Susan Woodward and Robert Hall propose a version (which they credit to Jeremy Bulow) of the “good bank/bad bank” approach:
The “bad bank” holds the equity in the “good bank” plus the bad assets, and owes the non-deposit creditors. The “good bank” continues to hold the good assets and be responsible for the deposits.
If the the “bad bank” can’t keep rolling over its credit, then it gets reorganized: the stockholders are wiped out and the bondholders and other creditors get so-and-so-many cents on the dollar depending on their seniority, just as in any other bankruptcy. But there’s no pressure on the Treasury or the Fed to bail out the “bad bank,” because the depositors’ money is safely in the hands of the well-capitalized “good bank,” which continues doing business just as before, whatever happens to its parent.
This seems obviously right to me, and — more significantly — to Brad DeLong, who calls it “the cleverest plan I have yet seen.” It’s obviously bad for the stockholders and creditors of the banks, compared to being bailed out by the taxpayers, but that looks to me like a feature rather than a bug. I have no objection in principle to temporary nationalization Swedish-style, but this looks like the right way to avoid having to do it.
Technical question: Since the “good bank” equity has actual market value, could the “bad bank” do a quick underwriting of a partial stake in the “good bank” to generate cash with which to pay off its short-term creditors?