Baumol's cost disease, also called the Baumol effect, is the tendency for the costs and prices of labor-intensive services with little scope for productivity growth, such as live performance, education, health care, policing and personal care, to rise faster than prices in general over long periods. It was identified by economists William Baumol and William Bowen in Performing Arts: The Economic Dilemma (1966) and set out formally in Baumol's 1967 article "Macroeconomics of Unbalanced Growth" in the American Economic Review.
The mechanism is simple. Productivity rises quickly in manufacturing and some other sectors, pushing up wages there. Because labor can move between jobs, wages in slow-productivity services must rise too, or those sectors would lose staff. Because output per worker in those services barely changes, their costs and prices climb relative to goods. The effect matters for measuring inflation with indexes such as the consumer price index, for public budgets, and for long-run growth in services-heavy economies.
Baumol's cost disease in plain terms
A barber today needs about as long to cut someone's hair as a barber did seventy years ago. A car factory today turns out many more cars per worker than it did then. Yet barbers' pay has risen roughly in line with everyone else's, because otherwise young people would choose other work.
So the price of a haircut has risen much more than the price of a car or a television. Nothing has gone wrong in the barbershop. The economy as a whole has become more productive, and the services that cannot easily speed up end up costing more in comparison.
How the cost disease works
Consider two sectors, a "progressive" one where output per worker grows and a "stagnant" one where it does not. If both pay the same wage, the price of each sector's output reflects its unit labor cost, the wage divided by output per worker. The relative price is:
- and are the prices of the stagnant service and the manufactured good.
- is the common wage.
- and are output per worker in each sector.
If output per worker in manufacturing grows at 3% a year and in the service at zero, the relative price of the service grows at about 3% a year. Worked example: after 25 years, , so the service costs roughly twice as much, relative to the manufactured good, as it did at the start. Nothing in this calculation requires waste or market failure; it follows from uneven technological progress.
Evidence
US price data from the Bureau of Labor Statistics show a long-running split: prices of services such as hospital care, college tuition and childcare have risen much faster than the overall CPI, while prices of televisions, software and toys have fallen. William Nordhaus, in a 2008 study of US industries, found that sectors with slower productivity growth did see faster-rising relative prices, consistent with Baumol's prediction. Across countries, the same logic helps explain the Balassa–Samuelson effect: services are cheap in poorer countries because wages there are low, which is one reason purchasing power parity adjustments raise the measured incomes of developing countries.
Implications for public finances and growth
Because governments pay for much of education, health care and public safety, the cost disease pushes up public spending as a share of GDP even if service levels stay the same. This puts pressure on fiscal policy and can be misread as pure inefficiency. As spending shifts toward slow-productivity services, overall productivity growth can also slow, a point Baumol called the "growth disease". In his 2012 book The Cost Disease, Baumol argued that rising productivity elsewhere means societies can still afford these services, since total income keeps growing.
Criticism and limits
Service productivity is not fixed. Recording and broadcasting changed music, and online learning, telemedicine and artificial intelligence may change teaching and health care. Measuring the output of services is hard, so some measured "stagnation" may reflect poor statistics, including incomplete hedonic adjustment for quality. Critics also point to other drivers of rising costs in health care and education, such as administrative growth, regulation, subsidies that raise demand, and rent-seeking by providers. Alex Tabarrok and Eric Helland, in a 2019 study, concluded that the Baumol effect explains much of the rise in US education and health costs, while others argue it explains only part. The internet economy offers a test: services that can be digitized have often escaped the disease, while those that need a person present have not.