For decades, economists and business leaders have puzzled over a contradiction: why has the digital revolution, despite transforming how we live and work, failed to produce the kind of sweeping productivity gains seen in past technological shifts? This enduring question, often called the “productivity paradox,” is now at the center of debate in the era of artificial intelligence.
Bank of America believes it has found an answer: replace people with processes.
The Productivity Puzzle
In theory, technological advances should drive economic growth by enabling people to do more with less. From the Industrial Revolution’s steam engines to the post-war boom powered by electrification, productivity gains have historically followed waves of innovation.
But since the 1970s, productivity growth in the U.S. and other advanced economies has slowed sharply, even as computers, the internet, and automation spread across every industry. The paradox has led to competing explanations—from measurement problems to the lag between innovation and adoption.
Now, in a world reshaped by AI and machine learning, the question looms larger than ever: will today’s technologies finally deliver?
BofA’s Thesis: Processes, Not People
Bank of America analysts argue that the breakthrough lies not in simply augmenting workers, but in rethinking work altogether. As they put it: “A process is almost free, and it’s replicable for eternity.”
Instead of focusing on human productivity per hour worked, the vision is to automate entire workflows. If a digital process—say, invoice reconciliation or fraud detection—can be codified, then scaled across millions of instances, the marginal cost approaches zero. Unlike people, processes don’t fatigue, demand wages, or require retraining.
This is not just about efficiency gains but about fundamentally redefining the unit of productivity itself. Where once businesses measured output per worker, in the future they may measure output per process.
AI as the Enabler
Artificial intelligence is central to this transformation. Advances in natural language processing, predictive analytics, and robotic process automation allow firms to design digital processes that were once impossible to fully automate.
- Customer service is shifting from human representatives to AI-powered chatbots that handle millions of queries simultaneously.
- Financial compliance checks that once demanded entire back-office teams are increasingly run by algorithms.
- Supply chain management is being re-engineered through predictive models that preempt disruptions.
The idea, according to BofA, is not to replace every human task but to replace categories of work with processes that scale indefinitely.
Implications for Labor
The vision is both promising and unsettling. On one hand, widespread process automation could finally break the productivity paradox, ushering in an era of sustained economic growth. On the other, it raises pressing questions about employment, inequality, and social stability.
If processes replace people at scale, millions of workers could be displaced, particularly in middle-skill, routine roles. Proponents argue that new kinds of jobs will emerge, as they have in past technological revolutions, but the transition could be rocky and unevenly distributed.
The Cost Argument
From an economic standpoint, the appeal is undeniable. Once designed, digital processes require minimal upkeep. They are endlessly replicable, immune to human error, and scalable across geographies. For corporations seeking to improve margins in a competitive global economy, process automation offers a near-ideal solution.
As BofA notes, “a process is almost free.” Compare that to the costs of recruitment, training, salaries, and benefits, and the business case for automation is stark.
Risks and Limitations
Skeptics caution that replacing people with processes is not without risk:
- Complexity: Not all tasks can be neatly codified into processes. Creativity, empathy, and judgment remain stubbornly human strengths.
- Resilience: Over-reliance on automated processes can create vulnerabilities, as system failures or cyberattacks could paralyze operations.
- Societal costs: A sharp displacement of workers could widen inequality and strain social safety nets.
Critics also point out that productivity gains are not automatically shared. The benefits of past automation waves often flowed disproportionately to capital owners rather than labor.
Toward a Process-First Economy
Despite these concerns, momentum is building. Corporations across sectors are piloting process-first models, governments are investing in digital transformation, and investors are betting that AI-powered automation will reshape balance sheets.
If BofA is correct, the productivity paradox may not be solved by making people marginally more efficient but by systematically reducing reliance on people altogether. Productivity, in this framing, is no longer about how much a worker can do in an hour but how many processes can run in parallel, endlessly, at nearly zero cost.
Conclusion
The idea that processes can replace people as the cornerstone of productivity marks a profound shift in economic thinking. If successful, it could herald a new age of efficiency, profitability, and growth. But it also risks deep social disruption if the gains are not managed inclusively.
Bank of America’s thesis reframes the paradox of productivity in the age of AI: perhaps the problem has never been the technology itself but the assumption that people must remain at the center of productivity. In a world of replicable, scalable processes, the very definition of work may be changing—forever.
