First up, Matthew Yglesias reports on the situation in Cyprus:
That’s not to say that departure from a currency union is the only scenario in which capital controls make sense. There’s a strong case that the move to free cross-border financial flows was a mistake. In theory, financial globalization could have been a good idea but in practice it did too much to undermine national-level bank regulation. But that’s still not the same as saying that you would have capital controls inside a currency union. After all, if a euro in a Cypriot bank account can’t be swapped for a euro in a Dutch bank account then in what sense are they really both the same currency? A Cypriot euro and a Dutch euro would in some sense be more different than a US dollar and a Canadian dollar.
Indeed, this bifurcation of currency units happened recently with BitCoin. The problem isn’t so much the capital controls in Cyprus, it is the subsequent runs that will happen in other weak EU countries as individuals and businesses flee (say) Greek or Portuguese banks to Germany (and maybe to dollar assets). Matthew gets this one horribly wrong:
To truly have a common currency you need common deposit insurance.
There is not really a compelling reason to publicly insure the asset side of the balance sheet unless you’re also insuring the liability side of the balance sheet (which we shouldn’t). The Euro shouldn’t be kept; but since it will be, there should be EU-wide enforcement of double liability. However, Yglesias does have the best short-term solution to the mess (and long-term solution to diplomatic relations).
The key point, however, is that when FDR tried to give voters what they thought they wanted [debt reduction and a balanced budget, per a Gallup poll], he plunged the economy back into recession, and paid a heavy political price.
Sometimes Krugman acknowledges this, and sometimes not; but attributing the 1937-38 recession to the contraction in fiscal policy is a big stretch, even given Keynesian multipliers. Even the oft-cited increase in reserve requirements seems too small to generate the downturn. The policy responsible for the brunt of the downturn was actually the Fed’s sterilization of gold inflows (and here).