Thread by @Brad_Setser

We have about a year's worth of trade data since the trade war started. That's a period of fairly modest U.S. tariffs (25% on $50b, 10% on $200b). The tariffs also overlap with China's own 2018 policy tightening, which may well have mattered more than the trade war.

1/x

The y/y change in the trailing 12m sum is slow to move (it matches y/y data in December). But if you just look at China's imports of manufactures, I am not sure you would say "trade war" rather than "Chinese slump"

(and a semiconductor price cycle)

2/x

The export data by contrast screams trade war -- China's exports to the US are down, while its exports to the rest of the world (over the last 12ms v preceding 12ms) have held up OK.

3/x

That's always been the weakness in Trump's strategy.

Exports to the US are 20-25 of China's manufacted exports. Taking a whack at them doesn't impact China as much as it used to - as China 75-80% of China's exports don't go to the US and don't face tariffs ...

4/x

But China's non-American trading partners have taken a few lumps from China's slump (falling exports) and their relative openness (no tariffs + weaker yuan = more imports).

China's surplus in manufactures with the rest of the world has jumped by $130b in the last 12ms.

5/x

Some of that is chip prices. But not all. There is a reason why China's exports have held up better than China's imports. The nominal swing in manufactures with the RoW is about 1 pp of China's GDP. That's a big reason why NX contributed significantly to China's growth

6/x

the impact of the trade war can also be seen on the US side.

US non-petrol imports from China are way down, while imports from others have been OK. Usually imports from China track the broader data.

(I shifted from manufacturing to map to the US trade numbers)

7/x

and US imports have done better than US exports (taking China out of the data) even tho the base for imports is much bigger.

(higher frequency data shows a y/y fall in exports, the plotted measure lags)

(8/x)

net result -- while China delivered a nasty negative shock to the world's other manufacturers over the last 12ms, the U.S. has provided them with a pretty decent boost ($80b, 35 bps of US GDP)

(negative number here is a rise in the US trad deficit)

9/x

Puzzling fact to end on --

The bilateral trade deficit with China (which really is the deficit with the broader Asian electronics supply chain) hasn't been all that responsive to moves in the dollar.

But trade with the rest of the world has been very responsive

10/x

to sum up --

China's trade data is consistent with a weak Chinese economy (following policy tightening). Everyone's exports to China are down.

& the first round of Trump's tariffs have cut into the deficit with China, but haven't reduced the overall US deficit

11/x

But that is what has happened --

Trump doubled down on tariffs over the summer (and now perhaps is reconsidering)

& China is under pressure to do a bigger (and a hopefully a more effective for the rest of the world stimulus)

12/12