Korva is a scroll of the internet’s best ideas.
Consumers get most of AI's value
Mechanization and prices
Earlier waves of mechanization raised labor productivity in manufacturing, which drove prices down and saved consumers money. They also raised manufacturing wages, and wages in services had to rise to compete even though service businesses saw no productivity gains. That pushed the price of services up.
The share of household spending on food and clothing fell from 55 percent in 1918 to 16 percent in 2023, while the cost of knowledge-intensive services like healthcare and education has grown well above inflation.
What AI will do
Something similar will happen with AI. Knowledge-intensive services will get cheaper, so consumers can buy more of them, while services that require person-to-person interaction will get more expensive and take a greater percentage of household spending.
Projections put the gain in global GDP from AI at somewhere between 1 percent and more than 7 percent over the next decade, which is one to seven trillion dollars of new value.
Why do service prices rise as factories improve?
Economists William Baumol and William Bowen described this pattern in the 1960s, and it is called Baumol's cost disease. Their example was a string quartet, which needs the same four players and the same time to perform a piece as it did two centuries ago. Because those workers cannot produce more per hour, their pay can keep up with factory wages only if the price of the performance rises. Teaching, nursing, and haircuts follow the same logic.
Why did food and clothing take less of the budget?
German statistician Ernst Engel observed in 1857 that poorer families spend a larger share of income on food than richer ones. As incomes rise, food spending grows more slowly than income, so its share shrinks. Mechanized farming and textile mills also cut the real cost of food and clothing across the twentieth century.
More from Jerry Neumann
