The ratings failed because the mathematical foundations of the system failed. Portfolio theory is not a panacea. Government economists would have used the same math to arrive at the same results.

Really? Remember this?

From an April 2007 instant-message exchange between two S&P employees:

Rahul Dilip Shah: 'btw -- that deal is ridiculous'

Shannon Mooney: 'I know right ... model def does not capture half of the risk'

Shah: 'we should not be rating it'

Mooney: 'it could be structured by cows and we would rate it'

Shah: 'but there's a lot of risk associated with it -- I personally do not feel comfy signing off as a committee member'

That suggests that the math didn't mislead the ratings agencies -- they ignored the math in order to make more money.