Lots of electronic ink has been spilled over the past few years abcut the monstrous US current account deficit, and the correction which is completely certain but at the same time is also completely unforecastable. (How's that for a nice little economic paradox?)
This is how it is supposed to work. First, investors become less interested in holding US assets (presumably because they don't like the rest of the following scenario). As a result, the dollar loses value against other currencies. Third, the weaker dollar makes US imports expensive, and US exports more competitive. Fourth, the US trade deficit improves as a result, and the US current account balance with it.
However, we've already seen some of steps numbers 1 and 2, because the dollar has indeed weakened substantially over the past few years. Yet the current account deficit keeps getting bigger, not smaller. What gives?
Menzie Chinn points out an interesting paper by Charles Thomas and Jaime Marquez that tries to help explain why. The point of the paper is that the US dollar has not actually weakened that much, if you look at it the right way. I find the paper's conclusions quite persuasive from a technical standpoint.
But just as importantly, the paper raises a crucial point about exchange rates that many people miss, I fear. The exchange rate and the US current account balance are not related to each other in a simple sequential fashion, in the sense that first the exchange rate changes and then the current account balance responds. Rather, they are both "jointly determined", as economists put it, which simply means that they both affect each other. The causation doesn't just go one way.
In this context, what this means is that the results of the paper by Thomas and Marquez serve as an excellent reminder of the fact that we should not expect the value of the dollar to fall by a lot until the underlying causes of the US current account deficit (namely insufficient national savings, both by individuals and by government) change.
Once US consumption growth slows, and/or consumption growth in the rest of the world picks up, the US current account deficit will be able to fall. This will coincide with a loss in the rest of the world's interest in US assets. And so (with a little lead time to allow for a bit of a J-curve effect) that is when we should expect the value of the dollar to fall, and not before.