EH.Net Mailing List Archive: EH.Res

============= EH.RES POSTING========================= 
 
SAYING GOODBYE TO THE AGRICULTURAL REVOLUTION 
 
Gregory Clark 
University of California, Davis 
 
 
       INTRODUCTION 
 
British population more than doubled between 1770 and 1850, and living 
standards improved.  This implies domestic food output at least doubled 
between 
1770 and 1850, even allowing for imports.  Agricultural workers were 50% of 
all 
labor in 1770, but 25% in 1850.  Thus each farm worker fed twice as many 
people 
by 1850.  Food output per worker must similarly have doubled. 
 
Together these simple arguments generate an agricultural revolution precisely 
in the Industrial Revolution period.  Britain was twice touched in the 
Industrial Revolution years by the hand of providence.  This is remarkable 
given that there was no connection between the events of the two 
 revolutions, 
the one founded on the mechanical innovations of a bright few artisans in 
textiles in the north west, the other on the small scale improvements of 
literally hundreds of thousands of farmers throughout the land.  The 
Industrial 
Revolution must not be an accident, but the result of economy wide forces 
 that 
favored growth. 
 
 
       FIVE THESES 
 
Having assembled a mass of data on land rents, returns on capital, wages, 
 and 
prices I have come to the following conclusions about the agricultural 
revolution. 
 
1.      There was no agricultural revolution.  Not in 1770-1850, not in 
1600-1770, and not in 1200-1600.  Instead from 1500 to 1850 there was a long 
slow process by which measured agricultural productivity drifted upwards at 
 an 
average rate of less than 10% in each 50 years - a process so incremental 
 that 
it would be largely unnoticeable in any person's lifetime.  This drift 
 began 
long before the Industrial Revolution, and had no connection with the 
Industrial Revolution.  Its cause seems to be largely an improvement in 
 grain 
yields, some of which may stem from the decline in interest rates between 
 1600 
and 1750 which encouraged more investment in soil fertility. 
 
        The rent, wage and price data tell us nothing much happened because 
productivity growth has to show up in higher net payments to the factors of 
production.  Yet in 1700-49 wages relative to agricultural prices were at 
 90% 
of the level in the 1860s, land rents were at 65%, and returns on capital 
 were 
at 120%.  Thus in net there was little productivity growth  no more than a 
 25% 
gain. 
 
 
2.      Correcting the estimates of the growth of output from agriculture 
leads 
to much slower growth rates of output per person in England from 1700 to 
1860. 
At first it might seem mysterious that removing the agricultural revolution 
threatens the Industrial Revolution.  Agriculture is, after all, reckoned as 
only 18% of GNP by 1861.  But it turns out that given the way output growth is 
calculated in the Industrial Revolution period, removing the agricultural 
revolution from the scene cuts the growth of income per capita from 1760 to 
1860 from the already pessimistic 65% estimated by Crafts and Harley to a mere 
31%.  Crafts and Harley, to our surprise, are wild optimists!  For a slower 
growing agriculture gets much more weight in national income in 1760 or 
 1700. 
Correspondingly the fast growing industrial sector gets much less weight.  
 The 
Industrial Revolution looks more and more like an isolated phenomena of the 
textile industries, as opposed to an economy wide transformation. 
 
 
3.      The urbanization and industrialization of Britain in 1760 to 1860 
 was 
not spurred by the release of labor by capitalist agriculture.  Instead it 
 was 
compelled by the failure of agriculture to increase output in line with 
population, which led to huge imports of food and raw materials from abroad, 
and from the domestic coal industry.  These imports had to be paid for by 
 the 
production of tradable industrial products.  Many of these products were 
 made 
in the new power factories.  They would have been made in the old hand 
workshops had there not been the mechanical advances of the Industrial 
Revolution. 
 
 
4.      The logic of the simple argument in the introduction fails because 
 it 
equates food output with agricultural output.  British agriculture did 
 produce 
a lot more food in 1850 than in 1770, but it did so in part by reducing its 
output per head of the population of wood for building and fuel, of fiber 
 and 
dyes for clothing, and of fodder for horses.  If we do an accounting of 
 total 
domestic consumption of food, fuel, and raw materials in England in 1700 
versus 
Britain in 1850 we find that in 1700 agriculture produced 95% of domestic 
consumption, while by 1850 it was producing only 49%. 
 
 
5.      The reason researchers such as Bob Allen, Mark Overton, and Michael 
Turner have found confirmation on the ground that an agricultural revolution 
did indeed occur as expected is that they all have focused on grain yields in 
measuring output.  Grain yields did increase.  Yet by 1870 grain was only 
about 
a third of net agricultural output.  People focus on grain yields because the 
physical output of meadow and pasture land has been impossible to measure. 
Yet 
price data suggests that in this large section of agriculture there were no 
gains from 1600 to 1860.  The ratio of the price of animal products to the 
price of hay changes little over these years, suggesting no gains in 
conversion 
efficiency between feed and output.  And the rent of meadow in terms of hay 
increases little suggesting little increase in yields. 
 
 
 
 
 
______________________________________________________________ 
 
Gregory Clark                    
Professor 
Department of Economics         PHONE  530-752-9242 
University of California                FAX    530-752-9382 
Davis, CA 95616 
______________________________________________________________ 
=============== FOOTER TO EH.RES POSTING===================== 
For information, send the message "info EH.RES" to lists at eh.net.