Economists Confront AI’s Uncertainty in Economic Forecasts

The rise of artificial intelligence (AI) presents a significant challenge for economists as they attempt to incorporate this transformative technology into their economic forecasts. According to Erik Lundh, a senior global economist at The Conference Board, traditional economic models struggle to account for the unpredictable effects of AI on productivity and economic growth. While some impacts are evident, such as increased productivity and economic activity, the complexities surrounding AI make it difficult to quantify its overall impact on labor markets and investment.

Economists have established methods for projecting economic growth, but AI complicates these calculations. With automation increasingly taking over tasks traditionally performed by humans, isolating AI’s effects on employment becomes problematic. For instance, The Conference Board projects that U.S. GDP will grow at an average rate of 1.9% annually from 2025 to 2039, a decrease from 2.4% growth observed from 2000 to 2024. Lundh notes that AI is expected to mitigate some of this slowdown by enhancing productivity.

Despite the optimistic projections, Lundh acknowledges the uncertainty in measuring AI’s contributions. The Peterson Institute for International Economics forecasts a slowdown in global GDP by 2026, suggesting that while AI may provide some offset, it remains unclear how significant that impact will be. Lundh emphasizes that the integration of AI into economic models must consider both established variables, such as total-factor productivity, and the unpredictable nature of AI’s influence.

As AI technologies evolve, the way in which growth is measured may also change. Lundh highlights the dual effects of AI on capital investment and productivity. The construction of data centers and power plants to support AI infrastructure contributes to immediate economic activity, while the long-term productivity gains from these investments are still uncertain. He likens the situation to infrastructure projects, where initial costs lead to greater efficiency over time.

The potential pathways of AI’s influence on productivity can be categorized into two outcomes: increased output for the same input or maintaining output with fewer inputs. This creates uncertainty for businesses regarding workforce management. Companies might choose to reduce staff while maintaining revenue through AI, or they could opt to integrate AI to enhance productivity without cutting jobs. Lundh points out that research and development (R&D) spending could also be affected by AI, with the technology either streamlining processes or prompting increased investment due to higher returns.

The United States currently leads in AI investment compared to other nations, which influences Lundh’s productivity projections. He notes that while U.S. productivity is anticipated to rise, countries like China are also increasing their productivity projections due to substantial investments in AI. China’s upcoming five-year plan emphasizes developing advanced manufacturing capabilities and next-generation technologies, including AI. However, various factors, such as access to high-tech components and geopolitical considerations, will shape the effectiveness of these investments.

For developing economies like Vietnam, Bangladesh, and Kenya, AI poses both challenges and opportunities. Traditionally, these nations benefited from lower labor costs to attract manufacturing investments. However, as AI reduces the reliance on human labor, this cost advantage may diminish, complicating the path to industrialization. Nonetheless, businesses in these regions can leverage AI tools to improve efficiency, presenting a potential advantage in a rapidly changing landscape.

Venture capitalists express a desire to fund AI initiatives that have the potential to transform industries rather than merely improve existing software tools. Lundh acknowledges this perspective but emphasizes that the immediate impacts of AI will likely manifest in the services sector, given its dominant share of the U.S. economy. Innovations such as AI-driven call centers and automated accounting services represent significant disruptions in this area.

As AI technology continues to evolve, Lundh reflects on the uncertainties surrounding its future impact on economic analysis. He recognizes that rapid advancements in AI are reshaping the landscape for economists, potentially altering the profession itself. “It’s a wild future,” he states, underscoring the unpredictable nature of the economic environment shaped by AI.

The integration of AI into economic forecasting represents both a challenge and an opportunity for economists. As they navigate this complex terrain, continuous assessment and adaptation of traditional models will be essential to understand AI’s long-term implications on productivity and growth.