AI's Slow Economic Impact Due to Adoption Friction

This conversation on Bloomberg Surveillance, featuring former Fed Vice Chair Richard Clarida, and Francois Trahan of BMO Capital Markets, delves into the nuanced realities of economic forecasting and the adoption of technological advancements like AI. The core thesis is that conventional economic models and optimistic projections often fail to account for the friction and distributed adoption rates that characterize real-world implementation. The hidden consequence revealed is the significant lag between theoretical potential and practical impact, particularly concerning AI's role in productivity and economic growth. This analysis is crucial for investors, business leaders, and policymakers who need to understand the temporal and structural barriers to widespread technological benefit, offering them a more grounded perspective to navigate future economic landscapes and avoid premature bets on transformative technologies.

The Lag Between AI's Promise and Economic Reality

The prevailing narrative around Artificial Intelligence often paints a picture of imminent, widespread productivity gains that will fuel non-inflationary growth. This optimistic outlook, however, overlooks a critical systemic dynamic: the pace and breadth of adoption. Francois Trahan of BMO Capital Markets highlights that while AI's potential is undeniable, its current impact is largely confined to very large corporations. This is a crucial point because, as Trahan notes, "Small business America creates three quarters of jobs in the US." The implication is stark: until AI adoption becomes pervasive across the vast majority of businesses, its aggregate effect on productivity and the economy will remain muted.

This isn't a critique of AI's capabilities, but a systems-level observation of how technology diffuses through an economy. The immediate benefit is captured by early adopters, creating a visible, albeit narrow, success story. However, the downstream effect--the broad-based economic uplift--is contingent on a much slower, more complex process of integration across smaller enterprises. This creates a temporal disconnect: the theoretical promise of AI-driven productivity is often projected as an immediate reality, ignoring the years it will take for this technology to permeate the economic fabric and truly impact job creation and inflation.

"People say, 'Alright, we can't have non-inflationary growth because we don't have excess capacity.' Why can't AI play that role with productivity? That sounds logical when you think about it. When you dig into the data, you realize that it's not feasible in 2026."

This quote underscores the analytical rigor required to move beyond surface-level logic. The "logical" assumption that AI will automatically fill capacity gaps fails when confronted with the empirical reality of adoption rates. The consequence is a miscalculation of future economic conditions, potentially leading to misallocation of resources or misguided policy decisions based on an overestimation of AI's near-term impact. The advantage for those who grasp this nuance lies in their ability to make more durable investment and strategic decisions, avoiding the hype cycle and focusing on the longer-term, more gradual integration of AI.

The Limits of Models and the Uniqueness of Individual Views

The conversation also touches upon the limitations of economic models and the influence of individual perspectives within institutions like the Federal Reserve. Richard Clarida, former Vice Chairman of the Fed, discusses Governor Waller's view on the neutral interest rate. Waller's belief that the neutral rate is significantly lower than the committee's consensus is presented as a minority view. Clarida's framing of this as a "delicate answer, nicely said" hints at the internal dynamics and differing philosophies within policy-making bodies.

While models are essential tools, Clarida points out they are "backward looking." This is a critical insight into the challenges of forecasting and policy-making in dynamic environments. The neutral interest rate, for instance, is a theoretical construct that is difficult to measure precisely and can shift over time. Governor Waller's individual conviction, while not currently the committee's consensus, represents a potential divergence that could influence future policy discussions.

"Models are tools, but they're backward looking. And in particular, I think often times the argument that Governor Waller has made is he has a personal individual belief that the neutral interest rate in the US is well below the current level and well below where the committee believes."

This highlights a key systemic tension: the reliance on historical data and established models versus the need to account for evolving economic conditions and unique individual insights. The immediate consequence of such differing views is a degree of uncertainty in policy direction. The longer-term payoff, however, comes from acknowledging and debating these diverse perspectives. It forces a more robust examination of assumptions underlying the models and can lead to more adaptive policy responses. For market participants, understanding these internal debates within the Fed provides a competitive advantage, allowing them to anticipate potential shifts in monetary policy that might not be immediately apparent from consensus forecasts. The "conventional wisdom" here might be to trust the aggregated model, but the deeper insight is to recognize the power of dissenting, experience-based views, even if they are currently in the minority.

Actionable Takeaways

• Immediate Action: Re-evaluate AI investment timelines. Instead of expecting widespread productivity gains by 2026, focus on how AI can solve immediate, specific problems within your organization, acknowledging that broad economic impact will take longer.
• Longer-Term Investment: Develop strategies to support AI adoption within small and medium-sized businesses, as this segment represents the largest portion of job creation and future economic growth. This pays off in 3-5 years as adoption accelerates.
• Discomfort Now, Advantage Later: Resist the urge to chase the "AI hype." Focus on foundational business processes that can be enhanced by AI, rather than betting solely on speculative, large-scale AI applications. This requires patience, which is a competitive advantage.
• Policy Awareness: Monitor dissenting views within central banks, such as Governor Waller's perspective on the neutral rate. These minority opinions can signal future shifts in policy, offering an advantage in anticipating market movements.
• Model Skepticism: Understand that economic models are based on historical data and may not fully capture the nuances of emerging technologies or unique economic conditions. Build flexibility into your own forecasting.
• Strategic Patience: Recognize that true technological transformation, like AI adoption across all business sizes, is a marathon, not a sprint. Prioritize sustainable integration over rapid, potentially superficial implementation.
• Focus on Implementation: For large companies already leveraging AI, the focus should shift from initial adoption to deep integration across HR, IT, and procurement to realize cost savings and strategic advantages, as IBM suggests. This is an ongoing investment, yielding returns over many quarters.