In this quick blog post Krugman wrote:
There are a lot of good things you can say about international trade. But it does not, repeat not, do anything to alleviate a shortage of overall demand. Yes, if you liberalize trade countries will export more. But they will also import more. If you’re worried about C+I+G+X-M, it’s a wash, because X and M rise equally.
I said in the comments (still awaiting moderation):
On most matters I have the intrepidity to challenge Paul Krugman, but even I hesitate to question him when it comes to a Keynesian analysis of international trade. Nonetheless I must throw caution to the winds and ask:
Dr. Krugman, are you saying that if all countries sealed off their borders completely, world Y (real output) would remain unchanged?
UPDATE: I used a reductio ad absurdum in my comment, but somebody else (Scott Wentland) really hit the nail on the head:
Imports are subtracted from exports…but they are included in consumption. The imports are a wash, but exports do add to GDP…
I have already submitted a Mises Daily on this issue, so I can’t include this point. It’s consistent with what I was saying, but I think Wentland boiled down my entire article to two sentences.
The standard GDP equation measures C as the total amount of money spent on consumption goods by people within the country’s borders. That’s why you need to include the (X-M) component in the first place, to adjust the C and I figures for Americans who spend on foreign goods, and for foreigners who spend on American goods.
So let’s say there initially are 1000 unemployed Americans, and then a new trade deal “creates jobs” so that they can ship $50 million worth of novels to South Asia. Americans collectively then import $50 million of electronics goods from South Asia. That influx of electronic goods into the US is counted as (a) an extra $50 million in consumption spending by US consumers and (b) $50 million worth of imports used to offset the $50 million increase in exports.
Thus, GDP goes up by $50 million, and the “new demand” from Asia allows 1000 previously unemployed people to get jobs paying $50,000 each. If you don’t understand how this is possible, imagine the newly employed workers spend their $50 million on local merchants, who then use the extra income entirely to buy $50 million more in electronics goods from South Asia. There is no reason for anyone in the US to reduce his original consumption spending, and the people who are buying more Asian goods have exactly that much more income to finance it. The “trick” is that you’ve got 1000 people working who previously weren’t doing anything.
Of course in reality, things would be a lot messier. I’m just putting some meat on Wentland’s point, which shows Krugman’s argument is simply wrong.