Still trying to make a silk purse out of a cow's ear

Ok, we're all depressed about the state of the economy. But here's some evidence that things are better than one would have expected given where we were at the beginning of 2009. Using data from 1980Q1 to 2009Q1, estimate a second-order autoregression model for GDP growth. Use the model to forecast GDP growth in 2009Q2 and Q3. Actual GDP growth is considerably better (by almost 3 percentage points) in 2008Q3 than the model predicts.


Do the same for employment growth. While we're still losing jobs, job losses are 2.8 percentage points smaller (on an annual basis) in Q3 than the model predicts. That translates to about 935,000 more jobs that would have been lost in Q3 alone if the economy had behaved as it normally does. The same computation for Q2 gives us 368,000 more jobs in that quarter than expected.

What does this tell us? Maybe that the extraordinary monetary and fiscal policy actions of the last year (plus other unusual things - strong recovery overseas and so on) have pushed GDP growth into positive territory and "saved or created" 1.3 million jobs.