Job polarization, where the majority of the job gains are concentrated in both the low- and high-wage categories, is best seen in the occupational data. The U.S. Bureau of Labor Statistics releases occupational data once a year, as opposed to industry data every month. The 2016 occupational data was released today, March 31st. What follows is a brief update on polarization in the U.S. Next week I will dive into the state level data. See our office’s 2013 report for a more thorough and complete look.
Overall, the same job polarization trends we have seen this century continued in 2016. Job growth was strongest among low- and high-wage occupations. The good news is that middle-wage jobs are also growing, however they took a step back in growth in 2016. This slowdown is concentrated among the traditional blue-collar occupations. Installation, Maintenance and Repair jobs, and Transportation jobs saw slower growth. Construction did too which is interesting given the ongoing (slow) housing recovery. Production jobs — essentially the manufacturing jobs that do the manufacturing — saw very little growth. Some of this is to be expected given the fallout of the oil crash to end 2014 and the strong U.S. dollar.
The second chart shows the actual change in employment by wage group since the start of the Great Recession. The biggest problem with job polarization is that middle-wage jobs decline the most in recession and don’t come back all the way in expansion, particularly as a share of the economy. Job opportunities for middle-wage jobs is a declining share over time, particularly since the turn of the century. That said, middle-wage jobs themselves don’t decline forever. As the Federal Reserve Bank of New York noted last year, the actual number of jobs created were strongest among middle-wage occupations in recent years. This is good news, even as growth rates remain lower for such jobs.
When it comes to middle-wage jobs there are a few important things to keep in mind. First, they impact both men and women, particularly those without college degrees. While many of us know the story of the manufacturing decline hurting men without college degrees, the same thing is true for women and administrative and office support occupations. In fact, here in Oregon the relative decline of office support jobs for women and production jobs for men have been equal in recent decades.
Second, the outlook varies depending upon which middle-wage job you are looking at. Some middle-wage jobs are driven by population gains. The more people you have the larger the demand for artists, clergy, construction workers, police officers, plumbers, teachers, and the like. Locations with stronger population gains have seen better growth in these types of middle-wage jobs. The other types of middle-wage jobs can broadly be considered business support occupations. From our report:
Administrative Support, [some] Sales and Transportation all act as suppliers of labor and services to other businesses or employees. With increases in business operations, including headquarters, the demand for such occupations will increase even if technological advancements continue to eliminate a portion of these jobs. This provides an opportunity for continued investment into activities that foster both an entrepreneurial business climate and also recruitment and retention efforts of existing firms. The loss of significant headquarter operations in Oregon over recent decades has decreased the demand for some of these business support firms and workers.
Finally, if you want to slice the data differently, you can look at changes between manual and cognitive, and routine and non-routine jobs.
In the short-term, middle-wage jobs are expected to continue to grow along with the economy. Some cyclical rebound is still likely/needed for both construction workers and teachers in particular. However, over the longer term, the relative share of jobs is expected to continue to decline as high- and low-wage jobs see stronger growth.
I will have a lot more on job polarization here in Oregon next week and plan on diving into other states in the near future, including an update on other measures like state level Total Employment Gaps.
Mark McMullen
Josh Lehner
Oregon Office of Economic Analysis 
Josh, as always this is fantastic information. I do have to make one comment regarding middle wage jobs. One of the definitions you tend to leave out is that it middle wage is a very limited range of incomes. In fact I would suggest that occupations like construction no longer fit into the criteria of the middle wage as it is common for construction worker to make over $30, $40 and even $50 an hour and in recent years their hours have been well over the normal 2000 per year; thus putting them well into the high wage earners. This then changes the perception of your comments to appear as if the middle class or middle wage earners are struggling to find work versus simply moving up and out of that wage range.
By: John Killin on March 31, 2017
at 10:56 AM
Hi John, thanks for the comments and insight. You’re right. There can be quite a range of wages within any given occupation, or in this case occupational group. We categorize them based on median wages (half of workers earn more, half earn less, obviously), but also take into account the academic literature about routine vs non-routine, manual vs cognitive tasks, as that is the real driver of the changes over time. As for construction itself, in 2016 median wages were about $47,000 which is at the high end of the middle-wage jobs with teachers at $48,000. Clearly there are many high-paying construction jobs, particularly in nonres occupations. Those are also very skilled jobs too. But just to give some idea of how these sort out, the lowest high-wage job — Scientists — have median wages of $57,000, then it jumps to $62k for business and finance, and then $75k for legal, with all others in the upper 70s or into the 80s, again for median wages in those occupational groups. So while there is clearly overlap depending upon the exact job and each individual worker’s experience and skill, the data does sort pretty well overall, even if it is clearly a simplification at a 10,000 foot view.
By: Josh Lehner on April 3, 2017
at 10:34 AM
[…] Source: Job Polarization, 2016 Update | Oregon Office of Economic Analysis […]
By: Job Polarization, 2016 Update | Oregon Office of Economic Analysis | eClips on March 31, 2017
at 12:44 PM
[…] perspective on the future of work comes from the Oregon Office of Economic Analysis, where they’ve found that jobs with predominantly routine tasks have declined, but […]
By: TedX ReBlog: Will automation take away all our jobs? – Oregon Auditing on April 5, 2017
at 9:00 AM
[…] Last week we took a look at the newly released occupational data and job polarization across the U.S. Today is a relatively quick graphical update for Oregon and some of our regions across the state. At this point I don’t think there is a whole lot more to add in terms of the narrative. Please see last week’s post and our previous report for a more complete look. As such, I’ll get straight to the charts. […]
By: Job Polarization in Oregon, 2016 Update | Oregon Office of Economic Analysis on April 6, 2017
at 9:35 AM
[…] Last week we took a look at the newly released occupational data and job polarization across the U.S. Today is a relatively quick graphical update for Oregon and some of our regions across the state. At this point I don’t think there is a whole lot more to add in terms of the narrative. Please see last week’s post and our previous report for a more complete look. As such, I’ll get straight to the charts. […]
By: Job Polarization in Oregon, 2016 Update | eClips on April 6, 2017
at 11:10 AM
The data is helpful but the analysis is incomplete. At $50k, Middle Wage, most individuals would be living pay check to pay check in a rental home. If they have a family, then the domestic budget is very tight.
By: Dan on April 12, 2017
at 9:49 AM
Hi Dan, thanks for the comment. You’re right. We’re not talking about a middle class lifestyle or how much purchasing power any individual job has. That’s a whole different discussion and set of assumptions. We’re focusing on the middle part of the income distribution, or wage distribution. Roughly speaking, low-wage jobs make up the bottom 20-25% of the wage distribution, middle-wage jobs the middle 50-60% and high-wage jobs the top 20-25% of the distribution.
By: Josh Lehner on April 12, 2017
at 10:00 AM