We have seen the enemy, and it is the robot.
A wave of fear over automation is driving dire predictions about robots rendering swaths of Americans unemployable and has created a vogue for outlandishly stupid policy proposals. A popular idea in Silicon Valley is for a so-called universal basic income — the government would write everyone a monthly check — to cushion the unemployment effect of automation (never mind that such a policy would itself have an unemployment effect). Bill Gates wants to tax robots, which makes about as much sense as taxing personal computers back in 1991.
The rising tide of hysteria about our technological future has to count as one of the most pernicious features of the current intellectual climate. The only way to raise our standard of living is to increase productivity, which is impossible without innovation. This process always destroys some jobs but, by raising wages and cutting prices, makes it possible for new jobs to spring up in their place.
This, in a nutshell, is the story of the American economy, as Robert Atkinson and John Wu describe in a new study for the Information Technology & Innovation Foundation. It’s not a tale of stasis but of gales of technological change constantly making the economic landscape anew. It’s hard to see, as examples of past occupational shifts cited by Atkinson and Wu make clear, how anyone would want it any other way.
Do we really want to recover the tens of thousands of pinsetter jobs that used to be filled by boys and young men at bowling alleys prior to the adoption of the automatic pinsetting machine in the 1940s? Or return to 1950, when there were more than 100,000 elevator operators, an occupation that essentially disappeared after the widespread adoption of self-service elevators? Or rekindle the glory days of motion-picture projectionists, who numbered more than 30,000 in 1940 and now, in the age of multiplexes, fewer than 5,000?
What technology giveth, it taketh away. Atkinson and Wu note that occupations associated with railroads — engineers, conductors, etc. — increased 600 percent in the 1850s. They kept growing until the 1920s, when the advent of cars and trucks began to erode the railroad market; the trajectory in railway jobs has been downward ever since.
The number of telephone operators and data-entry clerks has declined by about 50 percent since 2000, although we don’t lack for communication or ready information.
This sort of occupational churn is inevitable when an economy is becoming more productive, and indeed, the average American worker — equipped with once-unimaginable technological tools — now produces in an hour what his predecessor 100 years ago produced in a day.
The key finding in the Atkinson and Wu study is that, despite the alarm about automation, the American labor market in recent decades has had its lowest rate of occupational churn since 1850. Instead of being overrun by robots, the American economy is lagging in terms of a technological change. Tellingly, productivity growth has increased at just a 1.2 percent annual rate from 2008 to 2015, a sharp drop from the 2.8 percent annual rate from 1947 to 1973.
Atkinson and Wu believe that another wave of technology is coming, but perhaps not for decades. “Rather than too much innovation or productivity,” Atkinson and Wu write, “the biggest risk is that economies will not be able to raise productivity fast enough to adequately raise per-capita incomes, especially in an era where nations face growing elderly populations.”
We can argue about the best ways to train higher-skilled workers or to cushion the blow to workers of economic change, but technophobia is a formula for stagnation. If we survived the advent of cars, the telephone and electric lighting — truly revolutionary changes around the turn of the 20th century — surely we can endure whatever transformative innovations that will, once again, make our economy and workers more productive than ever.
Don’t fear the robot.
— Rich Lowry is the editor of National Review. He can be reached via e-mail: email@example.com. © 2017 King Features Syndicate