This article asks why the trade gap appears stickier, i.e. is not moving back towards zero, as quickly as it did in the past. Several reasons are cited, the most prominent being the increased proportion of total trade with countries that peg their exchange rate and interfere with the automatic adjustment mechanism:
A Stickier Trade Gap By Daniel Altman, Economic View, NY Times: ...Last year was the eighth in a row with a record-setting deficit in the nation's current account, which includes the trade balance and other income from abroad. Yet once upon a time, that big deficit turned around, and it took a mere five years. Could that, too, happen again this time? ... From 1987 to 1991, the annual current account deficit fell from a peak of almost $161 billion — equivalent to about 3 percent of the domestic economy — to less than $3 billion. This time, however, the challenge is bigger: the deficit of $805 billion for 2005 was about 6 percent of the domestic economy. And there are other factors that could make a turnaround much more difficult.
First, consider what took place in the late 1980's and early 1990's. The dollar was falling rapidly against foreign currencies, partly as a result of coordination by the governments of the world's five biggest economies. As the dollar fell, American investments became less attractive to foreign investors; the same returns would be worth less when converted into their home currencies. The returns themselves were falling, too. ...
Clearly, the United States was becoming a relatively less attractive place to invest. The value of the dollar and investment returns held hands as they jumped off a cliff. At the same time, the current account deficit narrowed. Americans couldn't afford as many foreign goods, and foreigners could afford more American goods.
Now things are heading in the same direction — sort of. While long-term interest rates remain stubbornly low ... in the United States, they are finally starting to rise in Japan and Germany. Those changes may be starting to make a difference in foreign investors' preferences, said Maury N. Harris, chief United States economist at UBS Securities. "When you look at the monthly Treasury international capital flows data, your capital inflows aren't as strong as they were, let's say, six months ago," he said. Mr. Harris said he also expected to see the dollar's value decline this year, by about 9 percent relative to the euro.
But even if exchange rates change with the euro and Japanese yen, it probably won't solve the problem the way it did from 1987 to 1991. The big difference is the "changing shares of who we import from, and very few changes in who we export to," said Catherine L. Mann, a senior fellow at the Institute for International Economics... "It's important, because who we import from increasingly is from countries that have exchange rates that have not moved very much."
Back in 1987, the nation's current account deficit with Japan and Europe, which had fairly flexible currencies, was 62 percent of the total deficit. Last year, it was just 32 percent. In 1987, the current account deficit with the rest of Asia, Africa and Latin America was about 50 percent of the total. In 2005, the share of the deficit for those regions, where many countries link their currencies to the dollar, was an enormous 71 percent. (The numbers in each year don't add up to 100 percent because of surpluses with other countries and international organizations.) These changes matter, because countries like China protect their currency links by keeping plenty of dollar-denominated securities in reserve in their central banks. ...
"The U.S. capital markets are where people want to invest their money right now," [Columbia] Professor [Robert] Hodrick said, "and the performance of our economy has been really extraordinary, so there's no reason to think that that's a bad idea." ... For the current account deficit to turn around in a hurry in this climate ... the United States economy would have to become genuinely weak while that of the rest of the world was strong. Some forecasters do see growth slowing here in the next year or two, while Germany and Japan pick up, and ... the current account deficit could start to shrink by the end of next year. But ... without really big differences in economic growth, a return to balance in the current account could take a decade — not just five years, as before. Given the alternative, that might not be so bad.
Echoing a point Brad Setser also makes, since the investment in the U.S. is largely from foreign central banks, the focus on business conditions and asset returns as an explanation for the trade balance misses a point Larry Summers was trying to make in his speech. Summers believes "an international facility in which countries could invest their excess reserves without taking domestic political responsibility for the process of investment decision and ultimate result" is needed to allow foreign central banks to diversify their reserve holdings.