Georgetown University’s Center on Education and the Workforce published an interesting analysis on the growth of employment during recovery from the recent recession.

The report counters what it describes as media portrayals of the recovery as being built on the creation of low-playing, low-skilled, benefitless, no-advancement positions as, say, baristas, Uber drivers and hamburger flippers.  Georgetown points to both the New York Times and the Wall Street Journal as among the culprits, citing articles written from 2012-15.  While this characterization may have been true in 2008-2009, the opposite has been the case during the five years since.

Over the past half-decade, job growth has been driven by “good jobs,”  which Georgetown defines as being in the upper third of their occupations by median wages.  Such positions pay $53,000/ year, or 26% more than the median for all full-time workers.  86% of “good jobs” are full-time, 68% offer health care benefits and 61% an employer-sponsored retirement plan.  Such benefits are typically add 30% in ecnomic value in addition to wages.

How can the media have been so wrong?

It’s because reporters have examined employment data by industry–looking at the types of products and services provided–rather than by the position being filled.  In other words, the reporters ended up counting a software engineer, an accountant or a marketing executive hired by Starbucks as a barista.

Looking at positions instead of industries, paints a different picture.

“Good jobs” have accelerated sharply since 2010.  Comprising 2.9 million out of 6.6 million total new jobs, they are dominating the recovery.  There are more “good jobs,” and more low-paying ones, today than there were in 2008.  Middle-wage jobs, however, are still 900,000 below their pre-recession levels (no explanation given by Georgetown for this).

The Georgetown report also shows that from 2010 on, workforce participants without at least some college have actually lost jobs across all wage categories–high, average and low–even though employment was expanding rapidly.  There are 39,000 fewer workers in “good jobs” who have high school diplomas or less (during a period when 3.1 million net new employees were hired), 280,000 fewer in average-paying jobs (2.1 million hired), and 159,000 fewer in low-paying ones (1.9 million hired).  It’s unclear how much of this replacement is due to employees upgrading their credentials, how much to changes in the labor pool, how much to changes in hiring practices…  In addition, college graduates made up 97% of the “good job” hires, 62% of the average-job hires and 39% of the low-paying hires.

I’m mostly interested in the economic implications of the Georgetown study.  But I find the report’s comments (p. 6) on the implausibility of the media articles to be interesting, and a little disturbing, as well:

“We find these media stories to be counterintuitive because they disagree with the well-established cyclical patterns of economic behavior. The consensus among economic researchers is that the economy has seen a strong shift toward college-educated workers since the early 1980s. The long-term shift in hiring, the increased economic value added, and the wage premium of college workers have persisted and strengthened for more than 30 years in periods of both recession and recovery. If the reports that the economic recovery was only producing low-wage, low-skill college jobs were true, they would suggest a profound reversal of structural trends in technology and globalization in place for decades. This seems unlikely given the weight of continued evidence to the contrary.”

I read this paragraph as addressing the issue of whether reporters just wrote pieces off the top of their heads or whether they may have simply been repeating the results of interviews with informed sources in the world of economics or government.  Georgetown seems to be saying pretty strongly that no credible person could possibly have told them anything remotely like they were printing.