> Actually, though, it was only a small proportion of the underlying
> mortgages that were defaulting (though larger than the norm). That led
> to a crisis in liquidity in the asset class as a whole because units
> were unwilling to buy what they were unsure of. Mark-to-market
> accounting records that as a plunge in the value of the whole range of
> mortgage-backed securities even though the bulk of the underlying
> payments will continue to come through and values are likely to
> recover in the future. It is quite possible that different
> market/institutional configurations could have handled it differently
> - and indeed much of the government rescue around the world is based
> on the assumption that demand for many assets will eventually recover
> - i.e. it's about liquidity rather than the worthlessness of the
> underlying revenue streams.
>
You're right, but I would put it somewhat differently. The specific market/institutional configuration that made default on a (relatively) small class of securities so disastrous was the existence of a parallel banking system for large firms and banks - the repo market - that used AAA-rated MBS *as money* under the false or unthinking assumption that the value of those securities' was as secure as that of Treasury bonds. When a particular type of asset depreciates unexpectedly, it makes a huge difference whether that asset had been regarded as a speculative investment or as a near-perfectly liquid form of payment. If it's the former, no problem, that's life. If it's the latter, it will cause an enormous panic.
It doesn't really make sense to say that a more transparent pricing mechanism for MBS would have averted the problem, because if market participants had thought such a mechanism was needed, they wouldn't have been using MBS as money in the first place. The whole point of choosing an asset to use as money is that you choose one whose value you don't need to devote any resources to determining. Setting up transparent pricing markets for an asset presupposes the need to gather costly information about that asset, which vitiates the whole point of using the asset as money. In just the same way, it doesn't make any sense to say 19th century bank runs could have been averted if small-town banks' loan portfolios and financial conditions had been more publicized and transparent. If they had been, nobody would have wanted to use those banks' checks and deposits as money, because they would always be afraid some savvier market actor would invest more in information-gathering in order to arbitrage the different risks of checks written on different banks and the less-informed actors - the local grocer who takes checks, for instance - would get screwed.
SA