Paul Hennessy/NurPhoto via Getty Images
Lots of people are worried AI is going to take a bunch of jobs.The fear that new tech will replace workers has been common throughout history. And yet here we all are, still with jobs to do.
For decades, there have been fears that new technology was about to wipe out a generation of workers.
The ATM was supposed to wipe out bank tellers. Spreadsheets were meant to wipe out bookkeepers. Robots were going to make humans redundant in manufacturing.
And yet, here we are in 2023, with unemployment in the US at 3.8%, and an estimated 9.6 million jobs available. So what happened?
That’s an especially important question to ask now as AI emerges as the next big transformative technology, impacting everyone from customer service agents to coders.
The answer can offer lessons for the future.
Tech typically creates more jobs overall
Simply put, technology creates more jobs than it takes away. Morgan Stanley addressed this in a recent report:
“During many prior periods of technological innovation, there have been predictions of tremendous job losses, and broadly what we have seen is the opposite driven by increased productivity, lower prices and also entirely new products and services.”
David Autor, a leading economist who’s followed automation and technology in the workplace, also wrote about this dynamic in an important 2015 paper that asked: Why are there still so many jobs?
He wrote then:
“Journalists and even expert commentators tend to overstate the extent of machine substitution for human labor and ignore the strong complementarities between automation and labor that increase productivity, raise earnings, and augment demand for labor.”
Tech can increase the need for a specific role by shifting costs and creating new demand
There are a few ways this plays out. In one scenario, the adoption of technology can increase the need for a specific role.
For example, there were fears that the advent of ATMs would put bank tellers out of work. And sure enough, a few years after the adoption of the ATM, there were fewer bank tellers per branch.
But the ATMs also reduced the cost of operating a branch. Banks responded by opening more branch offices, and bank teller employment continued to increase.
Here’s economist James Bessen speaking on a podcast about his research on this topic:
Well, the average bank branch in an urban area required about 21 tellers. That was cut because of the ATM machine to about 13 tellers. But that meant it was cheaper to operate a branch. Well, banks wanted, in part because of deregulation … but just for basic marketing reasons, to increase the number of branch offices. And when it became cheaper to do so, demand for branch offices increased. And as a result, demand for bank tellers increased.
Tech can also create new roles in adjacent professions
In a second scenario, tech might reduce the need for one specific professional, but help create new roles in adjacent professions.
In a recent article, I offered the example of Microsoft Excel. Spreadsheets reduced the need for bookkeepers, but massively increased the need for financial managers and accountants, creating many more new jobs than were lost.
Here’s Morgan Stanley:
“As adoption of this technology grew rapidly throughout the 1980s, especially after the introduction of Microsoft Excel in 1987, we saw a reduction in the number of Americans working as bookkeepers and accounting/auditing clerks (from ~2 million in 1987 to just above 1.5 million by 2000) — but we also saw a significant increase in Americans employed as accountants/auditors (rising from ~1.3 million in 1987 to ~1.5 million in 2000) and management analysts & financial managers (from ~0.6 million in 1987 to ~1.5 million in 2000)”
But those in an impacted profession can find themselves worse off
Technology then creates jobs overall. That’s why there are still so many of us employed at a time of historic technological advancement.
But the diffusion of technology does have the potential to disrupt a labor force in ways that negatively impacts workers.
Charter, the workplace-focused media and consultancy company, cites the example of women who worked as telephone operators in the early 1900s.
Mechanical switching made that role redundant, forcing women to find work elsewhere. And they did. This new technology did not reduce employment among young women.
But, per a 2020 working paper from the National Association of Economic Research (NBER), the disruption may have led to those women earning less in the future. From the NBER paper:
The reduction in telephone operators was offset by growth of demographically-similar jobs with similar or slightly lower wages (e.g., typists and secretaries, restaurant workers). The specificity of this countervailing employment growth in a select set of occupations that were demographically similar to telephone operators suggests task reinstatement restored the employment levels of future cohorts
And existing workers sometimes suffer as a result of increased competition
There’s a second way this can play out. I wrote earlier this year about white-collar workers facing their own Uber moment. This refers to the introduction of Uber to London, where taxi drivers were previously required to pass an intensive exam, capping the supply of taxi drivers.
Suddenly, with Uber, every driver with a cellphone could compete for business. This put downward pressure on taxi-driver incomes.
“Suddenly, knowing the name of each street in London was no longer valuable expertise, so that anybody with a drivers license could drive a taxi,” Professor Carl Benedikt Frey, the director of future of work at the Oxford Martin School, told me earlier this year. “The result was more competition for incumbent taxi drivers who saw their incomes fall by around 10%.”
How a new technology is implemented is key
Technology typically creates new jobs and industries, more than offsetting any jobs that are lost as a result of the technology’s implementation.
There is still lots of work to go around, as is true today.
But there’s no guarantee that the individual worker impacted by a new technology will get the new job that’s being created, or that others won’t see their incomes fall as a result of fresh tech-enabled competition.
So it’s less a question of whether we’ll have work to do in the future, and more a question of what that work is, and what it will pay. There are fears that AI will exacerbate income inequality.
This puts a huge emphasis on how the technology is implemented, who gets to participate in its upside, and what happens to those workers who do find themselves disrupted.
AI is already at the heart of union negotiations in Hollywood. Morgan Stanley has predicted the rise of generative AI will create “unprecedented demand for reskilling.”
In a policy memo updated in September, Daron Acemoglu, Autor, and Simon Johnson wrote: “Generative Artificial Intelligence (AI) will surely impact inequality, but the nature of that effect depends on exactly how this technology is developed and applied. Nothing about the path of this (or any) technology is inevitable.”