Divya van Mahajan

Where the Economy Thrives After AI

ai-daily-brief-podcast

Where the Economy Thrives After AI

Overview

This episode of the AI Daily Brief (hosted by NLW) presents and discusses an essay by economist Alex Imas titled “What Will Be Scarce? The Economics of Structural Change and the Post-Commodity Future of Work.” The central thesis is that AI-driven automation will not collapse the labor market but will instead trigger a structural shift—analogous to the agricultural and industrial transitions of the past—toward a “relational sector” where human involvement is itself the core value of goods and services. The episode argues that prevailing catastrophist narratives about AI and unemployment are historically uninformed and practically counterproductive.

Source video URL: not available (AI Daily Brief podcast/video, published 2025-04-26)

Prerequisites

  • Basic understanding of macroeconomics (supply, demand, labor markets, GDP)
  • Familiarity with the concept of creative destruction (Schumpeter)
  • General awareness of the ongoing public debate about AI and job displacement
  • Helpful but not required:
    • Karl Marx’s theory of the commodity form and alienated labor
    • Baumol’s Cost Disease
    • René Girard’s concept of mimetic desire
    • The economics of structural change (agricultural → manufacturing → services transitions)

Main Points

1. The AI Jobs Discourse Is Inverted and Misleading

  • The dominant public narrative spends nearly all its energy on AI’s destructive potential (job loss, unemployment cliffs, permanent underclasses) while treating benefits as vague and hand-wavy.
  • NLW compares this to a pharmaceutical commercial that leads with side effects instead of the cure.
  • Predictions of 15–50% unemployment are, in the host’s view, fundamentally wrong because they fail to model what the economy looks like after AI is fully integrated.
  • Creative destruction always makes destruction visible before creation; this does not mean creation won’t occur.

2. The Core Economic Shift: From Supply Constraints to Demand Constraints

  • AI dramatically reduces the marginal cost of producing many goods and services, eliminating supply-side scarcity.
  • The new binding constraint becomes demand and consumption capacity—particularly time and attention.
  • The key question becomes: In which areas could humans consume significantly more than they currently do?
  • Healthcare is offered as a primary example: most health services consumed today are reactive; an AI-enabled, lower-cost system could unlock vast preventative care demand.

3. The Starbucks Case Study: Automation Rolled Back by Human Value

  • Starbucks spent years automating and streamlining store operations to improve thin margins.
  • The result was customer dissatisfaction; the company reversed course.
  • CEO Brian Niccol cited handwritten cup notes, ceramic cups, comfortable seating, and more baristas as drivers of customer satisfaction and retention.
  • This real-world example illustrates that even in highly standardized commodity production, human presence and hospitality create value that automation cannot replicate.

4. The Commodity Form and Its Limits

  • Pre-industrialization: products were inseparable from their makers; transactions had inherent social components.
  • Industrialization created the commodity form: products abstracted from their producers, enabling disaggregation, global supply chains, and massive scale.
  • Marx described this as alienation; economists described it as an engine of prosperity. Both were correct simultaneously.
  • The prevailing fear is that AI perfects the commodity form—eliminating human labor from production entirely.
  • Imas argues the opposite: AI may trigger the decline of the commodity form as a share of economic activity.

5. Historical Precedent: Structural Change Economics

  • ~40% of the American workforce worked in agriculture around 1900; today it is under 2%. Food production increased dramatically.
  • Workers did not disappear—they moved to manufacturing, then to services.
  • Formal mechanism (Coman, Lashkari, Mesteri, Econometrica 2021): demand is non-homothetic—as people get richer, they shift spending toward sectors with higher income elasticity, not just more of the same goods.
    • Agriculture: low income elasticity (you can only eat so much)
    • Services/relational goods: high income elasticity (there is always a better restaurant, a more attentive doctor)
  • Income effects account for over 75% of observed structural change patterns; price effects account for only ~25%.

6. Baumol’s Cost Disease as a Feature, Not a Bug

  • Baumol’s Cost Disease: productivity gains in one sector make the relatively stagnant sector appear more expensive, causing spending to shift there.
  • In the AI context, this is reframed positively: the relational/human sector becomes relatively more expensive because commodity production gets cheaper, and that relative expense is what sustains employment.
  • The “stagnant” sector—one that resists automation because human presence is the value—absorbs a growing share of spending and jobs.

7. Mimetic Desire and the Relational Sector

  • René Girard’s mimetic desire: people desire objects not only for intrinsic properties but because others desire them; scarcity and exclusivity amplify desire.
  • When basic needs are met, spending increasingly goes toward goods carrying social meaning, provenance, status, and exclusivity.
  • This mimetic component is inherently income elastic: the richer people become, the larger the share of spending going to goods where the identity and humanity of the producer matters.
  • Walter Benjamin’s “aura” concept is invoked: mechanical reproduction destroys the aura of an artwork; AI-generated content similarly feels non-exclusive regardless of stated scarcity (NFTs are cited as a cautionary example).
  • The relational sector includes: nurses, therapists, teachers, boutique fitness instructors, personal chefs, bespoke tailors, craft brewers, live performers, spiritual guides, childcare workers, hospitality workers, and many local service providers.

8. The Post-Commodity Economy and Future Jobs

  • AI does not just perfect the commodity form—it triggers a structural reallocation away from commodity production and toward the relational sector.
  • Mechanism:
    1. AI automates commodity production → prices fall
    2. Real incomes rise
    3. Higher incomes shift demand toward high-income-elasticity relational goods
    4. Baumol’s Cost Disease reinforces the shift as relational goods remain expensive
  • Durable future jobs are not prompt engineers or AI monitors (transitional roles); they are in the relational sector where the human element is the product.
  • Emerging roles mentioned: experience designers, human-AI collaboration artists, provenance certifiers, community curators.
  • Historical comparison: 6 out of 10 jobs held today did not exist in 1940.

9. Important Caveats and Scope Limitations

  • Imas’s framework does not claim labor’s aggregate share must rise or stay constant—it may fall.
  • The claim is about sectoral reallocation: labor remains a substantial, non-zero share of the economy even if its total share declines.
  • The framework works best for developed economies where rising incomes can fund the transition.
  • For developing economies built on commodity production for rich countries, the outlook is more complicated and potentially more concerning.
  • Relational labor may still be sold to capital within fully capitalist relations of production; this is not a claim about the end of commodification in Marx’s sense.

Key Concepts

  • Creative Destruction: The process by which new technologies destroy existing economic structures while simultaneously generating new ones; destruction is typically visible before creation.
  • Commodity Form: The abstraction of a product from its producer, enabling industrial-scale production, global supply chains, and standardization; value lies in the product itself, not in who made it.
  • Structural Change: The historical process by which economies reallocate labor and spending across sectors as productivity in one sector rises dramatically (e.g., agriculture → manufacturing → services).
  • Non-homothetic Demand: The economic property whereby, as incomes rise, consumers shift spending toward qualitatively different goods rather than simply buying more of the same things proportionally.
  • Income Elasticity: A measure of how responsive demand for a good is to changes in income; high-income-elasticity goods see demand grow faster than income itself.
  • Baumol’s Cost Disease: The phenomenon where sectors that resist productivity gains (e.g., live performance, personal care) become relatively more expensive over time as other sectors automate, causing spending to shift toward them.
  • Mimetic Desire (René Girard): The theory that human desire is shaped not only by intrinsic properties of objects but by what other people desire; scarcity and exclusivity amplify desirability.
  • The Relational Sector: Imas’s term for the human-intensive, provenance-rich portion of the economy where the human element is inseparable from the value of the good or service (care, education, craft, hospitality, arts, community).
  • Aura (Walter Benjamin): The quality of authenticity and uniqueness possessed by an original work of art that is destroyed by mechanical reproduction; extended here to human-made goods more broadly.
  • Provenance: The documented origin and human history of a good or service, which carries economic value in contexts where authenticity matters.
  • Intolerable Abundance (Herbert Simon): The theoretical scenario in which automation eliminates labor scarcity so thoroughly that existing social structures for income distribution and democratic stability break down.

Summary

The episode’s central argument, developed through Alex Imas’s essay and NLW’s commentary, is that AI will not produce mass permanent unemployment but will instead accelerate a historical pattern of structural economic transformation. Just as agricultural mechanization moved workers into manufacturing and then into services, AI-driven automation of commodity production will raise real incomes, and rising incomes will shift consumer demand toward high-income-elasticity “relational” goods and services—those where human presence, judgment, warmth, provenance, and social meaning are inseparable from the product’s value. Baumol’s Cost Disease reinforces this shift, making the human sector relatively more expensive even as it absorbs a growing share of spending and employment. Mimetic desire—the human tendency to want what others want and to prize what others cannot have—ensures that relational goods maintain inherently high income elasticity, preventing demand collapse. The durable jobs of the AI era will therefore be found not in monitoring algorithms, but in nursing, teaching, therapy, craft, hospitality, care, and as-yet-uninvented relational roles. NLW closes by arguing that the current climate of catastrophist anxiety reflects a collective failure to model what the economy actually transforms into, and that engaging seriously with frameworks like Imas’s is an intellectual and civic necessity.