Ratings QOTD

From Crash of the Titans, pp. 33-34:

The largest chunks of these [created by Merrill in the winter of 2006-2007] CDOs still carried triple-A ratings, at least in name, because the credit rating agencies hadn’t bothered to recalibrate their antiquated ratings models. But almost no one was willing to buy the triple-A portion of these bonds from Merrill because the rest of the marketplace knew what the credit rating agencies and [Osman] Semerci [then Head of Merrill’s FICC area] didn’t know: that the entire world of mortgages had turned into radioactive waste. [UPDATE: emphasis mine]

Two quick reactions:

  1. Sh*t, I knew that by late January of 2007, and I wasn’t being paid to know it.
  2. Note that this paragraph actually highlights an implicit disagreement that Robert and I have been arguing through on this blog for the past few years: whether ratings are a signal or the primary signal that investors use. Or, as Andrew Samwick—yes, this is my day for agreeing with conservatives (though not libertarians) on root-cause analysis—points out:

    I don’t think potential investors in U.S. Treasuries relied too much on its previous AAA rating in actively valuing the bonds and bills. And even if they did, they should be only minimally bothered by its current AA+ rating. Potential investors have plenty of public information on current and projected cash flows of the U.S. government. In those circumstances, there is little value added by a ratings agency’s grade.

When the market disagrees with the ratings, the ratings lose. So what has been happening today in the post-S&P bond market?