Obviously, I am going through Macroeconomic Patterns and Stories chapter by chapter. It is really too bad that he positioned it as a textbook, because I think that the rest of the profession needs to read it.

Anyway, he signed the preface “8 February 2007,” and in chapter 9 he forecast the characteristics of the next recession.

As of 2006, there has been no “recovery” of jobs in manufacturing in either durables or nondurables, and employment levels in those sectors are very low by historical standards. Thus it seems unlikely that there could be another bout of substantial job loss in those sectors…

Exceptional job growth is confined to construction…if we go to recession levels we could lose something like 1 million construction jobs…with 133 million payroll jobs, that is only 1/133 = 0.8% of jobs

So, how did things turn out? Go to these slides from early 2009, and look at slide 27 “Manufacturing Forms the Largest Share of Job Loss (not this time!).” Indeed, by the end of 2008 construction had lost more total jobs than manufacturing (which is not typical), but the really eye-popping losses were in “other,” where over 2 million jobs were lost, more than double the job losses in manufacturing and construction combined.

On p. 124 of his book, Leamer writes,

it is mostly manufacturing and construction where the job losses occur. On average about 70% of the job losses occur in these sectors

He also regards the 2001 recession as an anomaly, because job losses were relatively higher outside of those two sectors.

My guess is that if you looked at things relative to trend, the huge share of job losses in non-manufacturing, non-construction and the relatively small share of job losses in manufacturing in 2008 would look even more striking.

This is not your father’s recession. Seeing these differences makes me even more skeptical of those who want to view today’s economy through the lens of a 1960’s era macro model.