The Fed may have crushed entry-level jobs more than AI
Add Axios as your preferred source to
see more of our stories on Google.
Illustration: Sarah Grillo/Axios
This has been a hard couple of years for people starting their careers, with the unemployment rate among young adults skyrocketing. It could be early evidence of AI reshaping the job market — after all, the bots are already quite good at tasks often assigned to entry-level office workers.
- Not so fast. The timing and details of the jobs slump tell a more routine macroeconomic story.
The big picture: The deterioration in the job market for young workers began before ChatGPT was widely available, and coincides with the Federal Reserve's aggressive tightening of monetary policy in 2022-2023 — which was meant, in no small part, to cool down an overheated labor market.
- So say economists from Google in a new paper that meshes with other analysis in the last few months suggesting that recent job losses have not been particularly concentrated in the areas most susceptible to AI-replacement.
Yes, but: That's a backward-looking story, and doesn't answer the question of how much AI-driven mass displacement of workers looms ahead. As Axios' Jim VandeHei and Mike Allen summarized, those threats are increasingly real.
- Arguably, it's more worrying that the U.S. is entering this period of AI disruption not from a starting point of labor market robustness, but after a three-year-running, garden-variety labor market slowdown that has hit young workers particularly hard.
State of play: The unemployment rate among 20 to 24 year olds reached a 54-year low in the spring of 2023, at 5.5% — but it has risen sharply since then, to 8.2% in December.
- It stands to reason that the advent of generative AI is part of the reason, as suggested by a much-discussed paper from last year pointing to particularly steep declines in employment of young people in AI-affected fields.
- Zanna Iscenko and Fabien Curto Millet from Google's economics team, in a paper published by the Economic Innovation Group, argue that the inflection point in employment of AI-affected workers began months before the 2022 rollout of ChatGPT.
What they're saying: The economic challenges for young workers in white-collar fields "are palpable and distressing," they write, but the emerging narrative around AI, they write, is flawed.
- "The most plausible explanation is that the data patterns observed are not early warnings of large-scale technological displacement, but rather the predictable consequences of a classic macroeconomic shock: the sharpest monetary policy tightening cycle in four decades."
- The Fed, of course, raised interest rates by more than 5 percentage points from March 2022 through the summer of 2023 to combat the post-pandemic inflation urge.
Zoom out: Iscenko and Millet observe that in business cycles, it is usually newer workers who bear the brunt of labor market deterioration. It stands to reason that this would be true in the Fed-engineered job market slowdown of the last three years.
- During economic downturns, they note, hiring stops and fewer people quit their jobs.
- "The result is that the primary entry points to the labor market and the key pathways for career progression disappear, leaving young workers stranded at the bottom of the ladder or unable to get on it at all," Iscenko and Millet write.
Employers generally say AI is not leading to widespread layoffs. But it is one reason they are holding off on hiring.
- That difference might help explain why the "low-hire, low-fire" labor market dynamic has persisted, at least in recent months.
- While AI is not yet replacing existing staff, companies are thinking about how the technology helps them limit the need to bring on more workers.
- It makes the labor market tougher for workers looking for a job than for those who already have one.
That is, at least, one takeaway from nationwide anecdotes collected by the Federal Reserve regional banks.
- "Multiple contacts reported exploring AI implementation primarily for productivity enhancement and potential future workforce management," officials wrote in the latest Beige Book, released Wednesday.
- "AI's current impact on employment was limited, with more significant effects anticipated in the coming years rather than immediately."
In the Boston Fed district, a staffing services firm "noted that some larger companies had made selective layoffs to reduce costs but said that AI was not a factor behind these decisions."
- But one IT services firm told officials that it "paused hiring plans as it considered using AI instead."
The intrigue: The Cleveland Fed reported "increased caution in hiring" across multiple industries in the district. Firms were requiring approvals to backfill empty positions, relying on subcontractors and "strategically using technology to limit hiring."
- In the New York Fed region, officials noted less demand for marketing professionals, "partly due to increased efficiencies brought about by AI."
The bottom line: The anecdotes suggest AI-related employment effects are still minor, with early signs that it is improving productivity for workers — at least, for now.
- Philadelphia Fed-area businesses "noted only incremental productivity gains, with a minor impact on employment levels either from AI use specifically or automation generally."
