China’s Stealth Trade Surplus

China’s trade surplus has soared in the last five years.

That basic statement maps to a host of well-known and easily verified realities.

China now runs, for example, a large trade surplus in autos, when it ran a deficit as recently as five years ago.  Net vehicle exports will top 6 million vehicles this year, net passenger car exports will easily top 5 million cars.

It dominates renewables manufacturing (so much so that President Trump decided to essentially give up and take the U.S. back to the age of fossil fuels).

China’s export volume growth has consistently exceeded global trade growth.

Moreover, it maps to standard economic theory: a large real estate crisis typically leads countries to rely more on exports to make up for the fall in internal demand (ask the IMF…)

Yet that surplus often seems to disappear when it comes to the statistics on global imbalances.

China’s 2024 current account surplus was around $430 billion, below $500 billion dollar surplus registered by the European Union (EU). As a share of national GDP, China’s surplus of just over 2 percent of GDP was smaller than the Korea’s surplus, as well as the investment income-driven Japanese surplus (Japan runs a deficit in customs goods trade).

 

So Europe, not China, seems to be the biggest contributor to global imbalances. That, at least, is what is implied by the data sets used by the IMF…

But something is “lost in translation” so to speak.

Compare the customs goods data with the data for goods trade in the balance of payments. Europe’s surplus jumped up, but back in the first part of the previous decade. But the 2020s have been marked by a huge rise in China’s surplus.

Now, there are reasons why goods trade in the balance of payments doesn’t map to goods trade in the customs data.* The main adjustment is a modest one: insurance and freight charges are moved out of the customs goods data. But there are occasionally more complex adjustments from the activities of multinational companies.

The primary reason the EU’s BoP “goods” surplus tops the EU’s customs surplus is that Ireland—and largely because of the activity of a large Irish subsidiary of a U.S. mobile phone company. Apple Ireland technically imports phones from China (and India) or resale (after a markup) to markets outside of the U.S.

China hasn’t really explained why its balance of payments goods surplus tumbled in 2021; the IMF insists that it is one continuous data series, even with the Chinese data revisions.

It (sort of) argued that this was also attributable to Apple’s activity, but that story didn’t check out.

Contrary to the arguments that SAFE made to the IMF, mobile phone exports from China aren’t appearing at an inflated price in the Chinese customs data (China claimed without any supporting evidence that phones delivered to Apple were reported in the customs data at Apple’s global wholesale price, and that the BoP reflected the lower actual payments Hon Hai received from Apple.  The actual customs data though shows China received about $400 a phone in markets dominated by Apple, consistent with estimates of the cost of iPhone production).

And, of course, services (in China’s case, essentially travel to Hong Kong and Southeast Asia) further reduce China’s goods and services surplus in the balance of payments.

There is an interesting general point here: analysts who want to be comprehensive tend to use the goods and services data.  Only Trumpian barbarians still look at bilateral goods data, so to speak, and only barbarians still use the raw customs data.  In the process many analysts tend to pick up the big downward adjustments that happen to the goods data in the balance of payments while often missing the very large rise in the customs surplus.

Such adjustments may (or may not) paint a more accurate picture of global trade flows and trade imbalances, but the net result of looking at the balance of payments data rather than the customs data is clear: a chart of the goods and services balance doesn’t show much of a difference between China and Europe.

China’s goods and services surplus sits at around $650 billion, while the EU’s surplus is only a bit below, at around $620 billion.

Now consider investment income.

Europe doesn’t run a big surplus, despite lots of foreign assets, largely because of the large profits U.S. firms report earning in the EU (~ $200 billion in Ireland alone…)

And China runs a deficit in investment income.  For reasons, well, that no one really can explain.

China’s net international investment position (NIIP) shows a positive balance of about $4 trillion on interest-paying assets (technically, non-FDI assets), up from $3 trillion in 2020. Global rates have gone up since then while Chinese rates have gone down, so it is hard to understand why China’s external income has done nothing over this period (the balance on FDI in the net investment income position of China is now only about a $400 billion deficit, which makes it hard to understand how the investment income balance ends up at a deficit of $150 to $200 billion).

 

The overall result is that the EU reported a bigger current account surplus than China in 2024, and that’s what enters into the IMF’s standard analysis.

There is pretty good case to be made, though, that the real imbalance these days is all in China—and that is its captured better in the customs data than the balance of payments data. The EU’s balance of payments data is distorted by, well, the big impact of U.S. multinationals operating in Ireland. U.S. big pharma inflates the EU’s customs surplus in goods, Apple inflates the BoP goods numbers, and both inflate FDI payments to the rest of the world, which masks the underlying income surplus from the EU’s true foreign investments, etc.   This is something Philip Lane and the ECB actually understand quite well.  And China’s data has become much more opaque over time, as no one can replicate China’s new balance of payments goods series (I don’t think the IMF has even tried), and no one has really figured out where China’s income deficit comes from (the IMF does recognize that China’s published data lacks any detail).

It thus will be interesting to see if the IMF manages to find a way to focus on China in the forthcoming external sector report or if its commitment to using the broad data and trusting national sources (a bit too much) will get in its way once again.

And it will be interesting if the G-7 can find a way to focus on China’s distortions to the global economy, or if Trump’s revolving trade war against allies will get in the way.

* An interesting example is the Airbus A321 exported from Germany. These are German exports in the German customs data and don’t appear in the French customs data, but since Airbus is a unified company, the planes appear as French exports in the BoP data. Airbus the company gets paid for the planes in France and German export to France.