The most interesting thing about yesterday's market action was the behavior of the Treasury market:
If the market collapse was really about investors reacting negatively to the United States' new, lower credit rating, why on earth would replace their stocks with a direct investment in that very same government? We can only know as much as the markets reveal, and here are the facts: (1) People sold stocks in a panic. (2) People bought Treasuries in a frenzy. This does not seem like behavior consistent with a lack of confidence in the United States.
It is, however, a paradoxical consequence of modern portfolio theory. Let's say the universe consists of two possible investments, A and B. A is a very low-risk investment; B carries higher risk. Investors allocate their funds among A and B, creating a portfolio with some desired expected risk and return characteristics. Now, news arrives that A actually carries slightly more risk than investors had previously anticipated. The rational reaction is to sell B and buy more A. Why? Because when A was "downgraded," the investment portfolio became riskier. A is still less risky than B, so the only way to restore the portfolio to its previous risk levels is to increase the allocation to A.
But there is a problem with this model: A's price will fall on the news, ceteris peribus, further compounding the allocation issue. If it falls sufficiently, the buying activity will be fierce enough to drive it even higher than it was originally. But in that case, the expected return on A will be necessarily lower than it was originally, and the risk of holding it even higher than at the time of the downgrade. Investors may end up worse off as a result.