1) Normally in underwriting the bank is taking a hopefully small risk that between closing the price and actually selling on the securities the market does not change too substantially. It does and has happened that the underwriting banks lose substantial amounts of money even in vanilla equity underwritings. I also disagree that there are terribly many reputational issues with bringing crap to the market, as long as it is crap du jour such as Internet stocks or dare I say it CDOs and CLOs.

2) With the original CLO, I suspect GS bought the loans off some clients packaged them up and sold them to other clients. Yes for a period of time those loans were warehoused but I highly doubt GS bought the loans because they thought qua loans there were a great investment they wanted to take a view on, it was just balance sheet rental for another client or clients. Classic middleman. Just like the pharmacist bought the box of Preparation H, held it for a short time till another client – you – came in and paid a slightly higher price. It was hardly a principal investment by the pharmacist.

3) I also suspect that in buying back the AAA tranche that it was not quite as arms length as you suggest. I suspect GS wanted to call, needed to buy out the AAA tranche and to make sure it seemed pucker got a third-party to run an auction which, within reason, GS would always have won. The third-party and the auction was so GS and the client could wave a piece of paper around saying look it was kosher.

The only vaguely nefarious way this can be viewed is possibly the reality was that GS in 2008 was trying to shrink its balance sheet, offered a meaty coupon to clients with the nod-nod-wink-wink we will call the notes when all this blows over. Of course the journalist in this case didn’t bother chasing up the interesting bit because he is more concerned in the moral outrage that a firm isn’t aiming to provide a service at a loss – something the NYT has been doing well for a while now.

Goldman Sachs had no client in the CELF trades, from start to finish, other than itself.

http://epicureandealmaker.blogspot.com/2 010/11/client-is-not-counterparty.html

As best I can tell right now, the LEAST risky securities over a ten year horizon are top-quality dividend yielding stocks. This isn’t to say that they are any less risky than usual, just that the normally risk-averse investments are even worse.

Disclaimer: presently allocated ~75% stocks (predominantly blue chip dividend stocks), ~15% TIPS and guaranteed-principal annuities, ~10% cash/CDs with 2 year maturity or less. Conventional bonds are very scary to me at current prices.

Incidentally, does this make other people as nervous as it makes me? I’ve been telling my mother-in-law that if she’s as risk-averse as I think she is, a 1% return — or frankly a 0% return — really isn’t so bad right now, and that the mindset that one is “entitled” (or something) to more is likely to cause pain at some point.