The future of inflation: AI's impact on wages and markets
The world of finance is abuzz with the potential implications of AI on the labor market and inflation. In a recent note to clients, Barclays' Head of Inflation Research Strategy, Jonathan Hill, delves into the complex relationship between AI, wages, and inflation. Hill's analysis highlights a growing consensus that US inflation dynamics are returning to their pre-COVID state, but with a crucial twist: the rise of generative and agentic AI.
The core argument revolves around the potential for AI to disrupt the labor market, leading to wage deflation. Hill explains that before the pandemic, the US core CPI ecosystem was stable, with core goods inflation around 0%, rent inflation at 3.0-3.5%, and core services ex-rents averaging 2%. However, with the fading of tariff pass-through and core goods returning to near-zero inflation in H2 2026, the focus shifts to rent and services. The data suggests rent inflation will be lower than pre-COVID, and with core goods inflation flat, core services ex-rents must rise to maintain the 2% target.
Here's where AI comes into play. The rapid advancements in AI model capability and the emergence of agentic architectures capable of multi-step tasks have shifted the labor market's probability distribution. Hill references a viral blog post that captures a widely held intuition: AI might not just substitute tasks but entire workflows. This is evident in corporate actions like Block's layoffs, explicitly tied to automation.
The author argues that dismissing AI's impact as unproven in labor data is a mistake. By the time labor market stress appears, repricing will have already occurred. The argument that AI tools are not yet mature is also challenged, drawing a parallel to a comedic GPS error in The Office. Observed improvement curves and roadmaps indicate a high probability of significant AI-driven disruption, which could impact wage momentum.
The implications for inflation are profound. If AI softens labor demand before boosting productivity, inflation risks are lowered. However, at some point, AI should become a positive productivity shock, leading to disinflation in the medium term. Yet, current inflation swaps remain elevated, and real rate swaps only briefly dipped below 1%. This suggests that investors may not fully grasp the potential of AI-driven disinflation.
The author concludes by emphasizing the need to consider the rising probability of AI disruption on inflation outcomes. The evidence is mounting, and AI's impact on wages and markets is a critical question that investors and policymakers must address. As AI continues to evolve, its influence on the economy and financial markets will only grow, making it a topic of utmost importance.
