Has the euro been a disaster, or what?
It seems pretty self-evident that it has been. Unemployment is 27.6 percent in Greece, 26.2 percent in Spain, 16.5 percent in Portugal, and 13.6 percent in Ireland, which, remember, is supposed to be the austerity success story. What's happened? Well, exactly what euro-skeptics feared would happen from the time the common currency was just an idea: a shock hit some parts of Europe worse than others, and there hasn't been any easy way to adjust. The ECB's one-size-fits-Germany policy has left crisis countries no choice but to try to pay-cut their way to prosperity -- which would be painful enough if it were even possible. But it's really not when interest rates are all but at zero.
In other words, the euro has turned a recession into a depression, because it doesn't let its countries fight either. They can't devalue their currency or cut interest rates or even run bigger deficits when they get into trouble, so their trouble gets worse. That's why Paul Krugman is rightly skeptical of a new Andy Rose paper that claims countries with "hard fixed exchange rates" (like the euro) have done no worse since 2006 than countries with inflation-targeting central banks. It just goes against all intuition that a less flexible monetary would work as well as a more flexible one during a global financial crisis. Against all evidence too. As Krugman points out below, higher sovereign debt levels have been a problem for some euro members, but not for anybody else. How could it be true that there's no difference between euro and non-euro countries?
Well, it's not. The Rose paper excludes euro countries from its sample. It only includes countries with either "no separate legal tender, a currency board, or a conventional peg" among its so-called "hard-fixers." It's an odd choice. See, the euro zone doesn't have a fiscal or banking union, so its currency union really shouldn't be treated as a whole. Euro members aren't so much parts of a federal state as countries with just particularly hard-to-break currency pegs. Leaving them out of the sample doesn't make much much sense, and unfortunately biases the results.
But that's not the only bias. You can see in Rose's table below that only 63 of the 83 countries he counts as hard fixers in 2006 were still hard fixers by 2012. And even that overstates things a bit. Three of those 63 countries that began and ended with hard fixed exchange rates temporarily ditched them in between. So only 60 of these original 83 hard fixers were hard fixers throughout.
What should we do with these 23 countries that dropped their hard fixed exchange rates? Presumably they did so because it would have been too painful to maintain their fixed exchange rates during the financial crisis. In other words, they faced a choice between higher unemployment and a weaker currency, and chose the latter. If that's the case, we can't ignore them or we'll be ignoring the countries that probably did the worst with hard fixed exchange rates. But Rose does ignore them. He basically looks at the 60 countries where hard fixed exchange rates worked, and concludes that they work. It's selection bias all the way down.
Fixed exchange rates work until they don't. But when they don't, they really don't. Like Great Depression don't. So it'd be strange if I said the gold standard had worked as well as other monetary regimes if you just disregard the 1930s. It's equally strange to say fixed exchange rates are working today if you just disregard the countries where they aren't. Because there are plenty of those.
Half a continent of them, in fact.