Mark Gertler and Nobuhiro Kiyataki write,

As balance sheets strengthen with improved economics conditions, the external …finance problem declines, which works to enhance borrower spending, thus enhancing the boom. Along the way, there is mutual feedback between the …financial and real sectors. In this framework, a crisis is a situation where balance sheets of borrowers deteriorate sharply, possibly associated with a sharp deterioration in asset prices, causing the external fi…nance premium to jump. The impact of the …financial distress on the cost of credit then depresses real activity.

Pointer from David Warsh, who has other interesting links. This strikes me as a paper that is reaching to find a theoretical justification for bank bailouts. The thing is, you have to assume that the intermediation that the banks were providing was useful in the first place, as opposed to distorting the allocation of capital. I have strong doubts about this approach.

Next, Brad Cornell writes,

just as a command economy is invariably less efficient at resource allocation and production than a market economy, a general stimulus program will, in all likelihood, lead to highly inefficient allocations, effectively burning resources at a time when they are scarce and particularly vital to restart and re‑align our beleaguered economies. To the extent that it is used to prop up declining industries, the stimulus could even prove harmful by delaying necessary adjustments.

Thanks to Mark Thoma for forwarding the link via email. A few months ago, I pointed out that Cornell is a Recalculationist.

Finally, let me recommend Garett Jones interviewed by Russ Roberts, particularly the latter half of the interview. Jones elaborates on his view that workers produce organizational capital rather than output.

The other theory of the recession is that it is a nominal shock. So we have three theories–financial shock, real shock (Recalculation), and nominal shock. They are not mutually exclusive. I suspect that those who might hope for a decisive piece of empirical evidence will be disappointed. For example, Roberts seems to favor a recalculation story as he points out that 22 percent of the jobs lost have been in construction (I wonder about how such a figure can be calculated, given many conceptual problems. Gross jobs losses? Net job losses? etc.) But if the implication is that 78 percent of the jobs lost were elsewhere, someone could easily use the point to make a different case.