Productivity Should Rise as Unemployment Rises: It Isn't

by: Casey Mulligan

As long as productivity continues to advance, this recession can remain mild or mediocre.

That's why the BLS release Thursday (drastically revising its earlier estimates) is discouraging. It reported that productivity declined slightly.

The chart below (Fig. 5 from my NBER working paper, revised with the new BLS data) compares productivity in this recession with previous ones. For the first three quarters, productivity was rising, making it quite unlike the 1970s and 1980s recession, and quite unlike the Great Depression of the 1930s.

Click to enlarge:

In this situation, a slight decline is bad news because productivity ought to rise when employment falls. This is why I calculate a "productivity residual" -- what productivity would be if labor usage had been constant. The chart below (Fig 6 from my NBER working paper, revised with the new BLS data). The productivity residual change 2008 Q3-Q4 looks a lot like the 1970s and 1980s recession.

Click to enlarge:

If the Q3-Q4 changes continue, this recession will look like no other since the 1930s in that both productivity and labor supply/labor distortions would be getting worse at the same time.