When the federal government, as part of the stimulus plan, says that it’s going to create or preserve 3.5 million jobs by the end of the year, or 600,000 jobs by the end of the summer, what are they talking about? How are they arriving at these figures?

It all comes down to formulas which the government has designed to estimate how much spending translates to how many new jobs. According to this model, job creation is basically a foregone conclusion: If you spend X, then you get Y jobs.

But the government spends money on job creation in more than just one way. According to a May 11 report (pdf) by the Council of Economic Advisors, there are three types of fiscal job stimulus. The report measures how much it costs to create one year of work for one person for each type:

• Government spending: For every \$92,136 of government spending, one year in one job is created for one person.
• Tax cuts: For every \$145,351 of government tax cuts, one year in one job is created for one person.
• State fiscal relief: For every \$116,603 in state fiscal relief, one year in one job is created for one person.

So the message here is that the cheapest way to create government jobs is through direct federal spending. We might consider taking this with an itsy bitsy grain of salt considering the fact that it’s the federal government that’s articulating these projections.  And the report itself notes that cheap is a pretty relative term.

“Since most workers earn much less than \$92,000 a year, the figure of \$92,000 per job-year may seem large…. A useful comparison is that the ratio of annual GDP to total employment in the economy is about \$105,000. Thus, a figure of \$92,000 per job-year is plausible and represents a very reasonable 'bang for the buck,’” it reads.

Now, we’re sure some very smart people have worked through these metrics, and we’re not saying these are disingenuous metrics (though other people have). But we would like to note just two things:

First, it would be nice if the experts at the Council of Economic Advisors would figure out a way to measure just how many jobs were really created after the fact. That way, the next time our economy tanks and we have to sink a trillion bucks into it, we’ll have a better idea of what we’re getting for our money.

Beyond that—and no doubt many will consider this to be a ridiculous proposition—why doesn’t the government just give the money to businesses that would like to expand, yet cannot afford the cost of new employees? The government could fund these new hires directly, until they start making money for the company.  The average cost per job is going to be less than \$92k per job, and we’d actually be able to objectively measure the impact of the plan.  Might that make sense, at least as a part of a stimulus plan?

At this point, the only concrete measure of jobs created by the stimulus are government jobs listed on USAJobs.gov. Currently there are just 61 government stimulus jobs listed on the site, down from 87 jobs when we last wrote about this.