Note To Readers: China Doesn't Seasonally Adjust Its Trade Statistics
Ironically though, a big increase in its trade surplus in February was also interpreted as signs of a dramatic slowdown since the big drop in imports indicated falling demand.
The truth is however that although these numbers support the view that economic growth is slowing in China, the shift isn't as big as the isolated numbers alome suggest.
The by far most important reason for both the very high trade surplus in January and the record trade deficit in February is the Chinese New Year, also known as the Lunar New Year. Because it is a very important holiday in China most businesses are closed during those days. Last year it was in February, this year it was in January, meaning that a lot of imports that was made in January last year was shifted to February this year. The Chinese New Year also shifted a lot of exports from January to February, but this effect was smaller. Add to that the fact that February in all countries had 29 days this year instead of 28 as in most years, and we have a very big seasonal effect.
One way to adjust at least for the Chinese New Year factor is to look at the aggregate sum for January and February. And as it turns out, neither exports nor imports changed much. Exports in China rose 6.9% while imports rose 7.7%, causing the aggregate trade deficit for the two months to increase from $890 million to $4.25 billion-
Note also that the fact that China had a trade deficit in January-February doesn't mean that it runs a structural trade deficit. What it means is that Chinese businesses in the beginning of the year exports less and imports more than later in
the year, particularly the months before Christmas when it delivers a lot of toys and other Christmas presents.
The big slowdown in growth of both exports and imports certainly supports the view that China's economic growth is slowing down. However, the slowdown isn't as dramatic as the interpretations of either January's or February's numbers suggest. If China had tried to use seasonal- and calendar adjustments of its data this misinterpretation hadn't arisen.