Nonroutine is the key

A good while ago I posted about research showing how at-risk routine work was compared to non-routine work. Specifically, routine manual work was the most exposed to disintermediation or technological replacement. This was followed by routine cognitive work (think administrators). People performing jobs that were non-routine, either cognitive or labor both were stable or growing. This research was from some time in the early aughts.

I came across some new research that brings the data up to date but with the same results. From Jobs Involving Routine Tasks Aren't Growing by Maximiliano Dvorkin.

Given this, it is important to classify occupations according to how routine their tasks are. It is also important to classify occupations by whether they use mostly cognitive skills or mostly manual skills (brain vs brawn). The following figure shows the evolution of U.S. employment across four types of occupations:

* Nonroutine cognitive occupations, which include management and professional occupations
* Nonroutine manual occupations, which include service occupations related to assisting or caring for others
* Routine cognitive, which include sales and office occupations
* Routine manual, which include construction, transportation, production and repair occupations2

Click to enlarge.

The picture is clear: Employment in nonroutine occupations—both cognitive and manual—has been increasing steadily for several decades. Employment in routine occupations, however, has been mostly stagnant.