I often hear back-and-forth arguments from pro and anti stimulus folks about CBO type reports on jobs “created or saved” by the stimulus being vindications or just re-runs rather than tests of the same models from before it was implemented. Steve Landsburg reports here on a study which found aspects of a “natural experiment” in how the different states spent stimulus money, concluding that “The number of jobs created or saved by the spending is about 2.0 million as of March, but drops to near zero as of August.” So all of you still blathering about forecasts, shut up. Folks like Niklas Blanchard excepted since he is being inductive rather than relying on a detailed model of his own. There is also the continuance of an interesting discussion there on the death penalty and structural models vs empirical analyses using instrumental variables.