I say this for four reasons:
1. Banks lost money on mortgage derivatives because of principal-agent failings. Principals - banks’ bosses - didn’t understand what agents (traders) were doing, and traders had incentives to take on excessive risk, because the gains from doing so - a life-changing bonus - exceeded the benefits of prudence.
2. Banks have been reluctant to lend to each other not so much because each bank fears its counterparty will not repay the money, but because they fear they’ll need the money themselves. This is because banks just don’t know what sort of losses they are sitting on. It’s impossible for managers of such complex organizations to know everything.
3. Banks are under-capitalized because chief executives have traditionally had incentives to maximize earnings by using leverage. Pressure upon them to be more prudent has been absent partly because when shareholding is dispersed, no individual shareholder has much incentive to rein in management.
4. Good financial innovation - of the sort advocated by Robert Shiller - has been lacking because it’s very difficult for anyone to own its beneficial effects; it’s a public good. By contrast, the gains from “bad” financial innovation - overly complex mortgage derivatives - are more appropriable. So we get more of it.
Of course, there are good reasons why capitalists’ apologists have not made the distinction between ownership structures and markets - because the former have always been less legitimate than the latter. But we should make the distinction now. What we’re seeing is not a market failure, but an ownership failure.
What’s the solution?
Not nationalization - which is a bad way of solving principal agent problems.
Perhaps instead, banks should make more use of internal markets. They should become more like venture capitalists, allocating capital to semi-independent divisions, which put in their own capital. This would restrain traders’ risk-taking, as they can not so easily hide behind the fact that losses are spread over the whole firm. And it would reduce the problem of asymmetric information between banks’ senior managers and trading desks, as there’s a simpler test of how well the latter do - whether they can hand over enough hard cash to cover their required returns.
Now, I’m being deliberately vague here. I just want to stress that ownership is the problem, more than markets.