Killing the Lie about 8% Unemployment

My overall goal in this blog is to encourage people to expect their leaders to tell them the truth. If they lie, they should be held accountable at the polls. But when they repeat the lie over and over, it takes real effort to provide the rebuttal. The lie is quick and catchy. The reality is nuanced and has to be explained. But it's worth the effort.

Yesterday in Cincinnati, Mitt Romney said the following: “The president said that if we let him borrow $787 billion for a stimulus, he'd keep unemployment below 8 percent nationally. We've now gone 40 straight months with unemployment above 8 percent.” This talking point has been called out by fact-checking sites so many times, it's simply amazing that it's still there. Glenn Kessler of the Post has given it multiple pinochios, Politifact has rated it "mostly false" every time it's been reported. 

The general explanation goes like this: In January of 2009, Christina Romer (chair-designate of the council of economic advisors) wrote a report outlining why a stimulus was needed. In the report is a now infamous graph showing what would happen if there was no stimulus and what would happen if there was one. 

Critics have repeatedly jumped on the fact that the curve peaks in March 2009 (end of the first quarter) at 7.9% This, they say, is the president's promise. It sometimes gets characterized as a campaign promise, which is wrong on two levels -- first, he didn't say it and second, this report came in two months after the election.

The fact-checking folks have observed that while Romer did put this in the report, she also had the appropriate "best guess" disclaimers. As Kessler observes, the first page of text in the report (which is page 2) closes with the following paragraph:

"It should be understood that all of the estimates presented in this memo are subject to significant 

margins of error.  There is the obvious uncertainty that comes from modeling a hypothetical 

package rather than the final legislation passed by the Congress.  But, there is the more fundamental 

uncertainty that comes with any estimate of the effects of a program.  Our estimates of economic 

relationships and rules of thumb are derived from historical experience and so will not apply exactly 

in any given episode.  Furthermore, the uncertainty is surely higher than normal now because the 

current recession is unusual both in its fundamental causes and its severity."

Clearly, even with the disclaimers, Romer's report was overly glowing. And from a purely political standpoint, it's awful to have something so easily quoted lying around. 

But I'm struck with an even more intriguing critique of the Romer argument that further underscores the deception of the Romney campaign and his surrogates. Romer's projection is not simply rosy with regard to the impact of a stimulus -- its rosy in terms of No Stimulus. The very chart that critiques delight in quoting misses the top end of unemployment by 1.2% (10.2 vs. 9.0). As others have explained, the data she was using was from the middle of 2008. By early 2009, things were worse than those previous projections would suggest.

I charted the data above against the actual unemployment rate for the same period. Here's what that shows:

What's striking is that the slope on the actual unemployment rate between 2008 and 2009 was far more aggressive that previous data would suggest. How do we add that reality to Romer's disclaimers? One quick way to do this is to imagine adjusting the bottom two curves upward so that the green line follows the blue one. To do that, I just added 1% to both curves. Here's how that comes out.

Even with this adjustment, Romer's prediction is still too rosy. There are two other pieces to understanding why the unemployment rate didn't fall as fast as she expected. First, there was much of the stimulus bill that got structured as tax cuts which helped it get passed by Congress. But in that very first page of the report critics like to quote, Romer says the following:

"Tax cuts, especially temporary ones, and fiscal relief to the states are likely to create 

fewer jobs than direct increases in government purchases.  However, because there is a 

limit on how much government investment can be carried out efficiently in a short time 

frame, and because tax cuts and state relief can be implemented quickly, they are crucial 

elements of any package aimed at easing economic distress quickly."

So the stimulus wasn't as stimulative as it might have been. Here's the second piece, two bullet points down:

"More than 90 percent of the jobs created are likely to be in the private sector.  Many of 

the government jobs are likely to be professionals whose jobs are saved from state and 

local budget cuts by state fiscal relief."

As I pointed out in my post from last week, government jobs have not been held safe. We have shed them at an unprecedented pace. What happens if you put those back in? Estimates are that it lowers the unemployment rate by about 1%. Here's how Talking Points Memo summarized it.

Can you see where the bottom right side of the TPM chart ends? Right where my adjusted Romer chart would come in.

So let's review: 1) the president never promised 8% unemployment; 2) the estimate was based on incomplete data that underestimated the size of the problem (which was acknowledged at the time); 3) tax cuts aren't as stimulative as other projects; 4) loss of government workers masks other improvements.

This claim is not just technically false as the fact-checkers point out. It is demonstrably false. It should not be repeated by anyone who has a commitment to truth-telling.