I am quoting his entire post, not to rip him off but because the post is short and unexcerptable. But DD is always worth reading. Please add him to your RSS.
Gahhhh, economics, my greatest pleasure but the source of all my frustrations. Today I am mostly being irritated by the theory of optimal currency areas.
Look guys. It is true that the USA has somewhat higher internal labour mobility than the Euro area. It is not true!!!!!1! that this internal labour mobility equilibrates regional economic differences to any economically significant extent. The Rust Belt happened. Michigan is how it is. The oil states have oil booms and oil busts. Florida is procyclical because of tourism. California had a tech boom and bust.
In actual fact, regional fluctuations in the USA are smoothed out by transfers from central government and by countercyclical monetary policy at the federal level. The USA is NOT an optimal currency area, or even close to being one.
Also not an optimal currency area: Canada. And Germany.
Exactly right. So why do econotechnocratic dreams go awry? Politics. Yes, politics. Politics in the U.S., politics in the EMU, politics in Canada and Germany even. So let's start paying serious attention to politics, eh? (I.e., if Ben Nelson doesn't vote for another stimulus package, maybe it's not because he's an evil person who knows nothing about economics, but rather because he's a self-interested person who knows something about being re-elected.)
This should be a recurring theme on this blog over the next few weeks (and probably months), so I'll keep this short for now. But I'd like to leave this open to readers: is there such a thing as an "optimal currency area", expanding the definition to include political as well as economic jurisdictions? If so, what would it look like?