Baby Busts and Growth Booms
| AUTHOR | Anupam Manur |
| DATE | November 2, 2026 |
| CATEGORIES | Macroeconomics |
In advanced economies, the shared economic anxiety is one of a shrinking, ageing population. The global birth rate has fallen from 3.78 births per 100 people in 1950 to 1.71 in 2025. The United Nations projects that populations will begin shrinking in most countries later this century. China, Japan, and South Korea now have some of the lowest birth rates on record and are projected to see substantial population declines and sharply older workforces by the middle of the twenty-first century.
The fear, flowing from conventional wisdom, is that older workers are less productive and less innovative, smaller populations generate fewer ideas, aggregate demand will be much lower and rising dependency ratios strain public finances.
A new NBER working paper by Daron Acemoglu, David Autor, Keelan Beirne and Andrew Scott examine seven decades of dramatic demographic change to test whether the fear is well founded - does falling birth rates actually cause economic decline. Their results, however, are counterintuitive.
Across countries, a one percentage point lower birth rate in 1950 is associated with about 23 log points higher GDP per working-age adult between 1970 and 2020. Across US commuting zones, lower birth rates predict faster wage growth, even after adjusting for the age, gender and education mix of workers. And crucially, aggregate GDP does not fall. In the standard growth model, a shrinking workforce raises output per worker but lowers total output, and the capital stock shrinks. The authors find the reverse on both counts: capital rises and total output holds steady.
What’s the mechanism at play here? Labour scarcity spurs labour saving technology innovation. When labout is scarce, it becomes expensive and to cut costs, firms have stronger incentives to develop and adopt labour-saving technology. Consistent with this, lower-birth-rate countries and regions show more ICT and labour-saving patents, a shift towards high-tech exports and employment, and faster TFP growth. A 1-percentage-point lower birth rate is associated with 15 log points higher total factor productivity growth over 30 years.
Can the past be a reasonable predictor of what’s to come? That’s unanswered by the paper. The ageing that is underway in China, Japan and Korea is at an unprecedented rate. And what happens when you have labour mobility? Does the fall in fertility rate in South India get compensated by labour mobility from the rest of the country?