Structural change, jobless recoveries and job creation

Structural change has in many ways become synonymous with jobless recoveries and in a way almost a bad word (phrase?).

In an interesting 2003 article, Erica Goshen and Simon Potter ask whether structural change can be responsible for jobless recoveries. Their chart 5 shows that since the recession of the 1980s, the percentage of total employment in industries undergoing structural change has increased substantially - from 50% to 80%

This finding is only suggestive since it doesn’t really get into the root of the problem, that is: What is it about structural change that makes jobs hard to create?

Here is a plot of job openings from JOLTS data via FRED from 2000 to current which seems to indicate that rate of job openings have reached a lower plateau(s).

JOLTS data does not go far back enough for us to see if the pre-2000 era had higher job openings. An indication that this might be true can be gleaned using openings data from the Business Employment Dynamics. Prior to 2000, the openings rate (measure differently as in the JOLTS) hovered between 2.0 and 1.5 percentages points but began declining in 1999 to about 1.0 percent. Because of the large pool of people looking for work, such a small decline can have large effects on the number unemployed and the unemployment rate.

Data from Business Dynamics Statistics compiled from the Longitudinal Employer-Household Database seems to also tell the same story. The job creation rate has declined over time. The net job creation rate however seems less pessimistic. The net rate subtracts from the job creation rate, the death rate of jobs and seems to indicate that while jobs are indeed harder to create, firms have also become somewhat more resilient in keeping jobs.

Of course, none of these pictures really address why jobs have become harder to create.