Five B-school professors from the University of Chicago write,

we believe that it might be even better if the Treasury plan was divided into two components — a plan to “liquify” certain moribund markets, thus allowing financial institutions to sell illiquid assets, and a plan to raise capital levels in financial institutions.

Banks need to be free to lend more money, and they are constrained by a lack of capital. I propose relaxing capitall requirements. The authors suggest an alternative of having banks raise more capital. They want to do this anyway, but the clever idea is that a government mandate could make it easier for banks to do so.

For an individual bank, going out to raise capital arouses suspicion (why are they doing that? are they in trouble?) As the authors put it,

The value of mandating this decision is that no individual bank sends an adverse signal to the market when it goes to raise capital.

This strikes me as too clever. I admire their use of game theory, but I doubt their sense of real-world context.I went back and read the paper that two of the authors wrote, and now I see that I misunderstood it. I thought they were taking the view that capital requirements ought to be countercyclical–allow banks to hold less capital in a downturn. But that’s a conventional view. Their novelty was to propose an insurance fund that could inject capital in case of a crisis. The insurance fund would substitute for capital. Off hand, I don’t see how it differs from capital. Either way, you are keeping funds tied up that cannot be used for loans but that can be used to cover losses.