale must reconcile the order, the payment and the logistics record.
1239 — the pilot bonded route with a simplified bond
Code 1239 is the pilot version of bonded import that lowers the up-front bond a brand must post to enter the zone. It gives smaller brands access to bonded fulfilment without tying up as much working capital, but it is available only in designated pilot zones and carries tighter supervision on certain categories. For a brand testing the China market with limited volume, 1239 can be the cheapest way to offer bonded-speed delivery; for a brand scaling fast, 1210 usually becomes the better long-term fit.
How the three compare on duty, speed and paperwork
On duty, 1210 and 1239 both suspend duty until sale, while 9610 collects at export. On speed to the consumer, 1210 and 1239 win because stock is already in zone. On paperwork, 9610 is lightest at launch but offers no bonded advantage, 1210 is the most mature and widely supported, and 1239 adds pilot-zone constraints. A brand should score the three against its own volume, capital limit and category before committing.
Matching a model to your product and channel
Low-volume test launches with thin capital favour 1239. Established brands with steady China demand and the working capital to hold bond favour 1210. Brands whose goods never sit in a Chinese zone — because they ship from overseas or export out — have no use for bonded routes and should stay on 9610 or a direct model. Category also matters: high-value or sensitive goods attract more supervision, and some pilot zones restrict them under 1239.
Switching models without breaking the ledger
If volume grows and a brand moves from 1239 to 1210, the transfer must be planned, not improvised. Freeze the affected product records, reconcile the open ledger to a clean closing balance, and move physical stock only after the new record is filed. An unplanned switch leaves two ledgers open at once and turns a routine upgrade into a customs finding. The broker should document the cut-over date so every future write-off knows which model governed which period.
Questions to ask your customs broker
Five questions separate a broker who understands transfrontalier e-commerce from one who only files forms. Which zones support 1239 for your category? What bond does 1210 require at your volume? How is the three-document match logged when a payment is subsidised by the platform? Who notifies a ledger discrepancy and in what format? And can the broker produce a parcel-level trail within a stated number of hours? The answers decide whether bonded fulfilment protects margin or quietly drains it.
Choosing 9610, 1210 or 1239 is not a back-office detail. It is the operating model for selling into China, and the right choice therefore keeps duty, speed and paperwork in careful balance while the brand scales.