Research Highlights Article
April 22, 2026
East–West preindustrial wage divergence
The effects of land ownership on wages in Japan, 1600–1870
Source: Katsushika Hokusai, Public domain
One of the central questions in economic history is why the Industrial Revolution began in England rather than somewhere else. East Asian societies during the eighteenth century were equally, if not more, sophisticated by many measures. One prominent explanation is that England's wages, which were considerably higher than wages in much of the rest of Europe and in East Asia, spurred innovation and investment in labor-saving technology. But a more fundamental question remains: why were wages lower in East Asia than in England?
In a paper in the American Economic Journal: Applied Economics, author Yuzuru Kumon argues that in the case of Japan, greater equality in landownership paradoxically led to Japan’s lower wages and GDP per capita.
During the Tokugawa period—1603 to 1867—Japan’s financial markets were relatively advanced, literacy rates were high, and urban economies were thriving. Edo (modern-day Tokyo) was the largest city in the world, home to one million people.
Yet, it was still a Malthusian world where people maximized the number of children conditional on resources. When households earned income above subsistence level, they tended to use those extra resources to have more children. A growing population then placed greater pressure on the fixed supply of agricultural land, reducing the marginal productivity of labor and pushing wages down.
Kumon shows the Malthusian mechanism at work by estimating the income–fertility curve, a central component of the Malthusian model. To do this, he drew on population registers known as the Shumon Ninbetsu Aratame Cho, collected by Tokugawa officials from 1660 to 1868 as a tool to identify and suppress Christian converts. The registers required every person in every village to declare their religion annually; it eventually took on broader administrative functions, such as the recording of landownership. Kumon’s sample included data for 334 villages across Honshū, the main island in Japan, which accounted for 82 percent of the population in 1750.
Preindustrial inequality
The chart below shows the Gini coefficient, a measure of wealth inequality, between 1300 and 1900 for Japan, the Republic of Venice, Piedmont, Germany, Sweden, and England.
"I think my contribution was to collect a lot of data from all over Japan and to try to make it into a more nationally representative data set, and also to use it to answer causal economic questions," Kumon said.
The empirical results confirm the central predictions of Kumon’s model. Landownership had a positive effect on household fertility as predicted by Malthus. Kumon shows that the differences in fertility operated primarily through earlier marriage rather than greater fertility within marriage, a finding that mirrors evidence from preindustrial England, France, and Germany. Another key feature of the fertility curve was its concavity—meaning richer households had more children but with diminishing returns at higher levels of wealth.
“If you distribute land and resources more equally, the concavity in the curve results in more children overall, increasing the population density in society,” Kumon told the AEA in an interview. “And the increase in population density leads to overpopulation, too many people. And when you have too many laborers, that decreases their wages.”
The results can also explain the puzzling evidence showing that a male laborer's daily earnings could sustain only two to three adults on a bare-bones diet. Landless households were found to have fertility below replacement level, consistent with their low wages. Yet, the population remained stable because many landowning households had fertility above replacement level. The balance between these two groups kept the overall population stable despite chronically low wages.
Kumon estimates that differences in landownership inequality alone accounted for roughly 27 to 48 percent of the Japan–England wage gap, rising to 45 to 67 percent under additional assumptions about differences in mortality.
The Malthusian mechanism is no longer working in today's world, but understanding what was happening in the past and how land was distributed could help to explain differences in cultural development, social development, attitudes toward women, all these kinds of things that could be having consequences on development today.
Yuzuru Kumon
The paper also points to implications that extend beyond the preindustrial period. Kumon suggests that the patterns of inequality established in this era may have left persistent cultural and institutional traces.
"The Malthusian mechanism is no longer working in today's world, but understanding what was happening in the past and how land was distributed could help to explain differences in cultural development, social development, attitudes toward women, all these kinds of things that could be having consequences on development today," Kumon said.
Overall, Kumon’s research helps to explain why wages were so low in preindustrial Japan compared to England and may also help to reframe the long-standing debate about what separated the economies of East and West before industrialization.
♦
“How Equality Created Poverty in Preindustrial Japan, 1600–1870” appears in the April 2026 issue of the American Economic Journal: Applied Economics.