Automation may displace some workers and raise the possibility of technological unemployment but could also lead to productivity gains that may increase market demand, and consequently increase labour demand. New research by Philippe Aghion, Céline Antonin, Simon Bunel and Xavier Jaravel suggests that the productivity effect of automation may prevail over the displacement effect, generating gains that are broadly shared across workers, consumers and firm owners.
Their study of firms and plants in the French manufacturing sector between 1994 and 2015 finds that both firm-level and plant-level employment increases after automation, even for low-skill industrial workers. They estimate that an increase of 1% in the use of automation technologies raises employment by around 0.4% after 10 years.
Following developments in automation technologies in recent decades, effects of automation in the labour and product markets are regularly debated, both from an academic and a public policy perspective – the taxation of robots being once such example. New research uses micro data on the population of firms and plants in the French manufacturing sector between 1994 and 2015 to provide an empirical analysis of the effects of automation on employment, sales, prices and the labour share.
The originality of the approach is to consider two precise and complementary measures of automation technologies: (i) the balance sheet values of industrial equipment and (ii) the plant-level records of the usage of electro-motive force, i.e. the quantity of machines used in the production process set in motion using electric motors.
Exploiting the timing of adoption of automation technologies, and contrary to the commonly held belief that firms that automate reduce employment, the researchers find that both firm-level and plant-level employment increases after automation, even for low-skill industrial workers. They estimate that an increase of 1% in the use of automation technologies raises employment by around 0.4% after 10 years.
Even if the absence of pre-trends in the event study is reassuring in order to interpret these results as causal, one may fear that simultaneous unobserved shocks – demand shocks for example - may actually explain the employment rise.
The study addresses these concerns using a research design that comes close to the ideal experiment that would randomly assign purchase prices of automation technologies across firms: it exploits productivity variations over time of foreign suppliers of industrial equipment that French firms are differentially exposed to through pre-determined importer-supplied relationships.
The results with this research design are very close to the event study results that exploited the timing of adoption of automation technologies: a firm that has a supplier of machines that becomes relatively more productive automates more, and at the same time increases its employment.
With this design, the researchers estimate that an increase of 1% in the use of automation technologies raises employment by 0.31% at firm level. Effects of automation on firm sales are similar to employment effects: an increase of 1% in the use of automation technologies leads to an increase in sales by 0.4% at firm-level.
Finally, in order to study business stealing and other equilibrium effects, the authors repeat the analysis at the industry level. Once again, the relation between automation and employment is positive, but seems to be stronger in industries that face international competition.
Altogether, these findings indicate that automation can increase labour demand and can generate productivity gains that are broadly shared across workers, consumers and firm owners. This is even more so the case in an open economy, where attempts to curb domestic automation in the hope of protecting domestic employment may be self-defeating due to foreign competition.
What Are the Labour and Product Market Effects of Automation? New Evidence from France
Philippe Aghion, Collège de France and London School of Economics
Céline Antonin, Sciences Po — OFCE
Simon Bunel, INSEE and Paris School of Economics (presenter)
Xavier Jaravel, London School of Economics
Contact :
Mail : simonbunel@gmail.com
Website : https://sites.google.com/view/simon-bunel/home
Phone: +33 6 31 05 45 32