What are the economic consequences of automation?
That’s a question that economists and pundits are desperately trying to answer as more and more businesses look to robotics, machine learning, and artificial intelligence to make production more efficient.
Two economists, MIT’s Daron Acemoglu and Boston University’s Pascual Restrepo, recently crunched a few datasets and concluded that when robots are introduced, human wages decline. Bloomberg took their data and created interesting visualizations that show the relationship between “industrial robot exposure” and employment, as well as those areas of the country that saw the biggest increase in robots (per thousand workers) between 1990 and 2007. For example:
“Bottom line: Robots do replace workers,” Bloomberg’s article concluded. “On the other hand, some industries that don’t automate end up losing workers anyway, because their costs are too high and their customers go elsewhere. For workers, robots are only part of the problem.”
Over the past few years, pundits have debated at length about ways to mitigate the disruptive effect of automation on human beings. In February, Microsoft co-founder Bill Gates went so far as to suggest a “robot tax.” His idea was met with immediate pushback from other public figures, most notably former U.S. Treasury Secretary Lawrence Summers, who stated in a Washington Post op-ed that such a tariff might act as “protectionism against progress.”
Even as debate rages about the possible job-killing effects of automation, it may take several years (if ever) for certain industries to feel the impact of the rise of the machines. Although researchers have made great strides in artificial intelligence and machine learning, not all of their initiatives have panned out as hoped. Despite lots of hype, for example, chatbots have so far failed to beat out flesh-and-blood customer service reps.
Nonetheless, artificial intelligence is advancing in fits and starts—and in the end, the consequences on jobs could extend far beyond factory workers. How many years may pass before professions with a lot of abstract thinking—such as programming—are widely affected by A.I.?