I’m not so certain that the current financial crisis indicates the ultimately “subjective”, “irrational” basis of finance capital - its dependency on “human” factors such as trust, confidence and so on. The “trust” and “confidence” operational in markets are not names for the affective states of actual traders, but labels for conjectures about the likely behaviour of other traders. The real basis for that behaviour is, in turn, the making and testing of further conjectures of the same kind.

For this reason, I’m inclined to agree with those analyses that suggest that the best response to the crisis is not to “bolster” the markets (either with cash or with talk) so that everyone feels better about their prospects but to enforce better disclosure, to improve the information available to all players of the game (even if this hurts those individual players who have painful* things to disclose). What has happened within the finance system is that the quality of information has deteriorated to the point where it is no longer possible to form sufficiently strong conjectures about how the entities being traded within the system will be valued in the future. A bubble can be supported by weak conjectures, so long as reasonably strong conjectures can be made about the future behaviour of the bubble itself, but when it’s a safe bet that the bubble will burst there’s nothing else left that you can bet safely on.

It would also follow that the purpose of future regulation should be to maintain the quality of information, even (especially!) if this impeded the growth of future bubbles. Obscurantist financial instruments should be outlawed; or restricted to a special sandpit (like the International Obfuscated C Code Contest) where the kinds of minds that enjoy devising them can run riot without causing further harm.

* i.e. quite possibly illegal