What to do about CDO ratings II

Robert Waldmann

An earlier post on this topic seems to have been eaten by blogger (update it’s back). This post is roughly half roughly retyping.

I have two proposals (last time I only had one). One is to give up on the ratings agencies. They were geese that laid golden eggs providing an immensely valuable service for a tiny fee, but we’ve killed them. Regulations could be based on firms which put a whole lot of money where their mouth is. Those would be firms which write CDSs. Personally I trust the vampire squid. If G-S is willing to write CDSs with huge notional value on something and sell them for y basis points per year, I’d guess they are pretty safe. GS may or may not be evil but they sure aren’t stupid. This reform would have to wait until the idiot CDS writer problem is solved (can you say AIG-FP ?).

Short of that, I’d say it would work fine to require the agencies to insure 0.01% of anything they rated. It takes a long time for instruments to actually default, but if the agencies balance sheets included CDS written and marked to market, they would be much much more careful.

A problem with this proposal is that the incentives are too long term. Ah that’s another hard problem. The solution is to make compensation above say $500,000 a year of officers of, for example, ratings agencies be paid only in stock which can’t be sold for 10 years (they can live on the dividends until then).

* it’s baaaack