Network News

X My Profile
View More Activity

Stupid stimulus tricks

I like it when Jon Chait stops being polite and starts getting real. Lucky for me, that's pretty much all of the time:

In response to the Congressional Budget Office's calculation that the stimulus increased employment by somewhere between 1 and 2.1 million, Veronique de Rugy has made a chart to show that the stimulus has failed:


"On this chart, I can't see that the private sector has gained any jobs," she comments. "In fact, it's lost 8.7 million in two years." I assume she's just being dishonest here, because anybody with an intelligence level even approaching that of an apple pie would recognize that the claims that the stimulus increased employment are set against a presumed baseline of significant job loss. Her chart is sort of like saying that the people who told me last autumn that I should buy a winter coat are all wrong, because here I am wearing the coat and I'm still colder than I was in October without it.

Indeed, de Rugy's blue line of private sector job level shows that the rate of decline slowed, and then eventually stopped, beginning with the passage of the stimulus. You could say that all the macroeconomic forecasters are wrong and this is just a coincidence. But you can't say that this chart proves, or even suggests, the stimulus didn't work.

By Ezra Klein  |  February 24, 2010; 3:16 PM ET
Categories:  Stimulus  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Why does Rahm Emanuel want a budget wonk all to himself?
Next: A cartoon is worth a thousand words


This chart also shows that anyone who says that any job growth is in the bloated public sector are making that claim up.

Posted by: billkarwin | February 24, 2010 3:25 PM | Report abuse

Is this what you would call "conflict"? Because, if so, yes, there is time when "conflict" can be entertaining. Also, if you consider Obama's pantsing the entire House Republican Caucus all at once to be conflict, I found that entertaining as well.

Hey--maybe you media people are right. Conflict can be fun. When it involves more than just slapping lipstick on a pitbull, that is.

Posted by: slag | February 24, 2010 3:39 PM | Report abuse

This supports the reminder from another commenter on the post re: Democrats becoming Republicans. The GOP always aims to "create their own reality." It is certainly alarming when they ignore or misrepresent evidence. Here, there's no representation or second-hand information. They provide a chart, and then willfully (unless the comparison of their intellect to apple pastry is true) provide a false narrative about what the data mean. Outweighing the empirical contradiction, that narrative becomes better known as conventional wisdom -- the "reality" on which many people base their beliefs on the stimulus, health-care policy, tax policy, etc. I'm with Podesta, Lakoff, etc. -- narratives matter, and Democrats need to learn that.

Posted by: warnkea | February 24, 2010 3:56 PM | Report abuse

Here's a possible explanation: the anti-stimulus crowd doesn't believe that government jobs are real. They are effectively ignoring the government employment line.

Posted by: weiwentg | February 24, 2010 4:05 PM | Report abuse

Here's a possible explanation: the anti-stimulus crowd doesn't believe that government jobs are real. They aren't. All those people will be laidoff when the stimulus money runs out!

Posted by: obrier2 | February 24, 2010 4:26 PM | Report abuse

Are Chait and Klein both dim? Or both dishonest? Or both?

De Rugy did not state "that the stimulus has failed" or that "the stimulus didn't work", as Chait casts it. At least she didn't make such claims in the post that Chait criticizes (without being honest enough to link to it.)

All de Rugy said was that one can't say that jobs are "added" when one has nothing to tally but continuing losses.

In other words, either Klein and Chait can't read, or they're blinded by their partisan ignorance, or both.

Posted by: msoja | February 24, 2010 5:07 PM | Report abuse

Of course Chait does link to her post--judging from that and the rest of msoja's "analysis," I'd say that some wingers really are that dumb.

Posted by: steveandshelley | February 24, 2010 5:30 PM | Report abuse

And I thought Ezra's comment was going to be 'what stimulus?'. I don't see any government job growth there, just net government job layoffs from 2009 to 2008. Hard to argue government employment is crowding out private sector employment when government employment is falling.

Posted by: stefan5 | February 24, 2010 5:33 PM | Report abuse

--"Of course Chait does link to her post"--

You are mistaken, or lying. Her post is on the National Review site, and there is no link to it from either Klein or Chait.

Chait's post includes links to the CBO, The New Republic, The New York Times, and, but does not link to the post he uses to build his strawman.

Posted by: msoja | February 24, 2010 5:46 PM | Report abuse

Agreed her chart can't prove a counterfactual.

However, the macroeconomic forecasters are probably wrong, at least in terms of magnitude and the slowing of job losses are coincidental (note that if 1.5 million jobs were created over the past year, that's 150,000 a month, suggesting the baseline trend was going to improve anyway).

What suggests the forecasters are wrong? Well, the large range of possible multipliers is a good place to start. Lots of multipliers have high estimate/low estimate variances of a factor of 2 or more. It's hard to be confident in the results of a model that suggests you might get $1 of GDP growth for $1 of government spending, or $2.5 of GDP growth for $1 of government spending, depending on calibration.

The main issue is the omitted variable problem - modelers can't possibly include every variable known to influence the macroeconomy, and so they have to pick. The variables picked often say more about the modeler's priors than anything else.

My model would include a central bank reaction function. The Fed is known to target a preferred rate of inflation. Back in January 2009, the Fed was continuing to target a low rate of inflation (roughly 2), for the next twelve months - we know the Fed is still targeting inflation because Ben Bernanke & Co strongly reject calls for an inflation target of even 3%/yr. The Fed would then see fiscal stimulus coming down the pike, expected to increase aggregate demand, output, prices and employment. An inflation targeting central bank then runs a tighter monetary policy than it otherwise would, so that aggregate demand does not rise.

As Ezra has said before, the stimulus spending did help people - things like COBRA and unemployment benefit extensions made the downturn more bearable. I'm far less convinced that it created anywhere near as many jobs as the models claim.

Posted by: justin84 | February 24, 2010 7:06 PM | Report abuse

One of the worst two-axis charts i've seen. The distribution on the left axis is 16 million jobs, or about 15% of the maximum value on the scale; on the right-hand axis the spread is just 400,000, or less than 2% of the maximum value on the scale.

Posted by: tomwoods | February 25, 2010 8:51 AM | Report abuse

So can we all agree that the stimulus did not create enough jobs to offset the recession. We can all agree that the stimulus failed to get unemployment significantly below 10%. I agree that's a different question that the vague, "did it work"? However, employment is certainly quite far below the projections in the models used to support its adoption.

Posted by: staticvars | February 25, 2010 1:07 PM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company