The graph above shows the relationship between sovereign debt levels, and S&P ratings. As Charles Franklin, the author of the post points out:
The most immediate and useful result is that there is very little relationship between S&P rating and the median debt level in each rating group. If debt were the primary element of the ratings, we would expect to see the red median markers move up from left to right. In fact there is almost no systematic pattern in the medians across groups.
Again, we have to live with the existence of the ratings agencies, and we have to contend with the power that they wield. But nothing justifies remaining silent when their ratings systems don’t make any sense whatsoever. There may be a political component to belittling S&P’s methodology–in the aftermath of their downgrade of our credit rating, we really don’t want their opinion to matter all that much, so that we can bring some calm to the markets, and so that we don’t have to pay higher interest rates–but that doesn’t mean that S&P’s methodology doesn’t leave a whole lot to be desired.
(Thanks to John Sides for the link.)