The Wall Street Journal pokes holes on the White House’s allegation of the stimulus having created “three million jobs” (hint: the key is in the saved or created part)
Three Million Imaginary Jobs
The White House says the stimulus worked beyond even its hopes. Seriously. (emphasis added)
Christina Romer went so far as to claim that the 3.5 million new jobs that she promised while the stimulus was being debated in Congress will arrive “two quarters earlier than anticipated.” Yup, the official White House line is that the plan is working better than even they had hoped.
We almost feel sorry for Ms. Romer having to make this argument given that since February 2009 the U.S. economy has lost a net 2.35 million jobs. Using the White House “created or saved” measure means that even if there were only three million Americans left with jobs today, the White House could claim that every one was saved by the stimulus.
The White House also naturally insists that things would be much worse without the stimulus billions spent on the likes of Medicaid payments, high speed rail projects, unemployment benefits and windmills. Mr. Obama said recently in Racine, Wisconsin that the economy “would have been a lot worse” and the unemployment rate would have gone to “12 or 13, or 15 [percent]” if government hadn’t spent all of that money.
This is called a counterfactual: a what would have happened scenario that can’t be refuted. What we do know is what White House economists at the time said would happen if the stimulus didn’t pass. They said the unemployment rate would peak at 9% without the stimulus (there’s your counterfactual) and that with the stimulus the rate would stay at 8% or below. (See the nearby chart.) In other words, today there are 700,000 fewer jobs than Ms. Romer predicted we would have if we had done nothing at all. If this is a job creation success, what does failure look like?
All of these White House jobs estimates are based on the increasingly discredited Keynesian spending “multiplier,” which according to White House economist Larry Summers means that every $1 of government spending will yield roughly $1.50 in higher GDP. Ms. Romer thus plugs her spending data into the Keynesian computer models and, presto, out come 2.5 million to 3.6 million jobs, even if the real economy has lost jobs. To adapt Groucho Marx: Who are you going to believe, the White House computer models, or your own eyes?
Or, as Milton Friedman used to say, “there’s no such thing as a free lunch.” The money government spends does create some jobs—the folks working on road projects, say—but that money has to come from somewhere, which means taxing or borrowing it from areas of the private economy that are nearly always more productive. This doesn’t mean that government spending is always a bad idea, but it does mean that government spending as economic stimulus is fanciful.
Harvard economist Robert Barro first blew apart Keynesian assumptions with his famous 1974 essay, “Are Government Bonds Net Wealth?” He and Charles J. Redlick, also of Harvard, recently updated this demolition in a new study for the Mercatus Center examining 50 years of defense spending in various countries. They find a multiplier effect of between 0.4 and 0.7. This means that government spending shrinks the private economy, because it “crowds out other components of GDP, particularly investment.”
This would certainly explain better than Ms. Romer’s computer models why a nearly $1 trillion stimulus has been followed by a mediocre economic recovery, a 9.5% unemployment rate, and almost no net new private job creation.
The rest of us call it “pulling numbers out of
their asses thin air.”