That Warm And Very Very Fuzzy Stimulus Job Spending
Written by Michael Lewis on December 10, 2009
By Michael Lewis
If it was hard to get a grip on true jobs development at President Obama’s employment summit last week and in his job-creation plans this week, it’s understandable. Even in Miami, real impact is fuzzy.
You don’t have to look farther than our front-page exclusive last week, "County stimulus net: 49 jobs at $4.7 million each," shockingly inefficient federal stimulus use.
Is that the best bang we’re going to get for $787 billion stimulus bucks in the plan that Mr. Obama signed into law Feb. 17?
Historically, federal job-creation spending has been inefficient, but wasn’t this going to be different as the US funded only shovel-ready projects to add jobs fast? And will upcoming action fare any better?
Or are lofty early goals going to boil down to massive spending producing few jobs — because that’s what the county figures show: 49 new or saved jobs for $230 million.
With 11.2% Florida unemployment and 11.8% in Miami-Dade, wouldn’t we have been far ahead just handing 4,600 local jobless people $50,000 public service jobs at the same cost?
And is this vast discrepancy between what US aid could do and what it actually does to add jobs — a multiple of almost 100 possible good-paying jobs for each one actually created — a national issue or a local aberration?
Look at how the stimulus money actually is or isn’t stimulating Miami-Dade.
Last December, as it ramped up to seek federal funds, county hall planned to apply for $1.5 billion stimulus aid for 420 projects that were set to start immediately creating 23,110 jobs at an average cost of $64,907 each.
By January, as a stimulus neared enactment, the county had upped the ante to $7.2 billion, almost five times as much, for a bit more than half as many projects, 215, that would create 10,962 jobs at $645,814 each, ten times the prior month’s cost per job.
And by November, when County Manager George Burgess tallied figures, Washington had OK’d $230 million for 35 funded projects that had thus far created or saved 49 jobs at an average of $4.7 million each, more than seven times as much per job as January’s aim and more than 72 times December’s.
To be fair, not all 35 funded projects are active, Mr. Burgess reports. He says once all are rolling next year more jobs will be tallied, though he offers no numbers or timeline.
But remember, the concept was to stimulate jobs immediately, not in 18 or 20 months.
But even the four projects Mr. Burgess cites as now fueling jobs vastly exceed the cost of directly paying the unemployed. They have created or saved 11 county jobs and 38 private ones at a cost of $37,295,849, an average of $761,139 per job.
And more federal stimulus aid for the county was pending long before Mr. Obama revealed new plans this week.
The Burgess report showed 47 more county requests pending for an added $348 million.
Contemporaneously, a federal tally said the stimulus had saved or created 640,329 jobs on 56,986 projects nationwide through October. Of those projects, just 7% had been completed, 9% were more than half done and 46% less than half, with 38% yet to start.
Florida, fourth nationally in jobs saved or created with 4,099 programs funded at $6.8 billion in one federal tally and inexplicably 3,966 programs at $11.7 billion in a second concurrent federal count, totaled 29,322.78 jobs, a terribly exact figure in a terribly inexact process.
It’s inexact precisely because jobs created or saved are reported by agencies and organizations that get the money, with no federal audit.
That failure to check allows recipients to overstate impacts with impunity, either through ignorance or intentionally, making exact job figures like 640,329 or 29,322.78 or even 49 almost meaningless.
For example, compare the county’s stimulus spending with the local Miccosukee tribe’s. While the county was reporting it saved or added 49 jobs at $4.7 million each, the tribe was reporting it created 34.4 at only $604 apiece.
So is the tribe 7,781 times as efficient as the county?
To assess that, look at what the tribe says it did with the tiny $20,785 Department of Labor award, only $7,244 of which it had then received.
The tribe used the funds for youth job training and employment. It reported hiring 10 Native Americans seven hours a day for 10 weeks at $7.25 an hour at a reported total of $23,375 (though mathematically that totals $25,375).
The tribe reported also hiring or retaining a secretary, demonstrators, teachers’ assistants, temporary and fulltime tribal outreach workers, a restaurant busboy, a file clerk, a facility supervisor, a marketing supervisor and a security officer.
All 34.4 jobs, let’s repeat, for $20,785 in stimulus.
Across the nation such aberrations abound.
One Kentucky shoe-store owner who got an $889.60 contract to supply work boots to the Army Corps of Engineers reported he’d created or saved nine jobs with the funds — topping even the Miccosukees in efficiency. But upon outside questioning, he said what had actually happened is that he’d supplied nine pairs of boots, not created nine jobs.
An Oregon community action agency that reported it had used $397,761 to create or save 205 jobs had actually used the cash to give out raises.
Examples of double-counting jobs or counting part-time student jobs as full time abound, making national figures on stimulus use as questionable as our local ones. Without auditing, whatever people who get the money say they did with it seems to have been good enough for Washington.
The stimulus was intended not to bloat government but to add jobs in the private sector, which is where the economy grows. But the opposite seems true.
As the private sector shed 6.9 million workers from the recession’s outset into August, state and local governments added 110,000 jobs.
And the Burgess report on stimulus use notes that of the $230 million approved so far for Miami-Dade’s government, $210 million was to stay with the county and just $20 million pass on to community organizations and contractors.
Clearly, the private sector’s need triggered the stimulus. But who knew so much would fall off the truck en route?
Now, when last week’s jobs report is the strongest since the recession’s onset and as new jobless claims fall, the President plans more help — but the main impact of the existing stimulus still seems far ahead.
And whether stimulus spending now is $4.7 million per job through the county or $7.25 per hour for summer jobs through the Miccosukee Tribe, it’s still unclear if either is triggering the fast, efficient creation of jobs for which the program was intended.
Without a serious federal audit, moreover, we may never know. Perhaps the unstated federal plan is to keep it all warm and very, very fuzzy.