Ricochet is the best place on the internet to discuss the issues of the day, either through commenting on posts or writing your own for our active and dynamic community in a fully moderated environment. In addition, the Ricochet Audio Network offers over 50 original podcasts with new episodes released every day.
The Luddites and technophobes have a point. Machines do displace workers. Always have. From the cotton gin, machine tools, and punch cards to combine harvesters, industrial robots, and business software. And it is this “displacement effect” that leads to scary forecasts about AI and robots leading to mass technological unemployment and underemployment.
But MIT’s Daron Acemoglu and Boston University’s Pascual Restrepo argue in a rich new paper, “Artificial Intelligence, Automation and Work,” that there is far more to the story. For starters, automation may allow tasks to be performed more cheaply, increasing demand for them. The introduction of ATMs was followed by more jobs for tellers because it reduced the costs of banking, and banks opened more branches. Or the productivity effect could be broader: Agricultural mechanization lowered food prices and created more demand for non-agricultural goods and the workers producing them.
But the heart of the paper is really about what Acemoglu and Restrepo call the “reinstatement effect.” Sometimes tech progress can create new tasks and more employment opportunities, just the opposite impact of automation. From the paper: “As tasks in textiles, metals, agriculture and other industries were being automated in the 19th and 20th centuries, a new range of tasks in factory work, engineering, repair, back-office, management and finance generated demand for displaced workers.” The Luddites often fail to factor this in.
Good news: Acemoglu and Restrepo think AI is no different than technologies of the past in its ability to create new tasks to balance out those that get automated, whether by boosting productivity (and incomes and consumer demand) or through reinstatement. They note, for instance:
A recent report by Accenture identified entirely new categories of jobs that are emerging in firms using AI as part of their production process (Accenture PLC, 2017). These jobs include “trainers” (to train the AI systems), “explainers” (to communicate and explain the output of AI systems to customers), and “sustainers”(to monitor the performance of AI systems, including their adherence to prevailing ethical standards). The applications of AI to education, health care, and design may also result in employment opportunities for new workers.
But there are no guarantees here. And the transition process will likely be difficult, just as it was with the first Industrial Revolution. A great history lesson here about “Engel’s Pause”:
The rapid introduction of new technologies during the British Industrial Revolution ultimately led to rising labor demand and wages, but this was only after a protracted period of stagnant wages, expanding poverty, and harsh living conditions. During an eighty year period extending from the beginning of the Industrial Revolution to the middle of the 19th century, wages stagnated and the labor share fell, even as technological advances and productivity growth were ongoing in the British economy, a phenomenon which Allen (2009) dubs the “Engel’s pause” (previously referred to as the “living standards paradox”, see Mokyr, 1990). There should thus be no presumption that adjustment to the changed labor market brought about by rapid automation will be a seamless, costless and rapid process. … It is perhaps telling that wages started growing in the 19th-century British economy only after mass schooling and other investments in human capital expanded the skills of the workforce. Similarly, the adjustment to the large supply of labor freed from agriculture in the early 20th-century America may have been greatly aided by the “high school movement” which increased the human capital of the new generation of American workers (Goldin and Katz, 2010).
The degree of difficulty of our 21st-century transition will depend on how quickly and appropriately we can educate and train our workforce, though the economists concede “there is little concrete information about what types of skills new technologies will complement, underscoring the importance of further empirical work in this area.” Failure here will not only make the transition more painful but also will crimp productivity gains. Maybe that’s already happening and is why productivity growth has been so weak despite all the tech advances we see around us.
The economists even raise the prospect that we are suffering from an “excess of automation” brought about by a weakness in worker skills and a tax code that subsidizes capital relative to labor. Or maybe too many of the advances are of the automating variety, rather than task-creating. (A similar theory has been proposed by Clayton Christensen.) This could be a sign that not enough brainpower is being applied to breakthrough discoveries and research. But the bigger point is that these technologies are not incomprehensible, unstoppable, unalterable forces of nature. What we do matters in making sure technology progresses and that it continues to benefit all of us.