Chad Stone is chief economist at the Center on Budget and Policy Priorities.
Everyone knows that the stimulus law didn't work, right? Except that it did.
Congressional Budget Office Director Doug Elmendorf told Congress recently, "Our position is that the  Recovery Act was not a failed program. Our position is that it created higher output and employment than would have occurred without it."
Of course, not all economists agree. But in a survey by the University of Chicago's Booth School of Business, 80 percent of the 40 or so economists surveyed agreed with the Congressional Budget Office, known as the CBO, that the unemployment rate was lower at the end of 2010 than it would have been without the stimulus law. The survey asked a second question about whether—accounting for future costs arising from financing the stimulus with debt—its benefits would end up exceeding its costs. Here, 46 percent thought that they would and another 27 percent were uncertain, leaving only a small percentage that did not.
The economists were also asked how confident they were in their answer. When the answers were weighted by each expert's confidence, only 4 percent of respondents believed the stimulus did not lower unemployment and only 14 percent believed that the benefits would not end up exceeding the costs.
The two charts below from the Center on Budget and Policy Priorities' chart book, "The Legacy of the Great Recession," illustrate the CBO's estimates of the success of the Recovery Act—officially, the American Recovery and Reinvestment Act—in keeping real (inflation-adjusted) gross domestic product (GDP) from falling more and the unemployment rate from rising more than they did.
The blue lines show the actual path of real GDP and unemployment, the red lines show the high estimate of how much worse things would have been without the stimulus, and the yellow lines show the low estimate. Clearly, the high and low estimates differ greatly, but that doesn't necessarily mean the truth lies somewhere in the middle.
The CBO says it chose those ranges "judgmentally to try to encompass most economists' views." The Budget Office's technical discussion says the upper portions are based mainly on one approach to combining economic theory and data to make an informed estimate and the lower portions are based mainly on another. Economists like me who judge the stimulus a success—including a large majority of the experts in the above survey—most likely would estimate the effects in the upper regions of CBO's ranges. As Paul Krugman points out, that's where the evidence leads you.
To be sure, the Recovery Act did not prevent a huge jobs deficit and output gap (the difference between what the economy produced and what it could have produced if there were not so much unemployment and idle productive capacity). But it prevented substantial economic waste by keeping output from falling more and unemployment from rising more than they did. It also reduced the economic hardship that comes with an economic slump. As the CBO observes:
Not only are the costs associated with the output gap immense, but they are also borne unevenly. Those costs fall disproportionately on people who lose their jobs, who are displaced from their homes, or who own businesses that fail.
Stimulus works to increase economic growth and employment when the economy is weak, as it is now. The CBO and almost all economists recognize that once the economy returns to full employment, deficit-financed spending does not work the same as in a weak economy and that continuing to run deficits will likely harm economic growth.
That's why most policy experts recommend that, once something happens to shake policymakers into cutting a meaningful budget deal, that deal should combine policies that support the flagging recovery in the short term with policies that promote longer-term fiscal stability—but that don't kick in until the economy is stronger.