Christopher L. Smith on Job Polarization

Recommended. A few highlights:

the inflow rate to middle-type jobs from unemployed formerly non-middle-type workers appears to be falling over time. Further, these transition rates appear to drop discretely following the 1990, 2001, and most recent recessions.

There is no evidence of an increase in the rate at which unemployed middle-type
workers transition to non-middle-type jobs (bottom middle plot). Instead, there
is an upward trend in the rate at which these workers remain non-employed,
though this is true of unemployed workers of all types…

For all horizons, the rate at which persons from low- or middle-type jobs transition to high-type jobs is rising over time… This is particularly true for the transition rates from middle-type jobs.

Conversely, the rate at which persons transition from low-type jobs to middle-type jobs is falling over time…

Later, he suggests that it is workers 55 and older who are transitioning out of (I assume losing) middle-type jobs, while it is younger workers who are not transitioning into middle-type jobs.

Pointer from Tyler Cowen, although I had seen an article on the paper somewhere else, also.