Goldstone J. (2003) Agricultural Revolution in China

Goldstone, Jack A. (2003) “Feeding the people, starving the state: China’s Agricultural Revolution of the 17th/18th Centuries”, paper for the EHES Istanbul conference, 43p.


This article is available online


The Chinese population jumped from 120 to 350 million between 1620 and 1800. Many historians have assumed that the necessary growth of the agricultural output had been reached through a process of “involution” (i.e. not through gains in labour productivity, but thanks to increased effort; p.1). This stagnation of the output per person has been seen as the reverse of what happened in England at the same time: the agricultural revolution (p.2).

The Chinese agricultural revolution

But the author finds a major inconsistency in this explanation of the rise of China’s total agricultural output: why would people not have increased their production earlier if all it required was extra effort? (p.3) He further remarks that decreasing returns may not have significantly hindered innovative farming techniques because, in the new double-cropping system (which replaced the former single-cropping one), the benefits from the second crop were not diminished by fixed costs (tax, rent). As a greater share of the output went to the peasants, in effect, from, their point of view, returns hardly diminished (p.5).

The great commercial expansion of the late Ming and early-to-mid-Qing period supports this view of a growing per capita output (p.7).  Although diverse types of double-cropping occurred over China (cereals and soybeans or cotton in the North, rice and mulberry or beans in the South), the author concentrates on the case of the Jiangnan province in the Yangzi delta (rice and rapeseed of beans; p.8). This later innovation was made possible by the importation of the early-ripening Champa rice from Indochina, but it certainly took several decade to fully acclimate this variety to the several local environments of the delta (p.10). Other improvements were needed (drainage, irrigation, matching new varieties with fertilizing regimen) which explains the rather slow development of the new system. Still the portion of double-cropped land in the Jiangnan went from 40% around 1650 to 70% a century later (p.11).

“By the early 17th century, most peasants were in no position to invest in improvement of their lands. Failures in the Ming court allowed large landholders colluding with officials to escape taxes, which then fell more heavily on the peasantry.” The new Manchu dynasty managed to dominate the provincial elites and freed most of the bond-servants. Terms of trade for the peasants bettered very significantly, in particular they secured long-term leases expressed in percentage of the rice yield (p.12). A vibrant land market arose.

“The spread of double-cropping was thus also accompanied by substantial increases in regional specialization, further boosting overall efficiency of agricultural production. Finally, and perhaps most importantly, a new long-distance interregional trade arose in fertilizer, with the Yangzi importing oilcake fertilize from the north” (p.13).

Cost-benefits analysis

The Jiangnan population expended from 20 millions in 1620 to 30 in 1750, and the urban dwellers went from 15% to 20. The average farm size was of 15 mu in the 17th century, and 10 mu one hundred years later; the typical holding went from serf-like bond-servant tenants integrated in large estates to free family-size farms (p.16). The average rice yield may have rose from 1.8/2.1 shi per mu to 2.5 (p.18). The market value of the winter crop averaged 0.7 shi. Rare in the 1600s, the use of rented oxen for ploughing and pumping spread in the 1700s (p.19). The use of animals and fertilizer both increased capital-intensity and (for the latter) labour-intensity.

Most importantly, although the amount of male labour markedly increased, agricultural female labour nearly disappeared. Consequently, female labour concentrated on textile production. As the additional labour required for winter crops took place in a previously slack season, cost opportunity for this new activity remained low (p.22). The result of this series of innovation was an increase of the amount a peasant retained per day of labour ranging from 15 to 45%. For the peasants, as the number of days of labour rose, this was a valuable improvement of their income. On the other hand, on the national point of view, gains in terms of productivity were low (p.24).



Thus paradoxically, the result of double-cropping is a involution for the economy as a whole but net increase of peasants’ income (p.25). This is made clear by the doubling of the provincial cotton cloth production over the century (from 50 to 100 million bolts) which was entirely caused by the release of rural female labour. Thus income per capita generated by spinning would have increased by 40% while productivity only rose by 15% (p.26).  As a result: “peasants’ net product in 1750 was over 20 shi per family, some 30% over the normal consumption requirements of a family of five.” Drop of the cotton price after 1750, would have erased most of the family surplus. Worse, further innovation (introduction of winter rice) suffered from major decreasing returns.

The elite and the state benefited only to a very limited extent of the production increase since rents and taxes were pretty much frozen. In the meantime, a tripling of the population caused a major rise of government spending (p.31). By 1800, most of the surpluses had been exhausted by massive demographic growth; the Qing government thus faced the daunting task of responding to the needs of a three-times larger population with a narrower surplus base to draw resources from.

Generalization and conclusion

“In Britain, rather the opposite distributional effect occurred. Instead of peasants growing more prosperous, in the 16th-early 17th centuries they increasingly were reduced to landless labourers as agrarian class relation allowed landlords to capture the lion’s share of productivity gain through rising rents.” The Netherlands and Belgium were also highly productive, but their allocation pattern was closer from China’s (p.32).

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: