Karl Rove, former Deputy Chief of Staff and Senior Policy Advisor to U.S. President George W. Bush, walks on the floor at the Republican National Convention on August 28, 2012 in Tampa, Florida.
I’m already weary of debunking Republican talking points about President Obama’s stimulus. Now Karl Rove’s SuperPAC has poured $11 million into an ad rehashing the GOP’s favorite talking point about the stimulus, that it was a failure because a report from Obama’s transition team suggested it would keep unemployment below 8% and drive it down to 5.6%. Here’s the real story behind that misconceived report—and that misleading ad.
As I explain in my book, The New New Deal, the report originated with Rahm Emanuel, Obama’s first chief of staff, who wanted an official-looking document full of jobs predictions to help him sell the stimulus on Capitol Hill. The economy was losing 800,000 jobs a month and shrinking at a Depression-level -8.9% annual rate, although no one knew it was that awful at the time; back then, the official GDP numbers were a much less scary -3.8%, before unprecedented revisions in the coming months. Anyway, Rahm wanted to show wavering congressmen that emergency action would produce results and create jobs.
The report was assigned to Christina Romer, who would lead Obama’s Council of Economic Advisers, and Jared Bernstein, Vice President Biden’s chief economist. Romer and Bernstein spent most of their time crunching numbers to estimate how much an $800 billion stimulus would increase employment compared to the no-action baseline, calculating the “delta” between what would happen with and without stimulus. They spent less time assessing that no-action baseline, mostly following forecasts from the Fed and two private firms. If anything, their no-action baseline was somewhat more negative than the prevailing economic consensus.
It wasn’t nearly negative enough. Page 4 of the report included the chart that has dogged Obama ever since, the chart that Rove’s American Crossroads replicates in its new ad. The chart suggested that unemployment would peak at 9% without the stimulus, but would remain below 8% with the stimulus. In fact, the jobless rate shot past 8% before the stimulus even kicked into gear, which obviously was no reflection on the stimulus. It was a reflection on the baseline. The economy was in far worse shape than Romer and Bernstein had realized.
The Romer-Bernstein report—just 13 pages, barely half of that text—was cluttered with caveats. “It should be understood that all the estimates presented in this memo are subject to significant margins of error.” And: “The uncertainty is surely higher than normal now because the current recession is unusual both in its fundamental causes and its severity.” But nobody remembers caveats. It even included a humdinger of a footnote about the baseline: “Some private forecasters anticipate unemployment rates as high as 11%.” But nobody reads footnotes.
Before releasing the report, Romer shared it with Obama’s economic team, as well as some academic peers. The only reader who raised concerns about including the unemployment prediction was her husband, another Berkeley macroeconomist. He reminded her that a fellow economist who had muffed an employment forecast based on an overly optimistic baseline once told them ruefully that forecasters should stick to predicting changes in unemployment, not levels of unemployment.
“Oh, for heaven’s sakes,” Romer replied. “If unemployment goes to 10%, we’ll have much bigger problems than this.” She was right about that. But Romer and Bernstein both say that no one inside the White House tried to block or even amend the report. Obama’s political team did exchange emails about the potential damage from the report’s (correct) suggestion that men would get more stimulus jobs than women, but no one seemed worried about putting the rosy employment forecast in writing.
“Rahm loved it,” Bernstein told me. “He said: Do more stuff like that!”
In fact, most analysts believe Romer and Bernstein did a good job with the delta, their prediction of how much the stimulus would improve the economy. The consensus from independent forecasters is that the stimulus added about 2 to 4 percentage points to GDP, the difference between contraction and growth. Romer and Bernstein simply blew the baseline, the economy’s starting point. They worried at the time that they were being too optimistic, but they figured they would have sounded like Chicken Little if they had strayed much further from the consensus of their peers.
“The fact remains that 8% was a conservative forecast at the time,” Bernstein says. “In hindsight, obviously, we just should’ve focused on the delta.”
Which brings me to that American Crossroads “Actually Happened” ad savaging Obama for the inaccurate prediction. Check out the blue line, the Romer-Bernstein forecast for unemployment in the Obama era. And then check out the orange line, the actual trend of unemployment in the Obama era. The pleasant-looking narrator correctly points out that the blue line is a lot lower than the orange line. But they both slope downwards! Karl Rove is pouring $11 million into graphic evidence that the stimulus worked. And it did work; the employment numbers in the second quarter of 2009 were the biggest improvement in 30 years.
Of course, “we got the delta right” is not the kind of slogan you want on a bumper sticker. The Romer-Bernstein report was a political disaster, gifting the Republicans an enduring talking point. But it correctly predicted that the stimulus would point us in the right direction. It just misunderstood how far we had to go to reach our destination. We haven’t gotten where the Obama team had hoped to be, but we’re way better off than we were four years ago, when the economy was in free-fall.
“If I were doing this over again, I’d be more politically astute,” Romer says. “But those were our best estimates at the time. We weren’t trying to puff up confidence or do the reverse. We were trying to do an honest forecast.”