Artificial Intelligence and the Economics of Adjustment

Source details

Canonical URL

Artificial Intelligence and the Economics of Adjustment

Other formats

• Markdown version
• Structured JSON version
• Bundle manifest

Bibliographic details

• Authors: DAN KATZ
• Published: June 25, 2026

Overview

• Author: DAN KATZ, first deputy managing director of the IMF.
• Published on June 25, 2026.
• Central claim: The aggregate labor-market impact of AI depends critically on how economies adjust; policies that facilitate reallocation and diffusion will determine whether AI raises growth and living standards or exacerbates disruption and inequality.

Historical evidence on technological shocks

• New general‑purpose technologies (steam engine, electrification, computing, internet) disrupted jobs and skills but ultimately raised productivity, lowered prices, increased real incomes, and supported higher employment.
• Technological adoption often required diffusion, complementary investment, and institutional adaptation; gains were "staggered and gradual."
• A systematic review published in 2023, covering more than 100 studies, found that the labor‑displacing effect of technology is more than offset by labor creation.
• Jobs evolve via:
  • Workers moving across occupations.
  • Shifts in skills required within occupations — example statistic: roughly 1 in 10 job vacancies in advanced economies now lists a new skill.

Expected impacts of AI on labor markets and the economy

• Mechanisms:
  • AI reduces the cost of analysis, prediction, communication, coordination, and action, making services cheaper and more scalable.
  • Lower costs → lower prices → higher demand → higher output and employment (Jevons paradox observed across sectors).
• Net impact likely to be positive if productivity gains reduce costs, expand demand, and create new tasks and firms sufficient to offset job losses in some categories.
• IMF staff estimate: approximately 40 percent of jobs globally could be affected by AI in some way (changed tasks, skill requirements, organizational structure), not necessarily eliminated.
• Emerging evidence suggests near-term gains may be strongest for high‑ and low‑skill workers, while demand for middle‑skill and entry‑level positions could weaken.
• Transition caveats: adjustment may be disruptive, uneven, and politically consequential.

Key uncertainties and risks

• Unknowns:
  • What the new work will look like, who will do it, and where it will be located.
  • Long-term trajectory of labor demand is "exceptionally uncertain."
• Labor market churn context (example, U.S.):
  • Total nonfarm employment is roughly 160 million.
  • Approximately 60 million hires and 60 million separations every year.
• Potential for an "Engels’ pause": a period where new technologies diffuse but wages or broad-based gains lag, leading to uneven distributional outcomes and political backlash.
• Policy-driven risks:
  • Policies that slow adjustment (protecting specific jobs, firms, or sectors) can delay reallocation, reduce productivity growth, and worsen labor-market outcomes.
  • Concentration of AI productivity gains in advanced economies could widen income gaps and concentrate economic rents geographically, redirecting capital away from lower‑income countries.
• Structural constraints in some developing economies that amplify risks:
  • Slow labor reallocation, barriers to firm entry/exit, limited access to financing, weak legal systems, ill-defined property rights.

Policy recommendations to facilitate adjustment

• Overarching principle: Favor structural policies that facilitate adjustment rather than prevent it.
• Recommended policy areas:
  • Labor market policies that support mobility and reemployment rather than preserve existing job structures designed for cyclical shocks.
  • Product market policies that promote competition and reduce barriers to entry to avoid regulatory capture and excessive market concentration.
  • Financial and legal frameworks that allow capital and assets to be reallocated efficiently and productively, including effective bankruptcy and restructuring frameworks.
• Active labor market policies and retraining:
  • Retraining and upskilling programs should be widely accessible and designed to maintain private sector incentives for businesses and employees.
  • Caution: historically mixed track record of many active labor market policies; governments may lack necessary information, incentives, and institutional capacity in fast‑moving environments.
  • Opportunity: AI can upgrade active labor market policies and the employment services industry via personalized education and reduced information frictions.
• Regulation of AI:
  • Apply guardrails where risks and externalities are clear (example areas: cybersecurity, protection of children).
  • Presumption against protective action unless strong evidence of harm or clear risks.
  • Prefer regulatory approaches that provide permission structures and regulatory certainty (especially in heavily regulated sectors like health care and finance) to enable productivity benefits.
  • Regulation should promote business dynamism and avoid erecting barriers that concentrate market power.

Recommendations for developing economies

• AI as an opportunity and a risk:
  • Opportunities: leapfrogging via digital service delivery, AI‑enabled diagnostics, automated compliance tools, improved tax administration, customs, and social protection delivery.
  • Risks: widening income gaps if advanced economies realize sustained productivity gains first; geographic concentration of economic rents; capital flowing "uphill" away from lower‑income countries.
• Policy tailoring by preparedness:
  • More developed and better‑prepared economies: focus on innovation and diffusion through R&D investment, improved access to financing, and a business environment that fosters innovative firms—plus regulatory frameworks that enable safe and widespread AI use.
  • Lower‑income and less‑prepared economies: prioritize building digital infrastructure (especially reliable and affordable power generation) and education with emphasis on earlier attachment to the labor market, lifelong learning, and skills that complement rather than compete with technology.
• Design investments to safeguard fiscal sustainability and align with absorptive capacity.

Role of the IMF

• Institutional priorities:
  • Strengthen surveillance of AI‑related structural changes.
  • Support members in designing reform strategies and assessing trade‑offs.
  • Facilitate international cooperation and knowledge sharing on AI‑related best practices.
• Internal use:
  • Deploy AI to enhance IMF operations—doing more with the same resources, improving the quality and speed of analysis, and increasing value delivered to membership.

Key statistics and figures (verbatim)

• Publication date: June 25, 2026.
• Systematic review year: 2023 (covering more than 100 studies).
• Roughly 1 in 10 job vacancies in advanced economies lists a new skill.
• IMF staff estimate: approximately 40 percent of jobs globally could be affected by AI in some way.
• United States: total nonfarm employment is roughly 160 million; approximately 60 million hires and 60 million separations every year.

Source: Artificial Intelligence and the Economics of Adjustment, DAN KATZ, F&D Magazine, Published June 25, 2026.

Content in this bundle

• Staff Discussion Note
  • Staff Discussion Note (Markdown version)
  • Staff Discussion Note (PDF)