Last week, Christina Romer, chairwoman of the White House Counsel of Economic Advisers, suggested that we think of the $787 billion American Recovery and Reinvestment Act as an extremely expensive course of antibiotics. "Suppose you go to your doctor for a strep throat," Romer said in a speech to the Economic Club of Washington, "and he or she prescribes an antibiotic."

If your fever goes up after you take the first pill, just as unemployment rose after the stimulus bill was enacted, that doesn't mean "the medicine is useless," Romer noted. It could simply be that "the illness was more serious than you and the doctor thought."

But it's also possible that your sore throat and fever are caused by a virus, not a bacterium, in which case the antibiotic will not help. Eventually, though, you will recover on your own, and you may mistakenly conclude that your doctor's prescription did the trick.

Such erroneous causal inferences are always a hazard when it comes to government spending aimed at alleviating a recession. Even if most or all of the money is disbursed after the recession has ended (which is typically the case), stimulus advocates can say the recovery would have been weaker without the spending. Since there's no readily available parallel universe in which to test that counterfactual hypothesis, they can never be conclusively refuted.

Still, Romer seems unreasonably sure that Dr. Obama's medicine is already kicking in. Although she concedes that "the evidence from the path of the economy over time can't settle the issue of what the effects of the Recovery Act have been," her answer to the question posed in the title of her speech -- "Is It Working?" -- is "absolutely."

I guess that depends on how you define "working." No doubt spending billions of dollars in borrowed money has some impact on the economy. But the idea that the stimulus package played a major role in what looks like an incipient recovery that may have begun in June is belied by a couple of inconvenient facts. Continued...