1.

A barber cuts no more hair in 1948 than in 1900 — yet his wages rose anyway. That single fact, observed by Jean Fourastié in 1949, became the most powerful explanation for why healthcare, education, and government keep getting more expensive, and why the economy itself may be slowing down.

2.

Wages equalize across sectors — even when productivity doesn't

• If the auto industry raises wages because robots doubled output per worker, retail managers quit for the factory floor — forcing retail to match pay without a single productivity gain
• Baumol's 1967 model proves the math: in a two-sector economy with mobile labor, wages must converge to the level of the high-productivity sector, or the stagnant sector loses every worker
• This is cross-elasticity of demand, not inefficiency — the stagnant sector is raising wages defensively, not wastefully

3.

Costs in stagnant sectors rise exponentially and without limit

• Unit cost equals wage divided by output. In the progressive sector, both rise at rate r, so unit cost stays constant. In the stagnant sector, wages rise at rate r while output stays flat — unit cost grows exponentially
• Nordhaus's 2008 study of 1948–2001 US data: "productivity trends are associated almost percentage-point for percentage-point with price decline" — low-productivity industries saw relative prices climb
• The chart at the top of the article shows the prediction in action: between 1998 and 2018, services became more expensive while manufactured goods became cheaper

4.

The economy shifts toward the sectors that can't get cheaper

• Baumol's thought experiment: if the physical output ratio of progressive to stagnant sectors holds constant, and the progressive sector needs fewer workers per unit, labor must flow from progressive to stagnant
• US agriculture followed this path: 21.5% of the workforce in 1930, 1.9% by 2000 — productivity gains in farming freed workers for services
• The post-industrial economy isn't a choice; it's a mathematical consequence of unbalanced growth

5.

Government spending is disproportionately hit because it buys services

• Most government budgets fund healthcare, education, and law enforcement — all sectors where productivity growth is persistently slow
• As stagnant-sector costs rise, the same services consume a larger share of public funds, crowding out other priorities
• Baumol himself warned in 1993: "Health care, education and the cost disease: A looming crisis for public choice"

6.

Aggregate growth slows as the workforce moves to stagnant sectors

• Baumol's original prediction: shifting labor from high-productivity to low-productivity industries reduces the economy's overall productivity growth rate, which drives economic growth
• Oulton's 2001 counter-argument: if services provide intermediate inputs to manufacturing (e.g., business services), the stagnant sector's growth could boost the progressive sector's productivity, raising aggregate growth
• Sasaki's 2007 model and Hartwig & Krämer's 2019 data analysis both rejected Oulton — the shift to services still decreases long-run growth

7.

The disease doesn't make services unaffordable to society — but it does to individuals

• Baumol's own healthcare thought experiment: if real national income doubles over 50 years, healthcare spending can rise 500% and the amount of income left for other purchases stays exactly the same
• The caveat is devastating: "the increase in costs disproportionally affects the poor" — individual wages may not keep pace with stagnant-sector price increases, especially with rising inequality
• Manufactured goods get cheaper, but the services that define quality of life — healthcare, education, legal help — become less accessible to those at the bottom

8.

The disease may not exist in the sectors where it matters most

• Rhoades and Frye found in 2015 that "relative academic labor costs have gone down as tuition has gone up" — the Baumol mechanism isn't operating in education, yet tuition still exploded
• Rossen and Faroque's 2016 Canadian healthcare study concluded the cost disease "is a relatively minor contributor" — technical progress and per capita income growth dominate
• If the two flagship applications — education and healthcare — show mixed or negative evidence, the theory's empirical reach may be far narrower than its intellectual footprint suggests

9.

The disease isn't a bug in the economy — it's the price of being human

• Both the original argument and the counter-argument share an unstated assumption: that stagnant productivity in services is a problem to be solved, a cost to be contained, a disease to be cured
• But the services most afflicted — healthcare, education, performing arts, law enforcement — are precisely those where the quality of the output is inseparable from the labor input. A nurse cannot change a bandage faster without changing the bandage less well. A professor cannot mark an essay faster without marking it less carefully. A musician cannot play a Schubert quartet faster without playing a different piece of music
• The Baumol effect doesn't reveal a flaw in the system. It reveals the irreducible cost of human connection, care, and craft — and the real question isn't how to eliminate that cost, but how to distribute it fairly in a world where everything else gets cheaper