guest post: the labor market seems to be missing high-skilled jobs, not high-skilled workers

The following is a guest post by Nate Wilmers.

We often hear that our big jobs problem is that workers without college degrees don’t have the skills needed for “today’s economy.” I’ve been thinking a lot about this chart (from Autor 2015), and I want to suggest that our problem for the last 15+ years has been basically the opposite. For some reason, the economy isn’t creating enough high value-added jobs for college grads and this is messing things up for everyone else.


In the 1980s we seemed to be headed toward a knowledge economy, with rapid employment growth among technicians, professionals and managers. In the 1990s, this high-end employment growth was coupled with growth in low-skilled service jobs, marking a period of polarization and “hollowing out the middle” in employment growth. It looked like we would end up with an economy of computer programmers and managers, tended by an underclass of more or less specialized servants. Though this is a darker future than promised by knowledge economy boosters, it’s now a pretty familiar picture.

But from 1999 to 2007 (and arguably beyond) something really unexpected happened. Growth in the share of highly skilled occupations plateaued while low skill job growth accelerated sharply. There was no increase in employment share above the bottom third of occupations. The economy just created a bunch of low-paying, low-skilled jobs. During the same period the 2 decade-long rise in the college wage premium abruptly stopped growing (Beaudry, Green and Sand 2014) and the rise in demand for non-routine analytical and interpersonal skills, ongoing since the 1970s, leveled off (Autor and Price 2013). The economy kept pumping out those more or less specialized servant jobs, but was way more stingy with everything else.

There are some parts of these trends that have been studied a lot. Computers, automation and trade with China undermined middle skill jobs like manufacturing assembly and clerical jobs. Some of the rise in the service jobs is rising demand from the rich and some is that we aren’t automating nursing care and wait staff.

But I’m puzzled about the recent flattening of employment growth at the highest skill levels. What happened to the knowledge economy?

There are two explanations that hinge on technology, one secular and one cyclical. In the 1980s and 1990s a bunch of IT investments were made, which, during implementation, needed college grads and replaced middle skilled workers. IT implementation had run its course by 2000, so this trend petered out and recent college grads had to work as baristas instead (Beaudry, Green and Sand 2014). Another explanation is that the bursting of the tech bubble led to temporary underinvestment in IT and as investment revives, so too will the knowledge economy jobs (Autor 2015). Since 2007, employment growth at the very top of the earnings distribution has increased a bit, and even as the college wage premium remains stagnant, the premium for post-graduate education has kept rising (Valletta 2016).

There could also be less anodyne causes. It could be that those at the very top of the income distribution, who have enjoyed such spectacular income gains over the last 35 years, are (1) getting better at opportunity and income hoarding, so they’re (2) keeping potential entrants out, jacking up pay, and in response (3) companies are figuring out how to do without these high skilled workers. This could happen through occupational licensing or through subtler means. Another possibility is that industry structure has changed in some way that makes broad growth of high-end occupations difficult: maybe dominant firms get big mark-ups (Barkai 2016), but can’t benefit from expanding and hiring more high-skilled workers.

So we don’t know why this is happening or how long it will last. But, it means many college graduates have been deprived of the good careers they were promised. Instead they compete with less educated workers for scarce jobs and drive credential inflation. It’s hard to see how wages for less educated workers can rise much if underemployed college graduates keep taking the entry-level service and clerical jobs. It’s also hard to expect lot’s of productivity growth absent the engineers and managers and whatnot that tend to make this happen.

Without knowing why this is happening it’s hard to propose solutions. There is still a big college wage premium, and probably still a lot of barriers to college enrollment and graduation. Backing off on expanding college attendance would be like closing high schools in response to the World War II collapse in the high school wage premium: it’s a good thing we didn’t do that. But, the occupational patterns reviewed here suggest that the usual benefits from minting more college degrees could be offset by low demand from employers. For all the talk about problems of/with low-skilled workers, it might make more sense to look for the problem at the other end of the labor market.

Nate Wilmers is a PhD student in sociology at Harvard University.

Author: Dan Hirschman

I am a sociologist interested in the use of numbers in organizations, markets, and policy. For more info, see here.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: