The most significant labor arbitrage today is not in low-skilled factory work but in high-skilled professional services. Raghuram Rajan highlights that a top Indian MBA costs one-fifth of a U.S. equivalent. This massive cost differential, combined with remote work, makes countries like India a hub for high-value service exports.
Related Insights
The decline in U.S. manufacturing isn't just about labor costs. A crucial, overlooked factor is the disparity in savings. While Americans consumed, nations like China saved and invested in capital goods like factories, making their labor more productive and thus more attractive for manufacturing investment.
Bringing manufacturing back to the US won't mean a return of old assembly line jobs. The real opportunity is to leapfrog to automated factories that produce sophisticated, tech-infused products. This creates a new class of higher-skill, higher-pay "blue collar plus" jobs focused on building and maintaining these advanced manufacturing systems.
While tariffs affect goods prices, immigration controls are reducing the labor supply, particularly in the service sector. This creates upward wage and price pressure on services, a subtle but significant contributor to overall inflation that is difficult to isolate in real-time data.
The primary economic incentive driving AI development is not replacing software, but automating the vastly larger human labor market. This includes high-skill jobs like accountants, lawyers, and auditors, representing a multi-trillion dollar opportunity that dwarfs the SaaS industry and dictates where investment will flow.
Wage stagnation is not accidental but a result of two concurrent policies. By sending manufacturing jobs overseas and simultaneously bringing in low-wage labor, corporations create a market where domestic workers lose nearly all leverage to demand higher pay for remaining jobs.
Just as NAFTA brought cheap goods but eliminated manufacturing jobs, AI will create immense productivity via a new class of "digital immigrants" (AIs in data centers). This will generate abundance and cheap digital services but risks displacing vast swaths of cognitive labor and concentrating wealth.
As agencies adopt AI to increase efficiency, clients will rightfully question traditional pricing models based on billable hours. This creates an "arbitrage" problem, forcing agencies to redefine and justify their value based on strategic insight and outcomes, not just the labor involved.
A powerful mental model for the future of work is a three-step pipeline. If a job can be done remotely in a high-cost country, it can be offshored to a low-cost one. Once offshored and process-driven, it becomes a prime target for AI automation. This positions remote work as a transitional phase, not an endpoint.
The immense salaries in software and finance may create a 'talent Dutch disease,' pulling the brightest minds from crucial fields like structural engineering. This reallocation of human capital could explain why productivity has stagnated or declined in industries that build the physical world.
Data from 2004-2023 reveals low unemployment in occupations that heavily utilize H-1B visas, such as tech and engineering. This suggests that foreign workers are filling a talent gap rather than displacing a large number of available American workers, challenging the narrative that immigration is a primary cause of job loss in these sectors.
