Arnold Kling  

The Stimulus Worked

New Richter Podcast... Futures Market Exposition...

The Council of Economic Advisers says so.

The reason that the Council has about as much credibility as Baghdad Bob is that chart that everyone has seen showing that unemployment with the stimulus is higher than what they predicted without the stimulus. Their response is to come up with a new prediction model.

We estimate a vector autoregression (or VAR) using the logarithms of real GDP (in billions of chained 2005 dollars) and employment (in thousands, in the final month of the quarter) over the period 1990:Q1-2007:Q4. We include four lags of each variable. Because the estimation ends in 2007:Q4, the coefficient estimates used in the prediction are not influenced by developments in the current recession. Rather, they show the usual joint short-run dynamics of the two series over an extended sample. We then forecast GDP and employment beginning in the second quarter of 2009 using actual data through the first quarter of 2009.

This model predicts that by now employment should be even lower than what it is. In retrospect, I will bet that the Council wishes it had used this model two years ago. I wonder if they will make any use of the model for prediction purposes going forward. Or, now that it has served its purpose, will they stick the model in a shelf in the back room.

Another takeaway I get from the report is that "shovel-ready projects" were the last component of the stimulus to kick in. Everything else is pretty much spent by now, but most of the "public investment" spending is ahead of us.

Comments and Sharing

CATEGORIES: Macroeconomics

COMMENTS (4 to date)
ThomasL writes:

They can promote a new model to justify the stimulus but only at the cost of implicitly acknowledging the correctness of the criticisms of their previous model -- ie, the model that they purport to have used to "scientifically" craft the parameters for stimulus.

It is nothing short of miraculous that their original model and their new model seem to recommend exactly the same actions, despite different outcomes.

Mike Rulle writes:

We know Christina Romer understands this is nonsense. Cherry picking models after the fact to suit ones political purposes is deceitful. At best, one can say she is so overwhelmed by researcher bias she is blinded. Yet, she is the same person that talks about "omitted variables" in critiquing Robert Barro's work on fiscal policy in the Korean War.

What about the counter factual of what would have happened had there been no stimulus? Or if there had been marginal tax cuts; or no Obama Care? These are unknowables. We only know she made a prediction and it was wrong. Her solution is to insist is was correct, if they only chose the right data mined model.

She should be villified by the economics profession regardless of one's political views. It's the equivalent of the Climate Change scandal.

Foobarista writes:

An old saying: if you torture the statistics enough, they'll tell you what you want to know.

frank cross writes:

The CEA actually uses two models. And somewhat different approaches are used by CBO and four private forecasters. All yield roughly the same results. Is there any actual evidence to the contrary?

Comments for this entry have been closed
Return to top