$2 Million stimulus dollars per job.

Via Len Penzo we learn of a Michigan factory that's received $300mm in subsidies and added 150 jobs. That's $2mm per job. A news release implies, but does not state, that this money is also paying for a second a plant, which will add another 300 jobs. It states that the first plant will fully employ 320, and makes further comments on increased staffing.

For the sake of argument, let's pretend that the multiplier doesn't exist and ignore other side effects (i.e. American competitiveness in high tech sectors). Mr. Penzo is correct that $2,000,000 per job is probably not the most efficient use of the Government's finite resources. What if the final result is $300mm for 620 permanent jobs (ignoring temporary work like construction and surveying).  Now we are down to well under $500,000 per job, and it is much more ambiguous. If each job lasts 10 years and pays workers (with bens) an average of $80,000 a year, we are at $5mm in salaries. This looks like a good deal depending on your discount rate.

Now, I don't know if this piece of stimulus was a good deal or not. My assumptions are probably wide off the mark and we don't know if the extra 320 jobs would appear nor if they would have without the stimulus money. The general point is that when attacking spending per job created, it would be nice to have an idea of what the baseline for a good project is.