The Obama Administration's Convenient Counterfactuals

The latest report on the effects of the $800 billion dollar stimulus from the Obama administration's Council of Economic Advisers notes that "evaluating the impact of countercyclical macroeconomic policy is inherently difficult because we do not observe what would have happened to the economy in the absence of policy." Another thing that makes it difficult? Neither of the two primary methods the administration uses to evaluate the stimulus's effects truly observe the policy that was put in place.

Instead, the administration's reports, like the reports put out by the Congressional Budget Office, rely on computer models and comparisons to statistical forecasting of what might have been. In the case of forecasting, the administration compares current GDP and jobs numbers to a non-stimulus baseline—what the administration calls a "sensible statistical forecast" of what would have happened had the stimulus not passed. The report's authors themselves decline to rely too heavily on the comparison, noting that "the estimates from a forecast approach have considerable margins of error." Can you know how far off you are from an alternate-history counterfactual that you can never prove?

The models are just as unhelpful. The mathematical simulations used to prove that the stimulus worked—creating between 2.4 million jobs and boosting GDP by somewhere between 2.3 percent and 3.2 percent—are functionally the same models used before the stimulus to predict that it would work. So it's not much of a surprise to find that rerunning the same model that predicted the stimulus would create jobs and expand the economy results in a report that, yes, the stimulus created jobs and expanded the economy.

The report explicitly compares the administration's approach to the one used by the Congressional Budget Office in its stimulus reporting. But the CBO's mandatory reports aren't much of a guide to the effects of the stimulus either. The only difference is that the CBO has been perfectly willing to admit this.

At a Q&A in March of 2010, CBO director Douglas Elmendorf described the CBO's approach "repeating the same exercises we did [prior to the stimulus's passage] rather than an independent check on it." At the same event, he was asked, "If the stimulus bill did not do what it was originally forecast to do, then that would not have been detected by the subsequent analysis, right?" Elmendorf responded: "That's right. That's right."

The CEA report declares that its existence is due to "the unprecedented accountability and transparency provisions" included the stimulus bill. But far from a transparent accounting of what we do and don't know about about the stimulus and its economic effects, the administration is content to gin up a convenient story about what might have happened in the absence of the stimulus and then claim that, thanks to its passage, we're better off.