Error: bad arguments

Dean Baker, director of the Center for Economic and Policy Research, scoffs at the idea that the U.S. might one day default on its debt:

Yes, with the unemployment rate about to hit 10 percent, Robert Samuelson yet again expresses concern over the country's biggest problem: the prospect of defaulting on its debt. Never mind that investors are prepared to hold U.S. government bonds for an interest rate of just 3.5 percent -- the lowest rate (except for earlier in this crisis) in almost 60 years. There is still nothing more important for Post readers to hear about than the risk that the U.S. government will default on its debt.

So basically, Baker is arguing that a U.S. sovereign default is likely because the Efficient Market Hypothesis is true. I.e., if bond markets today don't expect a default, then one won't happen, because markets are perfectly rational.

Am I really reading this? It's a little hard to believe my eyes, especially when Baker talks about the housing bubble in the same post. In 2007, markets didn't expect investment banks to default on the mortgage-backed CDOs they had issued. Then the defaults started, and expectations changed fast.

If our best argument that a U.S. sovereign default is nothing to worry about is that markets don't currently expect one, well, we're in trouble.