DeLong, Krugman and Risk Aversion

Robert Waldmann

Brad DeLong writes

I think the private-sector players in financial markets right now are highly risk averse–hence assets are undervalued from the perspective of a society or a government that is less risk averse. Paul judges that assets have low values beceuse they are unlikely to pay out much cash.

I have the same objection to this that I have to Brad’s analysis in general. He is explaining the market price but not explaining trading volume.

Trading volume in MBS based CDOs is very very low (markets have “frozen”). It is very difficult for this to happen at a price below the expected value of cash flows discounted at the safe rate of interest. For the private sector to hold all of the assets, it is necessary for the expected return to be high, maybe very high if the variance in returns is high.

However, for this to drive trading volume to essentially zero, it is necessary that no private sector entity wants to buy a little more of the asset at its current price. That means that the covariance of the return on the asset and the returns on every existing portfolio must be positive — otherwise a little bit more of the asset is a hedge and someone will buy a little bit more even at a return lower than the safe rate.

There are a lot of naked CDSs out there. They were cheap when bought, but are now expensive and, therefore, risky — if the economy makes it out of the crisis they will lose their value. If I owned a naked CDS on Mortgage based CDOs I would like to clothe it right now. I would certainly clothe it, that is buy the CDO, if the expected return on the CDO was above the safe interest rate.

If I owned as much of both CDS and CDO (a fully clothed or Burka portfolio) I would still be bearing counterparty risk on the CDO but just a bit of CDO (a G-string or not quite totally naked portfolio) would be nice.

Why are the people with naked CDSs buying G-strings ?