The potential impact of AI on productivity and economic growth is difficult to pin down and subject to great debate in academic and industry literature. While overall gains are acknowledged, the range of projections is wide. As examples, Daron Acemoglu, an economics professor at the Massachusetts Institute of Technology, argues that forecasting AI’s effect is hard to predict and requires speculation. Based on his framework, he sees a gain of 0.55% in total factor productivity over the next 10 years.1 In contrast, Goldman Sachs’ Joseph Briggs predicts a 9.2% gain in total factor productivity and 6.1% increase in economic growth over the next decade.2
Regardless of the wide range of estimates, AI’s asymmetrical profile is clear in our view, with far more potential for upside gain than downside loss. And while shorter term gains may come in spurts, or be obfuscated by other data, we believe AI adoption and gains will likely be more evident long term, as the technology is adopted and use cases are refined.
That said, AI could cause displacement in jobs and increase inequality. Potential AI productivity gains will likely differ across developed and developing economies due to the nature of existing work and respective states of technology. Within countries, impacts could also differ based on factors such as firm size, industry and adoption. Likewise, impacts on the labor market (acceleration, augmentation, automation) could vary.
As we look ahead to opportunities in an AI-Enabled Productivity world, we see potential winners and losers. Infrastructure (data centers, energy infrastructure) seems attractive, with potential opportunities to capitalize on this theme in private markets. Meanwhile, some of the losers may be highly indebted companies lacking the sufficient access to credit necessary to invest enough in AI in order to remain competitive.
About the CMA Process
Every year, Northern Trust’s Capital Market Assumptions (CMA) Working Group gathers to develop long-term financial market forecasts. The team adheres to a forward-looking, historically aware approach. This involves understanding historical relationships between asset classes and the drivers of those asset class returns, but also debating how these relationships will evolve in the future.
Our forward-looking views are encapsulated in our annual list of CMA themes, which — combined with our quantitative analysis — guide our expectations for long-term asset class returns. The CMA return forecasts are combined with other portfolio construction tools (standard deviation, correlation, etc.) to annually review and/or update the recommended strategic asset allocations for all Northern Trust managed portfolios and multi-asset class products.
The CMA Working Group is composed of senior professionals from across Northern Trust globally, including top-down investment strategists, bottom-up research analysts and client-facing investment professionals.
1 The Simple Macroeconomics of AI” by Daron Acemoglu, Massachusetts Institute of Technology, April 5, 2024. Total factor productivity is a measure of product efficiency, calculated by dividing total production (output) by average costs (inputs).
2 Goldman Sachs Global Investment research, Top of Mind Issue 129: Gen AI: Too Much Spend, Too Little Benefit?, June 2024.