Sixty years ago, an economist named Kenneth Arrow sat down with a seemingly obvious conclusion: workers get better at their jobs by doing them. This insight was simple, but Arrow, who later won the Nobel Prize, formalized it into a theory that had widespread influence. Learning, he wrote, “can only occur by trying to solve problems and therefore can only take place in activity.” Experience, he argued, was not only beneficial to workers but also an engine of productivity growth for businesses and the economy as a whole.
Now, as artificial intelligence eats away at entry-level jobs that once served as entry points to white-collar careers, researchers at the Federal Reserve Bank of Atlanta are revisiting Arrow’s 1962 paper and warning that companies racing to automate approaches to reduce wage costs may cut branches where they currently operate.
The unemployment rate among young degree holders currently remains higher than the overall unemployment rate, a reversal of recent workforce trends that many blame on artificial intelligence replacing entry-level knowledge jobs. Some college graduates are now facing unemployment rates similar to those of their peers without degrees, suggesting that a college education may become harder to justify and that the appeal of securing a stable position in an office job may lose its luster.
But if enough entry-level jobs are eliminated, those white-collar employers will start to hurt, too. That’s the conclusion of a paper published last week by researchers at the Federal Reserve Bank of Atlanta that analyzes the trade-offs on both sides of the field of automating entry-level office jobs.
Arrow believes that innovation and productivity growth are by-products of experience and practice. Fed researchers apply this framework to the drudgery of entry-level jobs, arguing that this experience is the foundation for building the expertise needed for senior positions. Crucially, the types of repetitive activities and skill-building that occur early in a young person’s career cannot be replicated in college or graduate school, with entry-level positions effectively becoming specialized crash courses to prepare employees and ensure that the company’s institutional knowledge remains intact.
“The tasks of filling entry-level positions are not just low-value work, they are lessons in building human capital for employees that will make them productive later in their careers,” the researchers wrote.
By automating more of these job roles, companies run the risk of weeding out capable senior staff they may need in the future, trading short-term cost savings now for long-term stability. Because Arrow theorizes that experience-based learning and productivity gains spill over and ripple throughout the economy, rather than being localized to one company, one company’s choice to automate entry-level tasks or roles will ultimately impact other companies in the industry as well.
