When President Barack Obama signed the American Recovery and Reinvestment Act, he promised that the $787 billion stimulus package would create or save 3.5 million jobs over two years, mostly in the private sector.
Without the spending called for in the law, the administration predicted, the unemployment rate would reach 8.6 percent. Today, the U.S. unemployment rate is 9.6 percent. What happened?
We spent a great deal of money. But if we wanted to spend it to benefit the states with the highest unemployment rates, we failed. Out of the total $787 billion, the federal government so far has allocated one-third, or $275 billion, in grants and contracts to shovel-ready projects. So far $190 billion of that amount has been spent, according to government figures.
With a few exceptions, the data show little correlation between the level of unemployment and stimulus spending. In fact, the opposite is true. The federal government has given far fewer stimulus dollars to states with high unemployment than it has to states with low unemployment.
Here is a chart that compares the five states with the highest and lowest unemployment rates and the amount of stimulus money per capita:
State Unemployment Rate Stimulus/capita Nevada 14.3 percent $561.55 Michigan 13.1 percent $648.91 California 12.3 percent $546.34 Rhode Island 11.9 percent $164.83 Florida 11.5 percent $475.67 Vermont 6.0 percent $522.42 New Hampshire 5.8 percent $852.53 Nebraska 4.7 percent $591.17 South Dakota 4.4 percent $1,084.73 North Dakota 3.6 percent $1,059.95
Does it make sense that the state with the highest unemployment rate, Nevada, is getting roughly half the per- capita amount of the state with the lowest, North Dakota? Sure, Michigan, Nevada and California are all getting more per person than New Hampshire, but only a smidge more, even though their unemployment rates are at least twice that of New Hampshire’s.
What’s going on here? For one thing, these are raw numbers that reflect only the amount of stimulus money spent. What they don’t do is tell us anything about the reasons it was spent.
In order to find what forces motivated the decision to spend, I ran a regression analysis that tested whether a certain factor is correlated to the outcome -- in this case, whether the unemployment level helped determine the spending allocations.
In this case, though, even after running a series of analyses, I found no correlation between unemployment levels and stimulus spending. (The data and regressions can be downloaded at mercatus.org.)
What about the possibility that these numbers don’t take into consideration money sloshing around federal agencies that will eventually go to the states? After all, that’s why data available on recovery.gov, which tracks stimulus spending, show that the District of Columbia receives a whopping $5,078.78 per person. That’s money that has been allocated to federal agencies, but some of it hasn’t yet gone to the states.
It doesn’t, however, explain why Alaska, with 7.7 percent unemployment, is getting $2,315.88 per person. Of course, Alaska always has been the Bermuda Triangle of federal spending -- some mysteries are eternal.
So if state unemployment levels weren’t the basis on which the federal government allocated these funds, what was? To me, it looks like it was just one: speed.
Back in February 2009, for all the talk about creating jobs, the administration wasn’t focused on distributing money to high-unemployment states, which, in theory, were the ones hurting the most. It was just trying to spend a massive amount of money as quickly as possible.
To achieve that, the stimulus bill distributed money among the states through existing channels -- such as the federal Departments of Education and Transportation -- whose main functions aren’t to address unemployment levels.
The Obama administration was wildly successful if its objective was to spend a lot of money in a short amount of time. Whether that money has done or will do anything for the people that need it most has proven far more elusive. As the saying goes, You can have it fast, you can have it good, or you can have it cheap -- pick two.
Sadly, we did worse than that: We only got one.
(Veronique de Rugy is a senior research fellow at the Mercatus Center at George Mason University. The opinions expressed are her own.)
To contact the author of this column: Veronique de Rugy at firstname.lastname@example.org