🔍 Read the full analysis: How AI Became An 80% Service Game on ThorstenMeyerAI.com
TL;DR
A November 9 report traces the growth of US service work to a productivity gap described by economist William Baumol, rather than simply to factory decline. It points to a study in which an AI assistant increased customer-support agents’ issues resolved per hour by 15% on average, while leaving open whether AI will reduce service-sector employment or change the work people do.
A November 9 report argues that AI is beginning to target work in the service sector, where productivity has historically been harder to raise, as services now account for about four in five US workers. The piece links that development to economist William Baumol’s account of why labor-intensive services can become more expensive even as productivity rises elsewhere, and cites a customer-support study in which an AI assistant increased issues resolved per hour by 15% on average.
The report’s employment figures show that the growth of service work predates the decline of factory jobs. Based on the US Bureau of Labor Statistics’ Current Employment Statistics, it calculates that services made up 62% of US payroll jobs outside farming in 1939, 72% in 1979 and 86% in 2025. The calculation excludes farm workers and self-employed people. Using a different measure that counts all workers, International Labour Organization estimates put the US services share at 79.6% in 2025.
The report explains the cost pressures through Baumol’s cost disease. In some industries, technology and scale let businesses produce more with fewer labor hours. In others, the labor is a large part of the service itself: a quartet still needs four musicians to perform, while teaching and personal medical care also depend on human time. The report says wages tend to rise across both types of work, even when productivity gains are uneven.
As an illustration of diverging prices, the source reports that since 2000 the official, quality-adjusted US price of televisions fell by 98%, while hospital services rose 274%, college tuition 188% and car repair 143%. It also cites a study of 5,172 customer-support agents, where an AI assistant raised the number of issues resolved per hour by 15% on average. That result measures task output in the study; it does not establish what happened to workers’ pay, jobs or the wider service economy.
Infographic 02 · Baumol's cost disease
Why the workers moved to services
Baumol's mechanism in four steps, and what it looks like in US jobs.
- 1966Baumol and Bowen: the quartetA 45-minute Schubert quartet still takes about three hours of human work
- 1967Two kinds of workWhere labour is a tool, machines cut it again and again; where labour is the product, they can’t
- WagesPay rises everywhereTeachers, nurses and musicians must earn more too, or they leave
- ResultCosts rise, people moveIf we keep buying the same mix, more and more workers end up in the slow sector
Service industries, share of US nonfarm payroll jobs
US jobs, annual average 2025 (millions)
Payroll shares exclude farm workers and the self-employed (BLS establishment survey). Health spending: national health expenditure as a share of gross domestic product.
Sources: W. J. Baumol, “Macroeconomics of Unbalanced Growth”, American Economic Review 57(3), 1967, BLS Current Employment Statistics (own calculation of shares), CMS, National Health Expenditure, Table 1.
Powered by Thorsten Meyer AI
AI Reaches Labor-Intensive Services
The development matters because services account for most US employment under the broad worker measure cited in the report, and many services rely on time and attention rather than easily scaled production. If AI allows workers to resolve more customer problems in the same amount of time, it could change costs and staffing in at least some service jobs. The cited study offers a measured example of improved hourly output, but not evidence that the same gains will apply across healthcare, education, repair work or other services.
The report frames the employment consequence as unresolved: productivity tools can reduce the amount of labor required for a given task, but they can also expand the amount or range of work performed. The customer-support result alone cannot determine which effect will dominate. Readers should distinguish the study’s reported 15% average increase in issues handled from broader predictions about job losses or cheaper services, neither of which is established by that figure.
Popular right now
KitchenAid Classic Series 4.5-Quart Tilt-Head Stand Mixer, Onyx Black
Samsung 50-inch QN70H Neo QLED 4K Smart TV with Austere 2.5 m 8K HDMI Cable
Dyson Omni-Glide Cordless Stick Vacuum, Gold/Iron
As an affiliate, we earn on qualifying purchases.
From Schubert to Service Prices
The source traces Baumol’s argument to research on the performing arts by economist William Baumol and economist William Bowen. Their 1966 book, Performing Arts: The Economic Dilemma, examined the financial challenges facing theatre, opera, music and dance. A year later, Baumol developed the productivity distinction in a paper titled “Macroeconomics of Unbalanced Growth: The Anatomy of Urban Crisis.”
The report uses a 45-minute Schubert quartet as a concise example: four musicians spend a combined three hours performing it, much as they would have in the past. By contrast, manufacturing has seen major productivity gains; the source recalls that the labor needed to produce a car chassis fell sharply after Ford introduced the moving assembly line. The comparison is not an absolute division: retail, for example, has changed through supermarkets and self-service. Baumol’s point, as presented in the source, is that productivity advances are easier in some activities than others.
The article also cautions against explaining service costs solely as waste or profiteering. It notes that haircuts and car repairs rose in price despite not depending on health insurance or student loans. At the same time, it contrasts US health spending of 18.0% of the economy with Germany’s 12.4%, suggesting that national systems and other factors affect the final bill alongside economy-wide productivity pressures.
„“Human ingenuity has devised ways to reduce the labor necessary to produce an automobile, but no one has yet succeeded in decreasing the human effort expended at a live performance of a 45-minute Schubert quartet much below a total of three man-hours.”“
— William Baumol, quoted in the report from his 1967 paper
Jobs and Costs Still Unsettled
The report does not establish whether AI-driven productivity gains will lead to fewer service jobs, different tasks, higher output or lower prices. The cited customer-support study provides an average change in issues resolved per hour, but the material supplied does not specify the study’s duration, the range of results among agents, or whether its findings generalize to other employers and service industries.
It is also unclear how much of the price growth cited for hospitals, tuition, haircuts and car repair reflects Baumol-style labor costs, rather than regulation, financing, market structure or other influences. The report itself notes differences in health spending between the United States and Germany. Its figures point to a broad pattern but do not isolate the cause of each price change.
Evidence Beyond Customer Support
The next test is whether the reported productivity gains hold across more workplaces and over longer periods. Useful evidence would show how AI changes service quality, worker pay, staffing levels and prices, not only how many issues a worker resolves in an hour. Results may differ between customer support and services where personal judgment, physical presence or sustained relationships are central.
The report presents the effect of AI on service employment as the question for the season. Until more evidence is available, the clearest confirmed point in its account is narrower: service work has grown for decades, productivity gains vary by industry, and one cited study found a 15% average increase in customer-support issues resolved per hour with AI assistance. Whether that changes the number of workers needed remains open.
Key Questions
What is Baumol’s cost disease?
It is an economic account of why costs can rise in labor-intensive activities where output per worker-hour is difficult to increase, even as wages rise alongside more productive industries.
How many US workers are in services?
The report cites International Labour Organization estimates that services accounted for 79.6% of all US workers in 2025. A separate calculation from BLS payroll data puts service-providing industries at 86% of nonfarm payroll jobs that year; the measures cover different worker populations.
What did the customer-support study find?
In a study involving 5,172 customer-support agents, an AI assistant increased the number of issues resolved per hour by 15% on average, according to the source report. The figure is a productivity result, not a finding about job cuts or wage changes.
Does the report show that AI will eliminate service jobs?
No. The report says it remains unclear whether productivity gains will reduce employment or shift workers into other tasks and services. The cited study does not resolve that question.
Source: ThorstenMeyerAI.com
