[A]fter August 2, Treasury can simply carry on with business as usual. Money will come in to its bank account; money will go out. And the balance will dip below zero: Treasury will have an overdraft at the Fed. . . . There’s no reason why this state of affairs couldn’t continue for months. Treasury would continue to spend money, as instructed by Congress in the budget, and Treasury’s overdraft at the Fed would continue to rise. The Fed, for its part, would have two choices when it came to cashing Treasury’s checks: it could either simply print the money, or else it could sell some of its assets — it owns $1.6 trillion in Treasury bonds — and use those proceeds instead. Either way, any bank presenting a check from Treasury could cash it, no problem.What would be the macroeconomic implications of this? Well, I know what everyone is going to say they are (Republicans: "Disaster! You can't just print money!" Democrats: "Eh."), but I myself haven't the foggiest idea. For his part, Felix seems quite excited about the prospect, suggesting it might be the "first-best solution" to the debt ceiling problem.
And would a Fed overdraft be strictly legal? Once again, I don't know. But frankly, I don't think it matters terribly much either. More than anything, the fact this idea is getting batted around indicates the inherent flimsiness of arbitrary legal rules, like the debt ceiling, when confronted with real, tangible disaster. The people in charge know something's got to give, and very few would choose to preserve a set of legal abstractions instead of the US economy. Cue the hunt for loopholes -- not even necessarily real loopholes, but just anything with a sufficient veneer of legality. The administration might not choose to overdraft at the Fed, but if Congress can't do its job, it'll find a similar escape hatch somewhere.