Despite what Lex says, the euro/ dollar does matter

Dean Baker’s critique of Monday’s Lex column on the dollar was on the mark.  Lex made rather selective use of the facts to argue that dollar depreciation (and renminbi appreciation) has little impact on the US trade balance. 

I suspect that the new conventional wisdom is closer to Lex’s view than my view.  So I wanted to lay out my critique of Lex's argument in some detail.  

Lex discusses both the RMB/ $ and the $/ euro.  And, unusually, rather than focusing on the RMB, I want to focus on the euro.   (Graphs are below the fold). 

Before doing that though, let me just say that Lex might want to look at the evolution of the eurozone’s trade balance with China before arguing that exchange rate changes don't matter.  

It is a natural test.  The RMB has been stable against the dollar for some time, but it has been anything but stable against the euro.  And it seems pretty clear to me that the acceleration in Chinese export growth to Europe in 03 and 04 followed directly from the RMB’s depreciation against the euro from 2002 on.  China now exports about as much to Europe as to the US – and runs a significant bilateral surplus with the eurozone.  The ECB puts out the data in table 7.6 of its monthly bulletin; it sure seems to me like the eurozone’s deficit with China went from around euro 30b in 2002 to euro 75b in 2005, largely because imports from China rose from a bit over euro 60b to a bit under euro 120b in three years.   

What of the $/ euro?   Lex argues that the dollar’s depreciation against the euro has had no impact on the US bilateral deficit with Europe.  That certainly seems to be the case if you eyeball the data. 

But looking at the overall balance can be deceptive.  The exchange rate is only one of many variables that influences the bilateral trade balance. Relative growth in domestic demand also matters.    Particularly for imports.   The exchange rate matters more for exports.  At least that is what the econometrics says.

Did the dollar’s fall have the expected impact on US exports to Europe.  Absolutely. Look at the following graph.   The dollar/ euro has been inverted so a fall in the graph is a fall in the dollar.

More importantly, the dollar euro has an impact not just on US exports to Europe, but US exports to the world.   American producers compete against European producers in a range of markets.  Aircraft and aircraft engines is just the most visible.  Again, did changes in the dollar/ euro have the expected impact.  Absolutely.  Just with a lag.  

Indeed, if I plot US exports (the rolling four quarter total) against the exchange rate with a one year (four quarter) lag, the link seems really, really clear. At least to me.

Changes in the dollar’s value aren’t sufficient to bring about a fall in the trade deficit.  Domestic demand and domestic savings also have to adjust.  But changes in the dollar’s value do have a big impact on US exports.   Despite what the Lex argues.  

US exports have been growing at about 12% recently.  Average growth over the past twenty years has been more like 6-7%.   

The US got its export boom from the falling dollar.  It just came in the midst of an even bigger import boom.  So it didn’t keep the deficit from rising.

Finally – for the sake of completeness, I’ll throw in the graph of the US bilateral deficit with Europe.  It certainly hasn’t fallen along with the dollar.   But that is because US import growth was strong, offsetting the weak dollar’s impact on exports.   As Dean Baker argues, the bilateral deficit likely would have been even wider if the falling dollar hadn't pushed US exports up.  Now that the US is slowing, the US bilateral deficit with Europe is set to stabilize if not narrow.  A small note – my graph of the bilateral balance is basically the same as the graph in Lex, I just inverted both the deficit and the dollar.

My data set ended at the end of q1.  But, given the June data, the overall stabilization of the overall US bilateral deficit with Europe should be more clear once we get the seasonally adjusted data for q2.  My data set also includes all of europe, not just the eurozone.  But that shouldn't matter that much, since most European currencies have moved with the euro against the dollar …