More than four years have passed since President Obama signed into law the American Recovery and Reinvestment Act (ARRA) of 2009. The stimulus package came at a cost of more than $830 billion, or about 22 percent of annual federal spending. It was supported by many Keynesian economists although criticized for being ‘too small’.
Although it seems unpopular to look back and ask whether policies of the past have actually worked, let us have look at the effects of the ARRA.
In a 2010 study, John Cogan and John Taylor have found little evidence in favor of the stimulus package:
The implication is not that ARRA has been too small, but rather that it failed to increase government consumption expenditures and infrastructure spending as many had predicted from such a large package. A consideration of the counterfactual event that there had not been an ARRA supports the hypothesis that state and local government borrowing would have been higher and purchases would have been about the same in the absence of ARRA.
Atif Mian and Amir Sufi have published another study in the prestigious Quarterly Journal of Economics in 2012. They estimate the effect of the Cash for Clunkers Program and exploit variation across U.S. cities in exposure to the program. They conclude:
We find no evidence of an effect on employment, house prices, or household default rates in cities with higher exposure to the program.
James Feyrer and Bruce Sacerdote also examine the effect of the ARRA in their 2011 NBER Working Paper. They use state and county level variation and find that each job created came at a humongous cost:
A cross state analysis suggests that one additional job was created by each $170,000 in stimulus spending. Time series analysis at the state level suggests a smaller response with a per job cost of about $400,000.
This finding is in line with a paper by Daniel Wilson published in the American Economic Journal in 2012:
Cross-state IV results indicate that ARRA spending in its first year yielded about eight jobs per million dollars spent, or $125,000 per job.
It is also in line with much of the previous research, as documented by Valerie Ramey in a meta study published in the Journal of Economic Literature in 2011:
I assess the likely range of multiplier values for the experiment most relevant to the stimulus package debate: a temporary, deficit-financed increase in government purchases. I conclude that the multiplier for this type of spending is probably between 0.8 and 1.5.
This would translate into something like $120’000 to $200’000 per job created.
By the way, these figures are not surprising since most economists estimated the ARRA to ‘save’ about three million jobs at most. At a cost of $831 billion, this suggests about $300’000 per job. I wonder whether the public support would have been large had Obama sold the program with this price tag.
Greg Mankiw has also been following the so-called recovery ever since the financial crisis. Instead of focusing on unemployment his focus is on the share of the population which is working. This takes into account that some people have simply stopped looking for work. During Bush’s presidency, the employment-population ratio increased slowly from 62 to 63 percent. In the recession of 2008-09, this ratio dropped to 58.5 percent. And ever since then, it has stagnated at this low level. (see figure here)
Finally, it is also worthwhile to reconsider Econ Stories’ 2011 video featuring Keynes and Hayek:
Spending is not free, that is the heart of the matter. Too much is wasted as cronies get fatter. The economy is not a car, there is no engine to start. No expert can fix it, there is no ‘it’ at all.
The question, I ponder, is who plans for who. Do I plan for myself or do I leave it to you? I want plans by the many, not by the few.
Swiss Economist 