America has not yet seen waves of widespread job loss due to automation. The U.S. economy created 209,000 jobs in August to continue its long streak of positive job creation, and the unemployment rate is down to 4.3 percent. Productivity growth remains sluggish, which would not be the case if more efficient machines were replacing droves of workers.
The lack of an aggregate effect on the labor market does not mean automation has not affected the labor market at all.
A new working paper by Grace Lordan of the London School of Economics and David Neumark of the University of California at Irvine finds that increasing the minimum wage lowers the share of jobs susceptible to automation held by low-skill workers. A $1 increase in the minimum wage lowers this share by 0.43 percentage points. Increases also adversely affect the workers' likelihood of being employed and hours worked.
The authors analyze how those effects differ by demographic groups and the intensity of automatable tasks.
Younger workers are the focus of much of the minimum wage literature because most of these workers are low-skilled. Unsurprisingly, the authors find that a $1 higher minimum wage reduces the share of automatable jobs held by workers 25 and younger by 0.94 percentage points. For workers over 40, a $1 increase in the minimum wage reduces the share in automatable work by 0.72 percentage points.
Some industries, such as construction and retail, have estimated responses close to zero. In manufacturing, a $1 increase decreases the share of automatable employment among low-skilled workers by 0.99 percentage points.
While younger workers see large effects in many sectors, the estimate for manufacturing is close to zero. In contrast, the share for older workers in automatable jobs in manufacturing decreases by 1.68 percentage points in response to a $1 minimum wage increase. Previous minimum wage studies might have overlooked the extent of this effect on this subgroup of older workers.
Along other demographic dimensions, female and African-American workers tended to see larger adverse effects, although the magnitude of these differences varied by industry.
The decline in the share of automatable jobs held by low-skilled workers does not necessarily mean those workers became unemployed or suffered declines in well-being. Workers may have shifted to occupations with fewer routine tasks that were less susceptible to automation.
The authors estimate that negative effects on employment shares in automatable jobs are associated with job loss and unemployment for affected workers. Overall, a $1 minimum wage increase reduces the probability that one of these affected workers remains employed by 0.12 percent. The magnitude again differs across industries and demographic groups. Older workers in manufacturing are particularly vulnerable, as a $1 increase lowers the probability of employment for affected workers by 0.78 percentage points.
These older workers are also more likely to be major contributors to their household’s income, and the net effect of minimum wage increases on the well-being of low-income families could be worse than previously understood. Hours worked and the extent that workers have to switch occupations are also adversely affected by minimum wage increases.
These effects could be getting stronger over time as the scope of automatable jobs increases due to technological developments and reductions in the prices of many labor-substituting machines. Restricting their analysis to the period 1995 to 2016 and comparing to the overall results, Lordan and Neumarkfind some evidence that this is the case. The adverse employment effects of pending minimum wage increases could plausibly be larger than those found in the paper.
Many of the proposed or passed minimum wage increases are venturing beyond the pale of historical experience in two dimensions: the increases are larger, and the effects of automation on low-skilled workers is likely higher. Despite the stated well-intentioned goal, minimum wage increases often harm many of the low-income families they are supposed to help, and the net effect could be even worse going forward.
Charles Hughes is a policy analyst at the Manhattan Institute. Follow him on twitter @CharlesHHughes.
Interested in real economic insights? Want to stay ahead of the competition? Each weekday morning, E21 delivers a short email that includes E21 exclusive commentaries and the latest market news and updates from Washington. Sign up for the E21 Morning Ebrief.