Lessons from the 20's

I just came across this snippet below, from Saj Karsan's excellent summary of the famous Security Analysis book, by Graham & Dodd. Apparently banks didn't learn the lessons from the 20's. I wasn't aware of these things either -- if only I had read this book more carefully when I first read it, a few years ago.

Saj writes (emphasis mine):
A popular type of guaranteed security is common in the real-estate mortgage market. A bank will sell investors a mortgage, and will guarantee payments on that mortgage. Unfortunately, for this type of self-guarantee to be worth anything, the following principles must be kept:

1) Loans must be conservatively financed
2) The guarantee must come from a company well-diversified

If loans are not conservative, then declines in real-estate values will result in deterioration in loan values. If the guarantor is not diversified, a general decline in real-estate values will serve to place the guarantor in receivership, which makes for a most dubious guarantee.

As simple as these principles are, in practice problems have arisen. In the roaring 20s, new and aggressive firms provided loans at levels so as to leave very little equity in the mortgaged property, despite the fact that appraisals were made at dubiously high levels. In order to compete with these firms, reputable ones would be forced to lower their standards. The industry spiraled out of control, and guarantees turned out to be useless in the ensuing carnage as firms were forced into receivership.
All this sounds exactly what happened recently. Last year, an ex-Lehman mortgage trader I know, talked about how investment banks had to lower their standards or lose business to more aggressive banks. The result we all know: rule number 1) above was violated and when RE prices declined (even though they never decline!), things went haywire. Many loan guarantors (AIG included) weren't well-diversified enough, violating rule 2).

All hail Ben Graham, the new Nostradamus.

No comments:

Post a Comment