Policy Analysis Nihilism
By Fester:
The future is unknown because it is the the future. All that is in the future is a collapsing profile of probabilities that unify as the near future becomes the now and the not so near future becomes a little bit closer to being the near future. However those probabilities and potentialities can be assessed and judgements can be made from proposed policy interactions.
If you are doing policy formation and or policy analysis for a living, there should be a few phrases that haunt your dreams. One of those phrases is the following:
WHAT IS THE COUNTERFACTUAL?
As soon as you identify the counterfactual, you can start asking about effectiveness and impacts of a given proposal. Not asking or disregarding the usefullness of counterfactualism and baseline assumptions results in policy analysis nihilism. Nothing is knowable, so fuck it.
And that is what Greg Mankiw is attempting to do with a recent post, and plenty of conservatives including those who should know better are following along:
This is hackery as Mankiw's tap dancing addendum indicates. If you have been following American politics and the economy at all, the baseline case is the do-nothing case. In that scenario, it is projected that the American economy will lose millions of jobs in a prolonged recession. The stimulus can and should be judged against this type of counterfactual.
The Congressional Budget Office, whose numbers are jibing very closely with the Obama Administration numbers provide their estimate of the counter-factual unemployment numbers. Their baseline has unemployment hitting 9.0% this Christmas and about 141 million jobs. There are other macro forecasts out there that are projecting roughly the same or worse numbers in the 'do-nothing' baseline case.
So here is an easy metric to see if the stimulus is working --- is the U-3 unemployment rate in December beneath 8.5%? If it is, then the stimulus is most likely working? Another metric is the numebr of jobs in the economy? Are there at least 142,400,000 jobs? Those are the low case numbers.
Now onto the 'create or save' metric that Mankiw is complaining about, by 2011 when most of the stimulus impact will have hit the economy, are there roughly four million more jobs in reality than measured against the counterfactual baseline as defined by the CBO?
Those are real and answerable questions. They are basic policy analytical questions and techniques that someone like Mankiw should be able to do in his sleep, or more likely assign one of his grad students to do in lieu of sleep.
The next line of objection is to the validity of the models and their underlying assumptions, but Mankiw is willing to use long term economic models in the form of large error band 75 year acturial projections to repeatedly argue for his favored economic policy proposal of gutting Social Security, so I don't think Mankiw is a policy analysis nihilist when the economic models support his preferences.
As soon as you identify the counterfactual, you can start asking about effectiveness and impacts of a given proposal. Not asking or disregarding the usefullness of counterfactualism and baseline assumptions results in policy analysis nihilism. Nothing is knowable, so fuck it.
And that is what Greg Mankiw is attempting to do with a recent post, and plenty of conservatives including those who should know better are following along:
The expression "create or save," which has been used regularly by the
President and his economic team, is an act of political genius. You can
measure how many jobs are created between two points in time. But there
is no way to measure how many jobs are saved. Even if things get much,
much worse, the President can say that there would have been 4 million
fewer jobs without the stimulus....
"how can the American people gauge whether or not your programs are working?... What metric should they use?", citing the 4 million job figure is a non sequitur, or more likely a diversion. A metric has to be measurable, and the actual number of jobs "created or saved" by the policy will never be measurable from any data source.
"how can the American people gauge whether or not your programs are working?... What metric should they use?", citing the 4 million job figure is a non sequitur, or more likely a diversion. A metric has to be measurable, and the actual number of jobs "created or saved" by the policy will never be measurable from any data source.
This is hackery as Mankiw's tap dancing addendum indicates. If you have been following American politics and the economy at all, the baseline case is the do-nothing case. In that scenario, it is projected that the American economy will lose millions of jobs in a prolonged recession. The stimulus can and should be judged against this type of counterfactual.
The Congressional Budget Office, whose numbers are jibing very closely with the Obama Administration numbers provide their estimate of the counter-factual unemployment numbers. Their baseline has unemployment hitting 9.0% this Christmas and about 141 million jobs. There are other macro forecasts out there that are projecting roughly the same or worse numbers in the 'do-nothing' baseline case.
So here is an easy metric to see if the stimulus is working --- is the U-3 unemployment rate in December beneath 8.5%? If it is, then the stimulus is most likely working? Another metric is the numebr of jobs in the economy? Are there at least 142,400,000 jobs? Those are the low case numbers.
Now onto the 'create or save' metric that Mankiw is complaining about, by 2011 when most of the stimulus impact will have hit the economy, are there roughly four million more jobs in reality than measured against the counterfactual baseline as defined by the CBO?
Those are real and answerable questions. They are basic policy analytical questions and techniques that someone like Mankiw should be able to do in his sleep, or more likely assign one of his grad students to do in lieu of sleep.
The next line of objection is to the validity of the models and their underlying assumptions, but Mankiw is willing to use long term economic models in the form of large error band 75 year acturial projections to repeatedly argue for his favored economic policy proposal of gutting Social Security, so I don't think Mankiw is a policy analysis nihilist when the economic models support his preferences.
Comments