Innovation is supposed to be the cure for economic doldrums. But what if it’s the cause? More specifically: Is it possible that the ever-increasing automation of everything from factories to retail sales to journalism will destroy more jobs than it creates?
It’s a question that economists and workers have been asking since at least the Industrial Revolution. And in the past, the answer has generally been a straightforward “no.” Automation makes certain low-skill human jobs obsolete, sure, but it also ushers in new categories of high-skill employment, from engineering and equipment operation to banking and blogging. Its greatest effect is to increase productivity, which should raise incomes and stimulate demand for new products and services.
Yet the current jobless recovery, along with a longer-term trend toward income and wealth inequality, has some thinkers wondering whether the latest wave of automation is different from those that preceded it.
Massachusetts Institute of Technology researchers Andrew McAfee and Erik Brynjolfsson, among others, see a “great decoupling” of productivity from wages since about 2000 as technology outpaces human workers’ education and skills. Workers, in other words, are losing the race between education and technology. This may be exacerbating a longer-term trend in which capital has gained the upper hand on labor since the 1970s.
Replacing manual labor with machines on farms and in factories was one thing, the worriers say. Those machines were dumb and highly specialized, requiring humans to oversee them at every stage. But the 21st century is witnessing the rise of far smarter machines that can perform tasks previously thought to be immune to automation.
Today’s software can answer your calls, organize your calendar, sell you shoes, recommend your next movie, and target you with advertisements. Tomorrow’s software will diagnose your diseases, write your news stories, and even drive your car. When even high-skill “knowledge workers” are at risk of being replaced by machines, what human jobs will be left? Politics, perhaps—and, of course, entrepreneurship and management. The rich will get richer, in other words, and the rest of us will be left behind.
All of which has brought John Maynard Keynes’ concept of “technological unemployment” back into the academic discourse, some 80 years after he coined the phrase. On Wednesday, Pew Research and Elon University released a report titled “AI, Robotics, and the Future of Jobs.” The report compiles and summarizes the results of a sort of expert opinion survey in which the researchers asked 1,900 economists, management scientists, industry analysts, and policy thinkers one big question: “Will networked, automated, artificial intelligence applications and robotic devices have displaced more jobs than they have created by 2025?”
The results of the survey were fascinating. Almost exactly half of the respondents (48 percent) predicted that intelligent software will disrupt more jobs than it can replace. The other half predicted the opposite.
The lack of expert consensus on such a crucial and seemingly straightforward question is startling. It’s even more so given that history and the leading economic models point so clearly to one side of the question: the side that reckons society will adjust, new jobs will emerge, and technology will eventually leave the economy stronger. Even Keynes in 1930 assured his readers that technological unemployment would be only a “temporary phase of maladjustment.” This view has been so widely held for so long that the dissenters have taken on a derogatory label: Luddites. The original Luddites were handloom weavers in England who smashed and burned power looms and mills on the theory that technology posed a fundamental threat to human well-being. Who’d have thought that half the mainstream experts in the United States in the year 2014 would share the Luddites’ basic view of automation’s effects on the labor market?
“Automation is Voldemort: the terrifying force nobody is willing to name,” declared one respondent quoted in the Pew report. “Good-paying jobs will be increasingly scarce,” said another, NASA program manager Mark Nall. “I’m not sure that jobs will disappear altogether,” allowed Justin Reich of Harvard University’s Berkman Center for Internet and Society, “but the jobs that are left will be lower paying and less secure than those that exist now.”
Is this dim view justified? I put that question to McAfee, who along with fellow MIT researcher Brynjolfsson helped to launch the current debate about the effects of automation with their 2011 e-book Race Against the Machine and 2014 follow-up The Second Machine Age.
McAfee, who specializes in the effects of information technology on business, is unreservedly sanguine when it comes to technology’s effects on economic growth. “I’m super-optimistic about the size of the pie,” he told me. But even he is increasingly nervous about its effects on employment, and middle-class jobs in particular.
“It’s fairly unambiguous at this point that the middle class has been getting hollowed out,” McAfee said. “The balance between capital and labor is shifting, and the best work on that shift clearly cites information technology as the reason.”
McAfee believes the education system has to change to prepare young people for a world in which most of today’s jobs are automated. He’s just not sure whether it can change fast enough. “The closer we look, the more it seems like some of these transitions have not just been, you know, 24 or 36 months and then everybody’s OK,” McAfee said. “They’ve been long, they’ve been difficult in some ways, and they’ve required pretty serious policy responses. [Columbia University economist] Joe Stiglitz talks about how the mechanization of agriculture may have made the Great Depression last as long as it did.”
It’s possible that today’s hand-wringing is simply a result of people underestimating the power of capitalist economies to adapt and thrive. They’ve certainly done so in the past. And remember, half of Pew’s respondents still think we’ll be just fine.
Still, there’s no guarantee that the future will resemble the past. It should be deeply unsettling to policymakers that so many of the smart people who think about these issues believe this time could be fundamentally different.
One more thing: Even if Keynes is right that the pain caused by automation is temporary, and that the economy will adjust in the long run, it’s worth keeping in mind another famous quote from the same great economist: “In the long run, we’re all dead.”
This article is part of Future Tense, a collaboration among Arizona State University, the New America Foundation, and Slate. Future Tense explores the ways emerging technologies affect society, policy, and culture. To read more, visit the Future Tense blog and the Future Tense home page. You can also follow us on Twitter.