Previous research has shown very little evidence that more intensive import competition from China has been broadly harmful to the economies of metropolitan areas. That's despite the fact that import competition in a given area is related to a sharper loss of manufacturing jobs. Indeed, the evidence I have reviewed previously suggests that manufacturing sector workers were less likely to be laid off over the last 15 years or so than workers in many other sectors. (There is healthy debate with other economists over this conclusion.)
Regardless of causes, the rapid loss of jobs in the U.S. manufacturing sector is undeniable, as is the intense focus on these types of jobs by the current presidential administration and other government officials at the state and local levels. Putting aside the question of whether this loss hurts the overall economy of a region, the question remains as to whether the loss of manufacturing jobs negatively affects people's personal psychologies and views of their lives. Is there a particular psychological price that is paid when manufacturing jobs disappear -- one that could in turn have an effect on how people view the world and vote for political candidates?
The hypothesis is that manufacturing downturns may create lower life evaluation, even for people or communities who apparently did not experience large losses in overall employment rates or average income.
For this research, I used data from over 173,000 individual adults living in the United States who were interviewed by Gallup in 2016 and asked to rate their lives on a 0 to 10 scale, where 10 is the best possible life and 0 is the worst. This scale, developed by Hadley Cantril many years ago, is highly correlated with various dimensions of the quality of life, including physical and mental health.
The focus of the research was to see if change in manufacturing jobs at the local level is a factor in helping explain variation in life ratings -- after standard variables such as children, age, race and educational attainment had been taken into account.
Manufacturing exposure for each county in the country was measured using three indicators obtained from government data:
Growth rate of manufacturing jobs from 2000 to 2015
Change in share of all jobs that are in manufacturing, from 2000 to 2015
Current (2016) share of all jobs in manufacturing
The results of my analysis show that living in a region that has lost manufacturing jobs is related to lower life evaluation, but there isn't anything special about manufacturing in this regard. It is bad to live in a declining area, no matter what the industries are that are declining. In fact, the effect of manufacturing decline is somewhat less significant than a decline in non-manufacturing sectors combined.
Importantly, there is no negative effect of living in a place whose job base begins to move from the manufacturing sector to other sectors. The data do show that manufacturing and production workers generally evaluate their lives lower than others, but this does not seem to depend on where they live and their level of exposure to regional job losses.
Overall, the data simply don't support the hypothesis that a decline in manufacturing jobs in an area has an unusually negative effect on how individuals who live in that area evaluate their lives -- any more so than a decline in jobs in general.
Additionally, the current share of jobs in manufacturing and the change in the share of jobs in manufacturing become insignificant -- in terms of predicting life evaluation -- after adding very basic controls, such as the county's college attainment rate. Only increases and decreases in manufacturing jobs remain significant, but overall growth and non-manufacturing growth both have a stronger effect on life evaluation.
The results described above are at the individual level, but an analysis of these same variables by region confirms the conclusion that shifting jobs from manufacturing to other sectors is not damaging to community life evaluation.
I should note that, as is often the case, there are limitations to this research. Most importantly it is difficult to isolate the causal effect of manufacturing decline, because many other changes can occur simultaneously even within the same local area that is suffering losses in manufacturing jobs -- and these other factors can either exacerbate or offset the loss of manufacturing jobs.
Still, the conclusion from the research seems clear. People living in areas that lose manufacturing jobs exhibit somewhat lower evaluations of their lives, but not if those losses are offset by gains in other sectors. There is no evidence in these data to indicate that the regional decline of manufacturing per se has resulted in lower life evaluations for people living in those communities.
This is important given the unique emphasis placed on manufacturing jobs by President Trump -- both in his presidential campaign and now in his presidency. Manufacturing jobs can be good jobs, of course, but there doesn't appear to be anything unique about them in terms of the way people look at their lives when other jobs are developed to replace them. Automation and other technological trends will inevitably reduce the share of manufacturing jobs in the future regardless of government policies. The key will be ensuring that the economy is working well enough to develop new industries and that training and educational programs are in place to ease the transition to other kinds of work.