Not surprisingly, stimulus advocates are already accusing me of miscalculations in my NY Post article. Mistaken criticisms include:
"The article has 2005 dollars to make the recession look little." That's incorrect, I used the same dollars to measure the recession as I used to measure the stimulus bill: 2009 dollars.
"The recession should be measured relative to an upward trend." For the purpose of a cost-benefit analysis of the stimulus package, that's wrong. It should be measured relative to what real GDP path the stimulus package might achieve. I have been generous in taking that to be real GDP constant (rather than falling, as it would have even with a brilliant stimulus in place).
"The stimulus package is cheap when measured per job year". The stimulus is $215,000 per job promised. Now if we took the people who supposedly would have gotten those jobs and asked how much sooner they were employed by the stimulus (as opposed to when they would have found a job on their own), it would be generous to the stimulus to say that it was six months. So that makes it $430,000 per job-year. And I am not counting the job-years inevitably lost by the stimulus when the stimulus projects run out and their employees have to go find something else. Nor am I counting the fact the law will fall far short of the 3-4 million jobs promised -- do these things and results will look more like $1 million per job. (Yes I am aware that the Obama people claim that workers getting the stimulus jobs would have remained unemployed for years without the stimulus, but that's ludicrous ... millions of people not getting employed by the stimulus will manage to find jobs more quickly than two years).
If you would like to see a spreadsheet with my calculations, click here.