Understanding the Impact of Robots on Employment — Part 1
By Siddharth Singh, 15th April 2015
The robots, they are coming. The International Federation of Robotics (a German organisation) forecasts a massive uptake of industrial robots in the next few years, particularly in China which is projected to witness an industrial robot growth of 174% between 2013 and 2017.
Industrial robots are employed when they are able to do more — and cost less — than their human counterparts. Indeed, raised productivity and therefore output (in the short run) are the motivation and outcomes of greater automation in production. Intuitively, this would at least lead to the restructuring of the labour force, if not aggregate job losses.
Jeffery Sachs (at The Earth Institute at Columbia University and the author of The End of Poverty) along with LaGarda and Benzell of Boston University explore the impact of robots in the economy through a theoretical framework in a working paper for NBER. They write,
“Our key finding is that an increase in robotic productivity will temporarily raise output, but, by lowering the demand for labor, can lower wages and consumption in the long run. In what we term a paradox of robotic productivity, innovations that increase the productivity of robotic investments can, after a generation, lower robotic and total output, and lower the well-being (lifetime utility) of all future generations. The mechanism for this immiserization is decreased wages of the workers with whom the robots compete.”
A loss of human employment owing to the replacement by robots is already being seen, and has been seen in the past. Bloomberg reports that in China, there have already been instances of job losses due to increased automation. For instance, a major manufacturer of air-conditioners plans to cut 6,000 of its 30,000 workers in 2015 to make way for industrial robots.
A study by Henry Siu of the University of British Columbia and Nir Jaimovich of Duke University reveals that in the United States, since 2001, routine jobs have been lost compared to non routine jobs. Siu speculates that these jobs have been lost to automation rather than outsourcing or any other reason. The graph below shows the percentage change in different categories of jobs.
In their working paper, Sachs et al conclude by discussing the role of the state in ensuring optimal outcomes.
“The fact that a rise in robotics productivity may immiserize future generations is paradoxical. After all, higher productivity enables society to produce more output for the same level of inputs. If the market response to robotic innovations does not lead to a positive result, this suggests that there may be a role for government intervention. We show this intuition to be correct. Immiserization may be overcome through redistributive policies of the state.”
The role of the state, they show, includes redistribution.
In China’s case, the government has been promoting greater automation as it seeks to be a manufacturer of higher valued products that need greater precision. The argument goes that this will also lead to job growth in the robot production industry.
On the other hand, in the case of routine job losses in the United States, Jim Kessler, the head of a think tank, proposes a policy response primarily involving skill development.
Postscript: As Sachs reveals, the word robot was derived from a dystopian science fiction book from the word ‘robota’, which means forced labour.
Siddharth can be followed on Twitter @siddharth3
