For the time being I trust the U.S. statistics.

The Chinese official data is mostly rubbish. As for the trade data: Since the beginning of the pandemic, China’s own official balance of payments trade surplus even diverges significantly from China’s customs trade surplus, particularly since 2022.

The obvious objective for cooking the books: China is artificially reducing its trade surplus.

We would think that if a foreign firm (or a joint venture between a foreign firm and a Chinese firm operating in China) is manufacturing goods in China for sale in China, the deal would not end up in China's official trade balance, because no good crosses the border.

However, Chinese officials seem to have a different view. In its balance of payments data, China basically reports a trade deficit with itself because of foreign firms producing in China.

Appendix VII of the International Monetary Funds's China Assessment in 2024 - opens pdf - which is when the change in China's statistics was evident - is very revealing.

The divergence seems to be mainly caused by the difference in methodologies to record imports and exports of goods in BOP {Balance of Payments] and Customs. In BOP, imports and exports of goods are recorded when ownership of the goods is transferred between residents and nonresidents regardless of the location of the goods. Customs records imports and exports of goods when the goods physically cross the border of China regardless of ownership of the goods.2 The methodological difference is particularly relevant in the recording of imports and exports related to global production arrangements (e.g., factoryless manufacturing) where nonresident enterprises (e.g., multinational enterprises) outsource part of production to contractors in China

Factoryless is, in this case, the wrong concept as the factories are all in China, they are just (partly) owned by a foreign company.

If the foreign firm then sells the goods that a contract manufacturer produced for it inside China, these goods are counted as an import in the balance of payments data.

As we can reasonably assume, the firms' sales prices for the goods are usually higher than prices the contractors have billed them. The result, therefore, is a trade deficit in the balance of payments.

Simply speaking, if a foreign firm in China produces goods for the Chinese domestic market, it generates a trade deficit, and, therefore, China produces a trade deficit with itself.

The linked IMF report states,

Since 2019, the Customs-based trade surpluses have been persistently above the BOP-based surpluses, with the gap widening significantly over time. In 2023, China’s BOP goods trade surplus was USD 594 billion while Customs recorded a surplus of USD 823 billion, a difference of USD 229 billion or 1.3 percent of GDP.

And:

Exports and imports arising from factoryless manufacturing seem to have been reducing China’s overall goods trade surplus in BOP. When a Chinese contractor sells produced goods to the nonresident enterprise that outsourced the production, exports of goods are recorded in BOP even if the goods remain in China (e.g., in warehouses). If the nonresident enterprise subsequently sells the goods in China, imports of goods are recorded in BOP. Given that the Chinese contractors’ ex-factory price for the nonresident enterprise (China’s exports) is normally lower than the nonresident enterprises’ wholesale price for Chinese distributors (China’s imports), these transactions result in a deficit in the goods trade balance in BOP. Customs does not record exports or imports for these transactions because the goods never cross the border. So, these transactions do not reduce the trade surplus recorded by Customs while they do in BOP.

All this, of course, makes no sense.

And this is one reason why you can't trust Chinese official data.