reference

The Projected Impact of Generative AI on Future Productivity Growth

Penn Wharton Budget Model analysis projecting AI's impact on total factor productivity (TFP). Estimates 42% of jobs exposed to AI automation, but only 23% of exposed tasks will be profitably automated. Projects modest TFP gains (0.01pp in 2025) due to slow adoption, scaling to ~10-15% GDP impact long-term.

View original source ↗

Source Information

  • Publisher: Penn Wharton Budget Model
  • Date: September 10, 2025
  • Type: Economic research/working paper
  • Framework: Based on Acemoglu (2024) task-level automation model

Key Findings

Exposure vs. Automation

  • 42% of current jobs have ≥50% of tasks potentially exposed to AI automation
  • Only 23% of exposed tasks will eventually be profitably automated
  • ~10% of current GDP likely to be impacted over time
  • Projected to grow to ~15% over next two decades

Cost Savings Estimates

Based on real-world AI adoption studies:

  • Range: 10-55% labor cost savings
  • Average: ~25% on tasks where AI is adopted

Productivity Impact Timeline

Year TFP Contribution
2025 0.01 pp
2030 ~0.1 pp (projected)
2045 Peak impact

Why Modest Near-Term Impact?

  1. Most businesses have not yet adopted AI at scale
  2. Adoption is probabilistic and gradual
  3. Only portions of roles automated, not whole jobs
  4. Integration and change management take time

Employment Effects Already Visible

Jobs with highest AI exposure (90-100% of tasks automatable):

  • Employment fell sharply in 2024
  • 0.75% lower than 2021 levels

Jobs with high AI exposure (90-99%):

  • Employment growth slowed significantly since 2022

Caveats Noted by Authors

The analysis does NOT account for:

  • AI-driven changes in product quality
  • Emergence of new products and labor tasks
  • AI's potential impact on innovation/TFP feedback loops

Relevance to MPDrexel

This research provides macroeconomic context for:

  1. The Efficiency Trap (The 2026 Jobless Boom & The Efficiency Trap): Efficiency gains alone yield modest macro impact
  2. The Case for Expansion: The "new products and labor tasks" excluded from this model represent the Growth Premium opportunity
  3. Realistic client expectations: AI efficiency gains are real but incremental; transformation value comes from new value creation