China+1 and inland China: where those who leave actually go, and what really changes

This guide finishes a sentence from the previous text. In the end of cheap labour we reached a rule that settles half the subject: the more of the final price is human hands, the more the country of origin matters. It was said there that where those who leave go belongs to a guide of its own, because the answer is of another kind. This is it.

China+1 answers a question that is not yours

The strategy was born as risk management, long before the tariffs: do not leave the whole of your production in one city, one supplier or one country. In that original form it is sensible and remains so. What happened afterwards is that it turned into something else, and it was that something else that reached Brazil through translated news.

The American buyer began facing a wall built by origin. When the cost of importing changes according to the stamp of provenance, changing the stamp becomes a financial project, which is why the business press of recent years is full of factories that crossed borders with exactly that stated aim. The next step came with it: once the destination country notices that only the last stage changed address, it starts treating transhipment as a category of its own, with worse treatment than the country of shipment gets. In other words, the race for the stamp has a ready answer on the other side.

Now read that from the Brazilian side. The import duty here is set by the tariff classification of the goods, the NCM code, and Brazil does not apply a blanket punitive rate against Chinese origin. Swapping Shenzhen for Haiphong, in the overwhelming majority of products, does not change a cent of what you pay in. The central financial motive of the American playbook simply does not exist in your operation.

What is left, and it matters: tariffs aside, three legitimate reasons still stand for a Brazilian importer to consider another origin. A product in which labour dominates the cost. A trade defence measure in force against Chinese origin in your specific product. And genuine risk concentration, when everything depends on one factory, one city or a single port. Outside those three, what usually happens is that the importer pays the cost of the move without collecting its benefit.

What actually migrated, and what stayed put

The migration is real, it is large and it is old. It is just selective in a fairly predictable way, and the rule from the previous guide explains exactly who went and who stayed.

What is made by hands went away. Clothing, footwear, costume jewellery, simple toys, bench assembly of a set, hand finishing. These are products in which the wage is the largest line of the cost, and where a difference in pay shows up directly in the price of the piece. Vietnam, Bangladesh, India, Indonesia and Cambodia absorbed a good deal of that, and it was not a passing fashion.

What is made by machine stayed. The electronic component, the injection mould, the metal sheet, the circuit board, the motor, the battery cell, the machine that makes the piece. At those stages the wage is a small fraction of the cost, and what decides is scale, automation and the ring of suppliers that the map of the industrial clusters describes city by city.

Out of that comes the fact that almost never appears in the news about China+1, and it is the most useful of all: the factory that left goes on buying from China. Fabric, trimmings, components, tooling and machinery cross the border to feed the new line. What changed country was the last stage, not the chain. Whoever imports from that factory has not swapped one supply chain for another: they now have the same chain with one more country in the middle.

And this has stopped being an impression. The 2025 figures from Vietnam's statistics office describe exactly that geography. The country moved 930.05 billion dollars in trade, with 475.04 in exports, 455.01 in imports and a surplus of 20.03. The detail that matters is in the composition of what it buys: 93.6% of everything Vietnam imports is production input, between machinery, equipment, raw materials and fuel, against only 6.4% of consumer goods. The country does not import in order to consume. It imports in order to produce.

And where that input comes from. China is Vietnam's largest supplier, at 186 billion dollars in 2025, and the Vietnamese deficit with it reached 115.6 billion, 39.6% above the previous year. On the other side of the counter, the United States is Vietnam's largest customer, at 153.2 billion. Read the two lines together, because that is where the story tells itself: Vietnam bought more from China than it sold to America. The last stage changed address. The chain did not.

Made in Vietnam is a legal status, not an address

This is the point where the subject stops being strategy and becomes paperwork, and where most people get hurt.

Origin, in foreign trade, is not the place where the box was sealed. It is a legal outcome, defined by substantial transformation rules. They vary with the agreement and the product, but they usually ask for one of two things: that the operation carried out in the country change the tariff classification of the goods, or that a minimum percentage of the value have actually been produced there.

In practice that separates two scenarios which look identical from the outside:

For the Brazilian importer this has consequences in two places. The first is the preferential agreement: origin is what grants the right to a reduced rate, which is why the subject comes up so strongly when we talk about importing from China into Mercosur. The second is trade defence, which is the subject of the next section and where origin really moves your cost.

The practical rule: if the operation only adds up because of origin, it has to be designed with your customs broker before the first shipment, with the description of the production process in hand, and not after the cargo has arrived. Declared origin can be challenged, and a weak origin is defended with production records, not with argument.

The one case where origin changes your bill

There is one situation in which switching country genuinely changes the number at the bottom of the Brazilian spreadsheet, and it is specific: when there is a trade defence measure in force for your product against Chinese origin. An anti-dumping duty is not import duty, it is an additional amount charged on a product of a given origin, and it can be high enough to sink a whole category on its own.

Three things matter before you build any plan on top of that:

• The measure is by product and by origin, and it is looked up by NCM code. Not by the commercial name, not by the supplier's description. The list of measures in force is public and the ministry keeps it up to date.
• The measure has an expiry date. An anti-dumping duty runs for a set period and can be extended through a sunset review. Building a new operation on top of a measure that expires next year is betting on a calendar that is not yours.
• The purely formal move is foreseen, and it has an answer. Brazilian trade defence law provides for extending the duty when production in the third country is, in practice, the assembly of parts and pieces from the origin under investigation. In other words, changing the address of the last stage is not a safe route around the measure, and the difference between a real factory and a kit assembly line is precisely what gets investigated.

Translated into the decision: if a measure exists for your NCM code, changing origin can be the most significant saving in your operation; if none exists, it almost never pays for itself. That check takes an afternoon and saves months of the wrong project.

The way out almost nobody writes about: inland China

Look at what is left. For a product in which labour weighs, the importer's real objective is to pay less per hour worked. Nothing in that objective requires crossing a border, and China is a continental country with an enormous internal wage gap.

The national average hides both ends, and the National Bureau of Statistics of China publishes the regional breakdown itself. In the release of 16 May 2026, with 2025 data, the picture is this:

The column that matters to the buyer is the second one, because the factory that answers a request for quotation sits among the private units. In it, the coast pays 37% more than the centre of the country. In the non-private column, which carries state-owned firms and multinationals, the distance is wider still, at 46%. It is that step which moved, over the past fifteen years, a gigantic slice of Chinese manufacturing into the country's own interior. Henan and Zhengzhou, Anhui, Jiangxi, Hunan and Hubei stopped being industrial periphery and became a destination for production lines, pulled by provincial incentives, by a workforce that started staying home instead of migrating to the coast, and by a rail and road network built for exactly that.

And here comes the part that dismantles the easy shortcut. A 37% cut is large, but it is not the cut Southeast Asia offers. The average monthly income in Vietnam's industry and construction sat close to 9.1 million dong, somewhere around 350 dollars, while the annual average in the private units of Chinese manufacturing was 76,055 yuan, close to 6,300 yuan a month. Even comparing Vietnam with the cheapest region of China, the distance is still measured in multiples, not in percentage points. Switching province does not match switching country on the cost of the hour worked, and anyone who says otherwise is selling something.

The two series are not comparable by construction, and that has to be said in full. The Chinese one is collected from enterprises and measures annual wages in urban units; the Vietnamese one comes from a survey of workers' income and is monthly. Add the exchange rate in the middle and the difference in sector composition. The figures serve for the order of magnitude, which is what the decision needs, and not for a cost spreadsheet. Anyone comparing seriously compares a factory quotation against a factory quotation, for the same product and the same specification.

So why does the interior pay off, if it loses on the hour worked? Because it is not selling the cheapest hour in the world. It is selling a meaningful cut in cost without charging the price of the chain, and that is where it wins:

• The chain stays on the same side of the customs border. The component, the mould, the surface treatment and the emergency supplier all remain inside the same country, in the same language, in the same currency and with no customs clearance between one stage and the next.
• The supplier is often the same group. Many inland plants are branches of coastal companies that kept engineering, sales and quality control on the coast and moved the labour-intensive part inland. You negotiate with people you already know.
• The rule of origin does not change. It is still a Chinese product, with the same paperwork, the same tariff treatment and no argument about transhipment.

And it has a price, which has to enter the accounts from day one:

• Your cargo leaves through a coastal port either way. A factory in the interior adds one to two thousand kilometres of overland leg before loading. That costs money and costs days, and the wage saving pays for that leg only in part, depending on the product.
• Local competition is thinner. Where thirty factories make your product, the quote disciplines itself; where three do, it does not. Fewer options also means less of a plan B when one of them is late.
• Export experience tends to be thinner too. Paperwork, packing for a long voyage, marking and conversation in English are points where the inland plant is frequently behind the coastal one, and that shows up on the first shipment, not in the negotiation.

Where the interior wins comfortably: labour-heavy products, bulky low-value products, and a large order with a comfortable deadline. Where it loses: anything sensitive to lead time, small orders with many shipments, and a category in which you are still testing suppliers and may need to switch quickly.

The three costs that never make it into the spreadsheet

Every origin comparison starts with a unit price spreadsheet, and it is precisely what the spreadsheet does not have that usually decides the outcome.

None of this is an argument against moving. It is an argument against moving on the strength of one column. Once the four items on the right enter the accounts, a large share of the moves that looked obvious stops adding up, and the ones that survive become a great deal more solid.

When China+1 really is worth it

It would be dishonest to close without the other half, as in the previous guide. There are situations in which leaving is the right decision, and they are identifiable:

• The product is dominated by labour. Clothing, footwear, hand assembly, hand finishing. Here the wage difference is large enough to pay for everything that is lost, and that is why this migration is already a settled fact and not a forecast.
• An anti-dumping duty is in force against Chinese origin under your NCM code. This is the case where the saving is immediate and verifiable, provided the new origin is real and will survive a check.
• The risk concentration is genuine. A single supplier, in a single city, for a product that carries your revenue. Here the second origin does not even have to be cheaper: it is insurance.
• Your own customer demands non-Chinese origin. It happens to anyone reselling to an American or European buyer who has their own restriction. In that case origin is a contract requirement, and the cost discussion is secondary.

Outside those four, and above all in a product of components, mould and machine, the maths rarely adds up. What does add up is looking at the Chinese map from the inside instead of looking at the map of Southeast Asia from the outside.

How to decide, product by product

The decision does not belong to the catalogue, it belongs to each item. Five questions, in this order, settle almost every case:

01 How much of your cost is human hands?

This question does not have two answers, it has three, and it is the one that separates the two routes in this guide:

• Labour dominates the cost (sewing, footwear, hand assembly, hand finishing): Southeast Asia wins on the hour worked by a margin the Chinese interior cannot reach. Here China+1 is real, and the migration has already happened.
• Labour weighs, but does not rule (assembly with components, semi-automatic lines, products with many operations and their own tooling): the Centre of China is usually the right answer, because 37% is enough to move the price and the chain stays within reach.
• Labour is a small fraction (components, mould and machine dominate): do not switch country or province over wages. Neither route pays for itself, and the road is another one.

Working out which of the three bands your product sits in is an afternoon's work, with your current factory answering in writing how many people touch the piece and for how long. Without that, the four questions that follow have nothing to stand on.

02 Is there a trade defence measure for your NCM code?

A public search, done by the exact classification and by origin. If there is one, note the expiry date too, and whether a review is under way, because that changes the horizon of the decision.

03 Where does the main component come from?

If the answer is China, switching country does not remove the dependence, it merely adds a border to it. Ask the candidate factory, in writing, for the origin of the inputs it buys.

04 What lead time can you genuinely absorb?

A new origin means a pilot batch, adjustment and rework before the first production order. Anyone whose year is already committed on the calendar has no slack for that curve, and is better off moving in the off-season.

05 What is your volume?

Small volume has little leverage in a young industry and pays dearly for the minimum order. Large volume commands attention and can negotiate the learning curve with the factory.

What still cannot be settled from a distance

Look at what comes out of all this. The questions that decide the move are not about price, they are about fact, and none of them is answered by a catalogue, by a video of a production line or by a video call booked well in advance.

Is the Vietnamese plant that quoted a factory, or a finishing shed that receives Chinese kits and closes boxes? You see it from what is on the floor: stacked raw material is one thing, sealed import cartons are another. Does the inland Chinese plant that quoted actually produce, or is it the address of a coastal company that subcontracts there and has never set foot on the line? Will the origin the whole operation presupposes survive a check, with the production process described and evidenced? And does the ring of suppliers exist within reach, or will the missing stage be subcontracted eight hundred kilometres away, on the subcontractor's schedule?

Those four are answered standing up, inside the factory, watching who operates and what is on the floor. That is what our team does at source, and it is the reason the trip exists. For anyone starting there, the guide on Shenzhen from the inside teaches you to tell a factory from a trading company, an agent and a sales office, the one on Yiwu beyond the market shows who is really behind the counter of a market, and the one on quality inspection before shipment explains what gets checked when the factory is new and has proved nothing yet.

BCVN — Brasil China Viagens e Negócios has connected Brazilian companies to the best suppliers in China and Asia since 2008. Find out about our services in commercial intermediation, quality inspection, factory audits and guided business trips to China's main trade fairs.

Keep reading