Goldman Sachs says hiring at technology companies has slowed by about 5 percentage points per year since 2022 compared with the historical trend, and that roughly half of this reduction can be traced to firms scaling back headcount following the rapid expansion from 2020 through 2022.
In a research note, the firm grouped the commonly cited explanations for the slowdown into three broad categories: a hawkish Federal Reserve causing slower growth and higher rates, labor shifts tied to artificial intelligence improving efficiency, and a post-pandemic correction for overhiring. Goldman Sachs’ analysis, however, finds decomposing the contributions of each factor is not straightforward.
On interest rates, the firm reports scant evidence that elevated rates are the principal cause. To probe this channel, analysts divided a universe of public U.S. technology companies into terciles based on how their interest coverage ratio changed over the rate-hiking cycle. The study found essentially no difference in the pattern of headcount evolution between the top and bottom ICR change terciles. In other words, companies most exposed to changing interest coverage and those least exposed appear to be hiring at broadly similar rates to the sector as a whole.
Goldman Sachs does identify a measurable but limited role for AI in the hiring slowdown. The note estimates that differences in occupational exposure to AI can only explain about 0.5 percentage points of the annual slowdown in tech employment growth since 2022. The firm also observes that AI-related layoff announcements have been credible: the companies that announced cuts attributed to AI actually reduced headcount in AI-related occupations by more than firms that cited other reasons for layoffs.
By contrast, the most compelling signal in the analysis links weak hiring performance to companies that materially increased staff levels during the 2020-2022 period. Goldman Sachs finds that headcount normalisation among these previously overhired firms can readily explain up to 2 percentage points of the slowdown in annual tech employment growth since 2022. Overall, the research concludes that while both AI and post-pandemic normalisation contribute to the hiring gap, the latter is three to four times more significant as an explanatory factor.
Crucially, the firm’s work does not point to a single, distinct shock that triggered the sector-wide deceleration in hiring. Instead, Goldman Sachs suggests that a combination of aggregate pressures and a broad-based slowdown across the sector explains a large share of recent labour market underperformance within technology.
The note also rejects the notion that AI-related layoffs are principally a case of corporate posturing. Given that firms announcing AI-focused reductions subsequently trimmed roles in AI-related occupations more than their peers, the firm sees those announcements as reflective of real adjustments rather than widespread ‘‘AI-washing.’’
In sum, the research frames the tech hiring slowdown as a multi-causal phenomenon in which post-pandemic headcount normalisation plays the dominant role, AI-driven occupational shifts make a minor but detectable contribution, and higher interest rates appear to have little discernible direct impact on hiring patterns across publicly traded U.S. tech companies.
Key takeaways
- Tech-sector hiring slowed roughly 5 percentage points per year since 2022 versus historical trends, with about half of that attributable to normalisation after overhiring in 2020-2022.
- AI occupational exposure explains only about 0.5 percentage points of the annual slowdown, though AI-related layoff announcements correlate with actual reductions in AI-related roles.
- There is little evidence tying the hiring slowdown to higher interest rates; hiring patterns were similar across companies with different changes in interest coverage ratios.
Impacted sectors
- Technology - primary focus given the analysis of public U.S. tech firms
- Labour markets - broader implications for employment dynamics in tech occupations
- Markets and corporate finance - potential implications for firms adjusting cost structures and capital allocation
Risks and uncertainties
- Attribution uncertainty - The firm’s analysis does not identify a single shock responsible for the slowdown, leaving some ambiguity about the precise mix of aggregate and sectoral forces at play. This uncertainty affects how policymakers and investors might respond.
- Measurement limits of AI impact - While occupational AI exposure accounts for a measurable portion of the slowdown, that estimate is limited in size and may not fully capture evolving productivity or role redefinitions within firms.
- Potential for future structural shifts - Continued headcount normalisation could persist if pandemic-era staffing levels remain elevated relative to long-term demand, affecting hiring momentum in the sector.
Note: All figures and conclusions above reflect the findings presented in the firm’s research note.