Full text of DOI:10.1111/1468-0289.00200

Economic History Review, LIV, 3 (2001), pp. 477–505
Farm wages and living standards
in the industrial revolution:
England, 1670-18691
By GREGORY CLARK
uch has been written on agricultural wages in England from 1670
M
to 1869, but this information has never been formed into one
national series of agricultural wages. Wilson Fox provides good evidence
based on farm accounts for 1850 and later. But for the years 1770-1849
the only national series available is the one Bowley constructed in 1898
mainly from wage surveys.2 While Bowley’s index is well founded in the
years after 1824, for the earlier period it relies on considerable inter-
polation, and takes no account of manuscript sources that have become
available in the past hundred years.3 In an unpublished doctoral thesis,
Eccleston calculated the day wages of workers on large agricultural estates
from five midland counties from 1750 to 1834, and gave an annual day
wage series for these counties.4 In another unpublished thesis, Richardson
similarly calculated the average wages in seven English counties from
1790 to 1840, in part from estate sources.5 But while the volume of The
agrarian history of England and Wales for 1750 to 1850 offers a number
of wage series on individual farms, it gives no overall wage series for that
period.6 For the years before 1750 the information is sparser. From
manuscript sources, Bowden calculated average winter day wages for
some decades in six counties for the years 1640-1749, but he had no
observations for the north of the country before 1690, and none for the
west in any decade.7
Hence, the first task this article undertakes is to produce a nominal
national agricultural wage index annually from 1670 to 1850, which
incorporates the available published wage information and manuscript
1 This research was funded by NSF grant no. SES 91-22191. For the midlands in the years
1750-1833 I benefited greatly from the pioneering work of Bernard Eccleston, and used some of
his data directly rather than going back to the sources. Peter Lindert generously provided the
material on Ardleigh. Without implicating them in any way in the conclusions, I thank the following
for advice, help, and responses to queries: Steve Broadberry, Joyce Burnette, Peter Lindert, Andrew
Oswald, Michael Turner, and anonymous referees. Robert Eyler and Mona Shraer provided excellent
research assistance.
2 Fox, ‘Agricultural wages’. Feinstein uses the Bowley series in his recent work on wages during
the industrial revolution, but he notes that ‘The most worrying feature of this series is the absence
of a reliable benchmark between 1795 and 1824’: Feinstein, ‘Changes in nominal wages’, p. 187.
3 Bowley, ‘Statistics of wages’.
4 Eccleston, ‘Survey of wage rates’.
5 Richardson, ‘Standard of living controversy’.
6 John, ‘Statistical appendix’.
7 Bowden, ‘Statistical appendix’, pp. 877-8.
 Economic History Society 2001. Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden,
MA 02148, USA.

478
gregory clark
sources, and which can be linked to the Fox series. Manuscript obser-
vations contribute about two-thirds of the information at the annual level.
A quinquennial index for these years for each of four major areas of the
country—the north, midlands, south west and south east—has also been
derived since these regions had very different wage trends in this period.
The strategy followed throughout is first to estimate a series on the
payment per day of labour to farm workers from wage payments outside
hay and harvest (44 out of 52 weeks in the year). I also check that the
series is at approximately the right level by comparing it with three
‘benchmark’ cross sections of agricultural wages. These are the 1834
poor law inquiry reports which collected wage information by circulars
in the winter of 1832-3, wages as reported in the Gardeners’ Chronicle
and Agricultural Gazette in April 1850, and Arthur Young’s wage reports
for 1767, 1768, and 1770.
Next, by examining what happens to the ratio of hay wages to winter
wages and of harvest wages to winter wages, and to the pattern of labour
inputs over the year, it is shown that these ‘winter’ wages are likely to
represent both the annual wages of male workers and the average cost
of labour to farmers. Having done this, it is possible to consider what
the series implies for the living standards of farm labourers and for the
movement of agricultural productivity.
I
The article estimates what an adult male agricultural labourer would be
paid for a day of labour from 1670 to 1869. This has necessitated the
extraction from farm accounts and secondary sources of records of pay-
ment per day of work for farm labour, including payments to workers
employed by a farmer for the whole year and those to workers employed
temporarily. A list of the manuscript sources is given in the appendix.
The form of many earlier accounts makes it impractical to try to dis-
tinguish between permanent and temporary workers. In earlier years and
in the north of England many workers were employed on annual contracts
as ‘servants in husbandry’ and received food, lodging, and clothing in
addition to a money wage. Although references to the payments to such
workers are frequently found in accounts, these payments have not been
used in constructing the estimates because of the impossibility of valuing
the in-kind compensation.
Sometimes wages are reported not by days but by weeks. In these
cases a week has always been assumed to be six full days. This assumption
was made because, for all but one account where the assumed number
of days in a ‘week’ could be calculated from internal evidence, the length
of the week was six days all the way from the seventeenth century to the
nineteenth. Thus in Somerset in March 1714 farm labourers were
recorded as being paid 10d. per day, or 5s. per week, implying that a
week was regarded as six days. Similarly on a farm in Bocking, Essex,
in March 1773, William Dod was paid £2 13s. 8d. for working seven
weeks and four days and a further 28d. for working two more days. At
 Economic History Society 2001

english farm wages, 1670-1869
479
six days per week his rate per day for the first payment would be the
same, at 14d. per day.8
There are two big problems with estimating day wages from farm and
estate accounts. The first is inferring when the employee referred to is
an adult male. Farms employed numbers of women, boys, and girls for
various tasks and paid them much less than adult males, but the accounts
rarely show the age of workers, and often not even the gender. To make
sure that adult male wages only were included, it is possible to rely on
the sexual division of labour that was practised in the English countryside
from the middle ages onwards. Thus the tasks of threshing, ditching,
hedging, mowing, carting, cutting wood and making faggots, which
together occupied a large portion of the agricultural year, seem to have
been almost exclusively the jobs of adult male workers. Farm tasks such
as these can safely be included in the wage index. And once a worker is
identified as performing such tasks he can safely be presumed to be an
adult male agricultural labourer. Some tasks such as reaping and weeding
were done by both men and women, at least in earlier years, and these
are included only where ancillary evidence shows that the workers were
adult males. Again, some tasks such as ploughing and stone picking often
involved boys, and again these are included only where other evidence
shows the worker to have been an adult male.
Nineteenth-century accounts often had a different form where a group
of what were obviously male farm workers would have their wages and
days worked detailed week by week, but the amount of the wages shows
some of the group to have been youths or elderly workers. Thus the
wages listed per day would be, for example, seven workers at 30d., three
workers at 28d., and one worker at 18d.9 In this case the wage has been
taken as the average of all wages that fell within two-thirds of the modal
wage. The effects of counting or not counting these outlying wages in
the mean wage were generally small.
The second problem lies in determining whether, in addition to the
wage, the worker received some of his pay as food, beer, cottage accom-
modation, an allotment, or the right to buy grain at low prices. Such
allowances are generally not recorded in these accounts. Detecting cases
in regular work where the worker was fed on the job is not as hard
where farmers employed workers both with and without food, since the
wage with food would often be no more than half the wage without
food. Thus if there are two sets of wages at very different levels it is
often apparent that one is for wages with food. In an estate in Cumberland
8 Somerset RO, Carew, DD/TB BOX 14/12, Essex RO, Tabor family, DDTA/A3. The following
other accounts for the years 1800 and earlier, listed in chronological order, showed a week of six
days. Staffordshire RO, Levenson-Gower, D593/F/3/25 (1681); West Yorkshire RO: Sheepscar,
Ingram/Irwin, TN/EA/12/11 (1691); Somerset RO, Parsonage, DD/X/REE/C/1308 (1692); Somerset
RO, Popham, DD/PO/32/3-14 (1706, 1713); Hampshire RO, Russell, 149M89/R5/6103 (1762-3);
Nottingham RO, Portland, DD5P/4/1 (1775-6); Essex RO, Tabor family, DDTA/A1 (1788-9);
Warwickshire RO, Conway, CR114A, 357 (1791-2); Durham RO, Salvin, D/Sa/E177 (1800). One
account showed a week varying from four to six days, with an average of 5.6 days. This was
Durham RO, Salvin, D/Sa/E167 (1715-17).
9 Cumbria RO, Dlons L3/5/55, Oct. 1846.
 Economic History Society 2001

480
gregory clark
in 1732, for example, the same person was paid both 9d. per day and
4d. per day for threshing, and the lower rate clearly involved food in
addition.10 Detecting wages that included food at hay or harvest time is
very difficult since these wages could vary a great deal from regular
wages, and food is a smaller share of wages in these circumstances.
Fortunately, in southern England at least, provision of food to workers
was relatively rare by the late seventeenth century, and may have been
unusual long before that. In the north of England, however, workers
were often fed at work even in the nineteenth century, and vigilance is
required to avoid including such wages in calculations of the mean.
Detecting from the accounts when workers in regular work received
beer has proved to be impossible, since beer was a much smaller sup-
plement to wages, and so cannot be identified from internal evidence.
Beer was often still provided even in 1832, especially at hay and harvest.
But evidence from the 1834 Report on the Poor Laws indicates that where
beer was provided it was worth about 10 per cent of wages in winter
and summer, and less than this in harvest. Thus changes in the degree
of beer provision will have some effect on wages, but not an especially
dramatic one.
Table 1. Amounts of day wage data in the wage
dataset
Work type
Individual
observations averaged by year
Observations
and place
All
12,746
7,812
886
526
472
468
334
232
211
191
180
177
143
140
3,022
1,730
405
517
216
347
84
206
70
136
70
97
Unspecified farm
Hedge
Farm work on estate
Thresh and winnow
Mow
Labour in garden
Harvest
Cart
Ditch
Plough
Dung
Dig
Reap
103
108
Note: ‘Farm work on estate’ refers to workers performing tasks typical of
farm workers, such as hedging or carting, on estate grounds.
Source: Agricultural wage dataset
The various sources of wage information used have been combined
into an ‘agricultural wages’ dataset containing 12,746 quotations of day
or weekly wages for farm workers for the years 1670-1850, though some
are at a much more disaggregated level than others. Hence the quotations
of day wages are reduced to 3,022 if averaged by place and year. The
breakdown of the most frequent work descriptions is given in table 1.
10 Cumbria RO, DPenn/204.
 Economic History Society 2001

english farm wages, 1670-1869
481
Table 2. Wage rates from the 1834 report
Season
No. of
observations
Wage/day
(d., money
only)
Wage/day
(d., valuing
beer allowances) (cash + value
Ratio to winter Ratio to winter
wages wages
(cash only)
of beer)
Winter
Summer
Hay
898
897
76
19.96
22.15
26.01
35.98
20.42
22.70
28.24
38.56
—
—
1.12
1.42
1.92
1.11
1.28
1.73
Harvest
172
Notes: Wages including beer were converted into equivalent money wages using the average difference between
wages with and without beer in specific parishes in each season. This was: for winter, 1.94d. per day (95 cases),
for summer 2.13d. (104 cases), for hay and harvest 2.67d. (6 cases). In the 15 cases where quantities were stated,
average consumption of beer or cider in winter was 3.5 pints per day.
Source: House of Commons, Report on the Poor Laws.
To construct the annual index of wages from 1670 to 1850, wages
outside hay and harvest are used, and defined as wages for all farm work
carried out between October and May, and for regular farm operations
such as ditching, threshing, and hedging carried out in the summer
months.11 Wages varied somewhat even in the period outside harvest,
those paid in the winter months being somewhat lower than even non-
harvest wages in the summer months. Table 2 shows for the 1832 wages
in the Report on the Poor Laws the level in winter, summer, hay time,
and harvest. Summer wages were on average 11 per cent greater than
winter wages, with hay and harvest wages higher still. Thus in calculating
average wages outside the harvest and hay periods from the sources, I
have increased wages paid from October to March by 4.5 per cent, and
reduced wages paid from April to September by 6 per cent. Some workers
were also employed on an annual basis and paid a fixed wage throughout
the year even in the higher-wage hay and harvest periods. In these cases
the winter and summer wages outside harvest were reduced by 7 per
cent to make them comparable to those paid to workers who benefited
from the higher hay and harvest earnings.
The data were divided into four regions—the north, the midlands, the
south east, and the south west—because there are indications that wages
moved in different ways in each of these regions in this period. Thus
the north went from being the lowest wage to the highest wage region
over this period, and the south east from being about 20 per cent above
the national average wage level to being 10 per cent below that level.
Table 3 shows the number of wage sources by decade for each region,
demonstrating that the amount of data is much greater for some regions
than for others and that the relative amount of information varies by
period.
11 Where the exact period of the work is not stated, but it is work usually done outside the hay
or harvest period, this has been counted as winter work.
 Economic History Society 2001

482
gregory clark
Table 3. Number of wage sources by decade for the wage series, 1670-1850
Decade
North
Midlands
South west
South east
All
1670-9
1680-9
1690-9
1700-9
1710-19
1720-9
1730-9
1740-9
1750-9
1760-9
1770-9
1780-9
1790-9
1800-9
1810-19
1820-9
1830-9
1840-9
4
3
8
6
5
6
10
7
5
8
6
5
4
7
7
7
5
6
5
9
12
6
11
9
11
14
21
16
12
15
23
24
19
20
13
4
1
3
4
7
9
4
7
5
7
4
2
4
5
4
6
9
7
11
11
12
15
13
9
12
11
6
9
5
4
7
25
20
32
37
32
36
36
36
31
45
31
23
30
47
46
40
45
34
12
11
8
11
8
6
Notes: The areas are composed as follows: North: Cheshire, Cumberland, Durham, Lancashire, Northumberland,
Westmorland, Yorkshire. Midlands: Bedford, Berkshire, Buckingham, Derby, Huntingdon, Leicester, Lincoln,
Northampton, Nottingham, Oxford, Rutland, Stafford, Warwick. South west: Cornwall, Devon, Dorset, Gloucester,
Hereford, Monmouth, Shropshire, Somerset, Wiltshire, Worcester. South east: Cambridge, Essex, Hampshire,
Hertford, Kent, Middlesex, Norfolk, Suffolk, Surrey, Sussex.
To construct the overall average wage which is not influenced by the
varying amount of data from each region, the following expression was
fitted to the data from each of the four regions:
ln(Wage ) =
␣ PLACE + ␾ DYEAR + ⑀
͸ ͸
it
i
i
t
t
it
i
t
where i indexes the source and t the year. PLACEi is an indicator variable
which is 1 when the observation is from source i, 0 otherwise. The
inclusion of the indicator for each wage source, along with the logarithmic
form for wages, allows for constant percentage differences in farm wage
rates. The estimation procedure calculates the movement of wages by
looking just at what happens within each series from each source over
time. The movement of wages across years is estimated by including
indicator variables for years DYEARt which are 1 in year t, 0 otherwise.
For the south west, where the data are thin before 1800 and the individual
series often do not overlap, this procedure did not work well, and the
indicator variables for each source were not used in the years before
1788 (in effect it was necessary to assume that the average wage level at
any time was the same across all sources in the south west in these
earlier years).12 The national wage is calculated as the weighted average
of the estimated wage in each region, weighting by the numbers of male
agricultural workers in each region recorded in the 1851 census tables
12 In the south west before 1700 the day wage data were supplemented by piece rates on threshing
and faggoting to fill in some years where there were no day wage data.
 Economic History Society 2001

english farm wages, 1670-1869
483
Figure 1. Average daily wages outside hay and harvest, 1670-1869
Note: The bold line shows winter wages as a weighted average by agricultural area of four divisions of England.
The light line shows a simple average of all the places reporting data for the year in question.
Source: Agricultural wages dataset
on occupations.13 Figure 1 shows as the faint line the raw average winter
day wage in each year. The bold line is the wage index calculated from
the above regression for each region, and averaged across each region.
The corrected index differs little from the uncorrected index in most
years. In the years 1670-1760 it is generally below the raw average,
because the high-wage south east is over-represented in the data in these
years. In 1770-90 the wage index is generally above the raw average.
Nominal agricultural wages are essentially flat in the years 1670 to 1730,
at an average rate of about 10.3d. per day in the winter. Thereafter there
is a rise to a peak in 1813, followed by a decline to the 1820s. With the
exception of notable downturns in 1822-4, 1834-7, and 1848-50, wages
are fairly steady between 1820 and 1850. Appendix table A1 shows
annual estimated nominal day wages outside hay and harvest from 1670
to 1850.
Figure 2 compares this series with the Bowley index for the years 1770
to 1850. The Bowley index was constructed using a few cross sections
of wages—1770, 1796, 1824, 1832, 1837, 1850—interpolated using rec-
ords for a small number of farms. For the years after 1820 the two series
move closely together. But for some of the earlier dates, and in particular
13 All farm workers between the ages of 15 and 65, including farmers, have been counted in this
calculation. The numbers were respectively north, 240,124; midlands, 284,676; south east, 334,163;
south west, 278,001.
 Economic History Society 2001

484
gregory clark
30
25
20
15
10
5
Bowley
Poor Law
Report
New index
Agricultural
Gazette
Arthur
Young
0
1770
1780
1790
1800
1810
Year
1820
1830
1840
1850
Figure 2. Winter wages in English agriculture, 1770-1850
Sources: app. tab. 1; tab. 5; Bowley, ‘Statistics of wages’
1795-1820, the Bowley series is very different. Thus at the extreme in
1807 Bowley reports wages at 27.8d. per day, while the new series reports
only 19.4d., a 44 per cent difference. Bowley is generally much more
optimistic about the level of rural wages for the revolutionary and Napo-
leonic war periods. Figure 3 shows that the deviation between Bowley
and this series in the years 1790-1820 is no accident of sampling in the
sources available for use here. The figure shows the Bowley series for
England by quinquennia compared with the movement of wages in each
of the four regions for which wages were estimated, with 1785-9 in each
Figure 3. Winter wages by quinquennia, 1790-1824
Sources: as fig. 2
 Economic History Society 2001

english farm wages, 1670-1869
485
case fixed at 100 since the two series are at the same level in this
quinquennium. In the periods 1795-9, 1800-4, and 1805-9 wages in all
four regions rise by much less than the Bowley series would predict, but
are generally similar across the regions. After 1810 the two southern regions
show lower wage increases than the northern regions, exactly what everything
else we know about the period would suggest should happen. Because all
four regions show very similar trends, and individually plausible trends, the
chances that this result stemmed from sampling error are very remote.
Bowley relied on information for only one or two farms in interpolating
between 1796 and 1824 and clearly these were not representative.
Table 4. Winter wages by region by quinquennia, 1670-1849
Quinquennium
North
Midlands
South east
South west
All
1670-4
1675-9
1680-4
1685-9
1690-4
1695-9
1700-4
1705-9
1710-14
1715-19
1720-4
1725-9
1730-4
1735-9
1740-4
1745-9
1750-4
1755-9
1760-4
1765-9
1770-4
1775-9
1780-4
1785-9
1790-4
1795-9
1800-4
1805-9
1810-14
1815-19
1820-4
1825-9
1830-4
1835-9
1840-4
1845-9
1850-4
1855-9
1860-4
1865-9
—
7.2
10.0
—
6.4
6.8
6.4
7.7
6.7
7.7
6.5
6.5
8.6
9.0
8.2
8.4
9.2
9.8
9.3
8.4
8.0
8.6
8.9
8.9
8.9
8.9
9.3
9.2
10.2
10.5
10.6
9.8
10.4
9.8
9.7
10.1
10.4
10.9
12.0
12.9
12.5
15.2
16.5
19.4
20.7
26.4
23.6
20.3
21.8
21.2
20.4
21.3
21.1
20.9
23.1
23.1
24.5
11.8
11.8
12.0
13.7
12.4
11.9
12.5
12.5
12.0
12.1
12.2
12.0
11.8
12.2
12.7
12.7
13.2
13.3
13.3
13.2
14.9
13.5
16.6
15.1
15.9
17.1
19.1
21.3
25.8
22.8
19.0
19.5
19.0
18.4
20.2
20.7
19.0
22.1
21.3
22.8
12.3
10.5
11.2
11.3
11.3
10.3
—
10.5
9.9
10.5
10.5
9.9
9.7
9.8
11.4
10.7
11.8
12.1
11.4
12.2
11.8
12.1
11.4
10.8
11.0
11.2
11.8
11.9
11.9
11.9
13.6
13.5
16.0
17.8
19.4
19.3
19.3
17.6
17.6
17.3
17.6
17.9
18.3
18.1
20.3
20.9
22.2
10.3
9.8
10.4
10.2
10.3
10.9
11.0
10.9
10.9
10.9
11.0
11.3
11.4
12.3
12.2
13.3
13.3
14.3
15.9
18.2
20.2
23.2
22.1
19.5
20.1
19.9
19.4
20.2
20.8
20.2
22.9
22.9
24.3
9.4
9.9
9.9
10.7
10.9
11.0
11.5
12.2
13.5
16.1
18.7
20.5
22.4
21.4
21.9
22.4
21.5
21.3
23.5
23.3
26.5
27.1
28.5
Notes: Nominal wages, d. per day. Some workers would also have received beer allowances.
Sources: Agricultural wages dataset; Fox, ‘Agricultural wages’.
 Economic History Society 2001

486
gregory clark
Table 4 shows by quinquennia the estimated movement of wages in
each of the four regions. The series from each region have been continued
for the years 1850-69 by matching them to the nearest corresponding
areas reported by Wilson Fox in 1850. There is considerable movement.
Wages in the north, which were 20 per cent below the national average
in the years 1670-99, were 10 per cent above the national average by
1845. This is in contrast to the situation in the south east, with wages
some 20 per cent above the national average at the earlier date, but at
the national average by 1845. And in the south west wages fell from a
position nearly 20 per cent above the national average in 1670-99 to 10
per cent below the average by 1845.
II
How good an indication is this series, constructed on the basis of an
average of 17 randomly located places per year, of the movement of
wages outside hay and harvest? The answer, at least for the years after
1767, seems to be that it is likely to be fairly accurate at the national
level. I measure how well the new series is likely to represent wage trends
by comparing its average level with average wages nationally at three
points where we have extensive data from across the country: 1767-70,
1832, and 1850. The 1832 benchmark is the best of all, since it comes
from a large sample of parishes. The national and regional wages for
that year, which are derived from the Report on the Poor Laws, and which
are shown in table 5, are based on 908 day wage observations and are
calculated by the same method used for the index. The Report suggests
an average money wage outside harvest and hay of 20.9d. per day.14 The
wage index estimates national wages in the same year as 20.7d., an error
of less than 1 per cent if the Report is taken as definitive. For 1767-70
Arthur Young gives 160 day wages, which imply an average money wage
of 12.0d., compared with the wage index that shows 11.4d, a difference
of less than 6 per cent. For 1850 the Agricultural Gazette reports 127
wages for April in 38 of 42 counties in England, which suggest a national
average wage of 18.5d., and the index is then at 19.3d., a difference of
only 4 per cent. Thus on a national scale we can seemingly expect that
the wage index will typically show wages within about 4 per cent of their
true national level in the years 1670 to 1850, since the number of farms
observed in each 10-year period is about the same throughout.
As expected, the regional wage levels deviate more from their respective
benchmarks. But only in one case does the deviation exceed 10 per cent
in any of the benchmark years. In 1832 the biggest of the four regional
deviations is less than 4 per cent. The estimates here thus also give a
rough picture of the movement of wages in the four different regions.
14 To calculate this wage it is assumed that winter wages covered six of the 10 months outside
harvest, and summer wages the other four.
 Economic History Society 2001

english farm wages, 1670-1869
487
Table 5. Nominal money wages outside harvest compared with benchmarks
Source
Period
All
North
Midlands
South east
South west
Arthur Young
Index
1767-70
1767-70
12.0
11.4
11.1
10.1
11.1
10.6
14.3
12.8
11.1
11.8
Report on the Poor Laws
Index
1832
1832
20.9
20.7
22.0
22.2
22.2
22.0
22.0
21.2
17.4
17.3
Agricultural Gazette
Index
1850
1850
18.5
19.3
22.0
22.7
18.7
19.9
17.7
17.8
16.4
17.7
Sources: Young, Six months’ tour; idem, Farmer’s tour; idem, Six weeks’ tour; House of Commons, Report on the Poor
Laws; Gardeners’ Chronicle and Agricultural Gazette, ‘The value of agricultural labour’, 27 April 1850, pp. 266-7;
Agricultural wages dataset
III
The above series shows what happened to wages per day outside the hay
and harvest periods. But the interest in wages in this context stems from
the information they provide in combination with other input prices on
the movement of agricultural productivity, and the information they
supply on living standards. If it is reasonable to assume that the typical
agricultural worker was employed for roughly the same number of days
per year throughout the period 1670 to 1869, then the day wage series
will also be an index of average farm labour costs and of the annual
earnings of male farm labourers. To see this, note that the total number
of days worked will be N, where
N = Nwinter + Nhay + Nharvest
and Nwinter is the number of days worked outside harvest, Nhay the
number of days worked in the hay harvest, and Nharvest the number of
days worked in the grain harvest. The total annual wage income per
worker, W, will be
W = wwinterNwinter + whayNhay + wharvestNharvest
whay
wwinter
wharvest
wwinter
= wwinter Nwinter
+
Nhay
+
Nharvest
ͩ
ͩ ͪ ͩ ͪ
ͪ
As long as the ratio of the hay wage to the winter wage and that of the
harvest wage to the winter wage remain constant, and the total number
of days worked each year outside hay and harvest does not change, total
annual earnings W will move proportionately with the winter wage.
Similarly the average cost per day of labour, W/N, will again be pro-
portionate to the winter wage.
Information from the wage accounts suggests that the ratio of harvest
and hay wages to wages in the rest of the year changed little from 1670
to 1850. Table 6 shows the ratio of money wage payments at harvest
and hay to non-harvest wages from the farm accounts used in the wage
index from 1670 on. The ratio does not change much over the course
of 180 years, and most of the apparent change may simply be sampling
error. The ratios in the farm accounts are also relatively similar to the
 Economic History Society 2001

488
gregory clark
Table 6. Money harvest and hay wages relative to money non-harvest wages
Source
Period
No. of places
Hay
wage/‘winter’
wage
No. of places
Harvest
wage/‘winter’
wage
Farm accounts
Farm accounts
Farm accounts
Arthur Young
Report on the Poor Laws
Agricultural Gazette
1670-99
1700-69
1770-1850
1767-70
1832
18
42
26
139
75
—
1.28
1.28
1.21
1.39
1.29
—
17
41
23
131
169
110
1.48
1.47
1.56
1.56
1.77
1.64
1850
Sources: see text
various benchmark cross sections that are available from 1770, 1832,
and 1850.
Note also that, since together the hay and harvest period covered only
about eight weeks, if workers were employed for most of the remaining
44 weeks of the year, any change in the ratio of hay and harvest wages
to winter wages would have very little effect on total earnings. For in
that case, taking winter wages from the 1832 Report, the 44 weeks of
winter employment would provide 78 per cent of annual money earnings,
so that any modest change in the ratio of harvest wages to winter wages
would have very little effect on annual earnings or on the average labour
costs of farmers. A 10 per cent change in the ratio of hay and harvest
wages to winter wages will produce a 2.2 per cent change in annual
earnings and in average labour costs to farmers. If weeks worked in
winter were constant but less than 44, then the annual wage and annual
labour cost would be more sensitive to the exact ratio of harvest wages
to winter wages. But even if the average worker experienced as many as
10 weeks of idleness in winter from lack of work, winter earnings would
still be 73 per cent of annual earnings, so that modest changes in the
ratio of harvest and hay wages to winter wages would have little effect.
Thus, as long as the number of days of employment outside harvest
is relatively constant from 1670 to 1870, and workers are not idle for
more than three months or so in the winter, winter wages will provide
a very good index of both annual earnings and average labour costs.
Only if the days worked outside harvest changed significantly would an
index based on winter wages be misleading. Suppose, for example, that
by the nineteenth century the average worker was unemployed for 10
weeks in the winter, while in 1670 work was available year round. In
that case, annual earnings in the nineteenth century would fall by 18 per
cent if the nominal winter wage stayed constant. In general, for every
increase of 10 per cent in the average level of unemployment annual
earnings will fall by 9 per cent, assuming that unemployment occurs only
in the winter and there is no replacement of wages by poor relief.
My claim is that male agricultural labourers were typically employed
for 300 days or so per year throughout the period 1670 to 1869.
This is against the general belief among agrarian historians that winter
unemployment was a significant problem for English agricultural workers
 Economic History Society 2001

english farm wages, 1670-1869
489
by the nineteenth century, and perhaps also at an earlier date. Hobsbawm
and Rude´, for example, note of the period around 1830 that ‘Under-
employment was constant, except perhaps at the height of the harvest,
and sometimes even then. Yet it is clear that the main burden of
unemployment was concentrated in the winter months.’15 It is not possible
here to argue this point to its full extent. But evidence both from the farm
accounts and from poor law sources suggests that winter unemployment in
English agriculture is likely to have been modest even in the nineteenth
century. Consider first the poor law evidence. Ardleigh in Essex was a
largely rural parish in the early nineteenth century, with 80 per cent of
males aged 15-59 reporting agriculture as their occupation in 1831. It
was also located in a region largely devoted to grain production where
the seasonality of labour demands was pronounced, and it was a parish
of consistently high poor law payments. The average payment per head
in 1831 was £1.43, which places Ardleigh in the top 11 per cent of a
sample of 926 rural parishes in terms of payments per head.16 The
overseers’ books for Ardleigh in 1821 list details of the names of those
receiving payments, the amounts paid, and the reason: ‘no work’, ‘ill’,
‘lame’, and even ‘lazy’. In addition, manuscript censuses for Ardleigh
survive for 1796 and 1821, listing the ages and occupations of everyone
in the first case, and of the head of household in the second. From these
sources it is possible to estimate the number of males aged 12-64 in
Ardleigh in 1821, the fraction of these employed as labourers or artisans,
and the share of these workers unemployed in each week of the year
1821. This share is given as an upper and lower bound in figure 4. The
lower bound shows the share calculated as the fraction of males explicitly
recorded as unemployed for each week, or whom the parish employed.
The upper bound shows the share calculated including as unemployed
those workers receiving relief where no reason was given, or the reason
was only ‘in want’. There is significant unemployment in the months
between December and July, when on average between 5 and 8 per cent
of male workers were unemployed. But this still entails an average annual
unemployment rate of between 3.9 and 5.6 per cent of the male labour
force.17 And this 4-6 per cent unemployment did not translate into a 4-
6 per cent decline in income since unemployment was concentrated in
the months outside the high earnings of the harvest season, and the poor
relief payments made up some of the difference. Assuming poor relief
replaced even half the lost wage income, the losses in income from
unemployment in Ardleigh in 1821 would certainly average less than 3
per cent of potential annual income. And it is necessary to remember
15 Hobsbawm and Rude´, Captain Swing, p. 74. Allen, Enclosure and the yeoman, similarly argues
that over the course of the eighteenth century ‘irregularity, casualness, and the seasonality of
employment increased’ (p. 288).
16 Payments per head in Ardleigh were £1.15, which again would place it well within the upper
range in terms of poor relief payments.
17 The reason why little unemployment was recorded in weeks 39-46 immediately after the harvest
might be that workers who were unemployed were expected to live on their extra harvest earnings
in these weeks, but even if the unemployment rate was the average for the months December-July
in all the 46 weeks outside harvest the average annual rate would be between 4.9 and 6.9%.
 Economic History Society 2001

490
gregory clark
Figure 4. Male unemployment in Ardleigh, Essex, 1821
Note: The weeks of the year are numbered from January to December
Sources: Essex RO, Chelmsford: Disbursements to the poor, 1821. D/P263/13/7-8; Census of Sept. 1796.
D/P263/1/5; Account of the population of Ardleigh, 28 May 1821. D/P263/28/1
that Ardleigh was a parish with unusually high poor relief payments.
Most parishes would have experienced less unemployment than this in
the nineteenth century, at least until the end of the old poor law in 1834.
The second source suggesting that close to full employment was the
norm over the period 1670 to 1850 is the material included in the farm
accounts themselves which contain information on the amount of wage
payments to workers in each month of the year. I did not systematically
collect this information, but for the accounts from earlier years which
tended to be simpler in form I sometimes recorded all the identifiable
payments to men in each month of the year. Aggregated, these payment
records suggest that the amount of daily or weekly male labour hired by
farms actually fluctuated little over the course of the year, at least before
1790. Figure 5 shows the average recorded expenditure per month on
labour from 55 farm accounts over 1,277 months for the years 1640 to
1789 measured relative to the average monthly payment over the whole
year. August payments were 55 per cent greater than those made in the
winter months, but since the harvest wage in table 6 was 48 per cent
greater, this implies that total employment in August was only about 5
per cent greater than that in the winter.18 There is no very obvious winter
slack season in labour hiring. Employment in December to February was
just as high as in the rest of the year outside harvest. The reason why
farms could hire workers at a relatively constant rate was that tasks such
as threshing—a major task of hired labour—or hedging and ditching
could be done at any time of year. Figure 5 also shows threshing and
18 Not all payments in August would be at the harvest rate, so the employment differential would
be somewhat higher than this.
 Economic History Society 2001

english farm wages, 1670-1869
491
1.6
1.2
0.8
0.4
0
Jan. Feb. Mar. Apr. May. Jun. Jul. Aug. Sep. Oct. Nov. Dec.
Figure 5. Labour payments to adult males by month, 1640-1789
Note: Payments have been adjusted to a standard month length of 31 days. The solid black area in each column
represents payments explicitly identified as for threshing. The white areas are payments explicitly for hedging or
ditching. The hatched area shows all other payments.
Source: Agricultural wages dataset
hedging payments. These were concentrated in the winter months when
other labour tasks were in short supply.
These labour payments do not include payment to the servants in
husbandry boarded throughout the year on the farm. Thus if any seasonal-
ity were going to appear in labour hiring it would be in these payments
to hired out workers. A fortiori, at least before 1790, the types of farms
represented by these accounts seem to have provided a relatively steady
demand for adult male labour over the course of the year. Since these
accounts tend to come disproportionately from the home farms of large
estates, this may not be the typical pattern of hiring for the farm sector
as a whole. But what these accounts do show is that farmers who wanted
to could arrange to hire a relatively constant stream of labour over the
year. And given that labour was cheapest in the off-season there was
every reason for them to smooth out labour demands as much as
they could.
Unfortunately, for 1790 and later I did not generally record the
amounts of day labour payments across each month of the year. But for
two farms in Dorset and Essex in the years 1837-40 I do have the annual
record of payments. These show just the same pattern as in the earlier
years in labour payments to male workers. Figure 6 shows the pattern
on these farms in comparison with the earlier years.
Because the issue of unemployment in agriculture is a large one that
cannot be settled here, in the remainder of this article I assume full
 Economic History Society 2001

492
gregory clark
1.6
1.2
0.8
0.4
0
Jan. Feb. Mar. Apr. May. Jun. Jul. Aug. Sep. Oct. Nov. Dec.
Figure 6. Labour payments to adult males by month, 1640-1789 and 1837-40
Note: The solid columns show the pattern on labour payments for the years 1640-1789. The light columns show
the pattern for 1837-40.
Source: Agricultural wages dataset
employment throughout.19 But from the data presented here readers can
construct their own real wage series, adjusting for the degree of unemploy-
ment they think reasonable in each period.
IV
Having generated a nominal wage series for agricultural labourers that
should indicate annual earnings, it is also possible to ask what happened
to the real purchasing power of wages in this interval. A rural cost of
living index was formed as a geometric index of the prices of each
component, with the expenditure shares of farm labourers’ families used
as weights for each of two sub-periods. Thus it assumes constant shares
of expenditure on each item as relative prices change. That is, if pit is
the price index for each commodity i in year t, and ␣i is the expenditure
share of commodity i, then the overall price level in each year, pt is
calculated as
␣
it
i
p =
t
p
͹
i
This implies that if the relative price of an item such as housing increases,
consumers adapt by reducing relative purchases of the item to the degree
to which the share of expenditures on each item remains constant. The
weights used in forming the index are derived from data given by Horrell
19 Clark and van der Werf, ‘Work in progress’, discusses other evidence on this issue suggesting
little change in labour input.
 Economic History Society 2001

english farm wages, 1670-1869
493
Table 7. Percentage of expenditure by category for agricultural labourers
Category of expenditure
Horrell
Assumed here
1787-1796
1840-1854
1670-1819
1820-1869
Food and beer
Bread
Flour
Wheat
Barley
Oats and oatmeal
Potatoes
Beer
Meat and fish
Bacon
Milk
Cheese
Butter
Sugar and treacle
Tea and coffee
Other food
Housing
Fuel
Light and soap
Services
77.0
34.8
5.3
0.0
1.0
3.6
2.0
1.0
9.2
1.3
4.0
3.5
3.9
3.6
2.4
1.4
6.0
4.0
4.8
0.1
0.0
8.2
68.6
8.8
24.7
3.0
1.4
2.2
6.0
0.0
3.4
2.8
3.2
2.6
3.3
3.1
2.6
1.6
10.1
4.5
3.3
0.7
1.0
11.7
77.0
0.0
0.0
38.1
2.4
7.1
0.0
1.0
9.4
1.3
4.1
3.6
4.0
3.7
2.4
0.0
6.0
4.0
4.8
0.1
0.0
8.2
69.6
0.0
0.0
37.4
1.4
2.8
6.2
0.0
3.5
2.9
3.3
2.7
3.4
3.2
2.7
0.0
10.1
4.5
3.3
0.7
0.0
11.7
Tobacco
Other (Clothing)
Source: Horrell, ‘Home demand and British industrialization’, pp. 568-9, 577
for the share of expenditures of agricultural workers in 1787-96 and
1840-54 and shown in the first two columns of table 7. Since I do not
have prices for some minor items such as tobacco the weights have been
modified to those in the last two columns.
In his paper on real wages in the industrial revolution period, Feinstein
instead uses a fixed weight index of the form
p =
t
␣ p
i it
͸
i
for each of three sub-periods. The Feinstein index assumes that if the
relative price of an item increases then within each sub-period there is
no effect on the relative amount consumed. The Feinstein index will
tend to show greater increases in the cost of living than the index
preferred here.
The decadal price levels for the major commodity groups used to form
the cost of living index, and their sources, are given below, in appendix
table A2. For bread and flour, the staple article that formed nearly half
of farm labourers’ expenditures, the price of wheat is used. Even though
wheat was only an input into making flour and bread, bread had very
different qualities that are very hard to control for over long time intervals,
and the cost of wheat was a very large share of the cost of flour and
bread. For fuel I use the price of faggots alone until 1770 since in rural
areas these were the main source of fuel until the nineteenth century at
 Economic History Society 2001

494
gregory clark
275
250
225
200
175
150
125
100
75
Feinstein cost
of living
Lindert and
Williamson
Farm workers’
cost of living
50
25
0
1770
1790
1810
1830
1850
1870
Year
Figure 7. Cost of living for farm workers, 1770-1869
Sources: Lindert and Williamson, ‘English workers’ real wages’; Feinstein, ‘Pessimism perpetuated’; app. tab. 1
least. From 1770 to 1830 I use a combination of the prices of faggots
and coal, and after 1830 coal alone.20 For light and soap I use the prices
of tallow candles and of tallow, the main input in making soap. Cottage
rents are estimated from housing of unchanged quality owned by charities
in parishes that had a population density of less than one person per
acre in 1801.21
Figure 7 shows the resulting index compared with Feinstein’s overall
cost of living index for all workers for the years 1770-1869 where in
both cases the average level of the indices in 1770-9 has been set at 100,
and compared with Lindert and Williamson’s earlier cost of living index.
Appendix table A1 shows the cost of living index for each year. All three
indices are very close in the years 1770-1810 but diverge thereafter. By
the 1840s the Feinstein cost of living series is about 13 per cent higher,
and the Lindert and Williamson series about 6 per cent lower. By the
1860s the Feinstein cost of living is 17 per cent higher than mine.
A number of things explain this deviation from Feinstein. One
important matter is the different functional form for the index. An
experiment conducted by reproducing roughly the Feinstein series with
a constant share of expenditure series suggests that that alone would
reduce the cost of living by about 7 per cent in the 1860s (see table
20 In the years 1770-1830 faggots and coal were each given 50% weight in the fuel series.
21 The method of estimation here is the same as in G. Clark, ‘Shelter from the storm: housing
and the industrial revolution’ (working paper, Univ. of California, Davis, 2000; http://www.econ.uc
davis.edu/faculty/gclark), with just a restriction to housing in rural parishes.
 Economic History Society 2001

english farm wages, 1670-1869
495
Table 8. Explaining the difference between cost of
living indices
Cost of living, 1860s
relative to 1770s
Feinstein index
1.479
1.376
1.347
1.314
Using constant expenditure shares
Replacing bread with wheat prices
Using rural weights
Using Clark rent series
Using fuel series with faggots
1.295
1.273
Clark series
1.262
Source: Feinstein, ‘Pessimism perpetuated’, pp. 652-3
8).22 This raises the question as to which assumption is better. Horrell’s
budget studies seem to suggest that constant expenditure shares approxi-
mate more closely to consumer behaviour than fixed quantity weights.
Thus the share of household budgets spent on tea, coffee, and sugar
stayed approximately constant from the 1790s to the 1860s despite tea
and sugar prices falling to about half their level relative to other goods.
The share spent on clothing rose despite clothing costs falling relative to
other costs. And the share of budgets expended on rent rose by only
about 30 per cent despite a rise of more than 60 per cent in housing
costs relative to other costs.
A second reason for the more rapid rise of the Feinstein series is the
use of bread prices from London until 1820. London bread prices
increased by 17 per cent more than wheat prices between 1770-9 and
1860-9. This might argue for using these bread prices, except that
London bread prices were fixed by an assize until 1815. Sidney and
Beatrice Webb’s discussion of how assize prices were fixed in this period
suggests strongly that they are not reliable.23 Thus the ratio of the price
of 4 lbs. of bread in London in pence to the price of a bushel of wheat
in England in shillings falls from an average of 1.36 in 1670-1769 to
1.14 in 1770-99, but then bounces back up to 1.32 in the years 1820-
69 when the assize was abolished. Again, table 8 shows the additional
effect of replacing these bread prices by wheat prices. Next, the weights
are somewhat different for the rural households. In particular, housing,
the most rapidly increasing cost, has a lower weight. Furthermore, the
rent series used here, although rising during the years 1770 to 1869 by
just the same amount as Feinstein’s, has a very different timing for the
increase, placing most of it in the earlier period when the expenditure
weight on housing was smaller. Finally, the fuel series used here, which
looks at the price of faggots as well as that of coal in the years 1770-
1830, shows more of a decline in prices. Together these effects are
22 This is not an exact calculation since I do not have the exact sub-series Feinstein used within
the food category.
23 Webb and Webb, ‘Assize of Bread’.
 Economic History Society 2001

496
gregory clark
Table 9. Farm labourers’ real wages by decade, 1670-1869 (1770-9 = 100)
Decade
North
Midlands
South east
South west
All
Feinstein
all workers
1670-9
1680-9
1690-9
1700-9
1710-19
1720-9
1730-9
1740-9
1750-9
1760-9
1770-9
1780-9
1790-9
1800-9
1810-19
1820-9
1830-9
1840-9
1850-9
1860-9
82
117
69
86
81
115
98
89
100
113
92
107
98
122
120
113
120
116
123
137
135
110
109
100
99
93
97
91
112
118
125
136
143
106
112
90
109
105
109
121
120
107
105
100
116
114
109
127
146
154
161
164
165
108
102
104
120
118
105
104
100
107
103
98
104
125
132
138
147
150
76
99
113
106
106
106
100
103
101
102
113
153
166
173
197
204
109
109
97
98
100
106
102
87
100
106
113
112
108
118
124
130
140
147
91
103
109
111
115
115
Sources: tab. 4; app. tab. A1; Feinstein, ‘Pessimism perpetuated’, app. tab. 1
sufficient to account for most of the difference between the cost of living
series. Since most of the modifications suggested above would apply to
an urban cost of living series also, the above discussion suggests that
Feinstein is too pessimistic about living standards for workers in the years
from 1815 onwards.
Table 9 and figure 8 show the resulting estimate of real farm wages
160
140
120
100
Male farm
workers
Feinstein-all
workers
80
60
40
20
0
1670 1690 1710 1730 1750 1770 1790 1810 1830 1850 1870
Year
Figure 8. Agricultural real wages, 1670-1869
Note: The figure also shows Feinstein’s index of real wages of all workers, assuming full employment
Sources: Feinstein, ‘Pessimism perpetuated’; app. tab. 1
 Economic History Society 2001

english farm wages, 1670-1869
497
by decade for England as a whole and for each of the four regions,
where the average of 1770-79 is set at 100. Real wages overall show
little sign of any trend between 1670 and 1819. Indeed, from 1770 to
1819 real wages are essentially flat. For its first 50 years the industrial
revolution does nothing to improve the living standards of farm workers.
But after 1820 a modest but sustained upward trend in real wages
appears so that by the 1850s real wages of male agricultural workers are
nearly 50 per cent higher than in the 1770s. This overall real wage trend
is a little more robust than that found by Feinstein for all workers, the
difference stemming principally from the differences in estimated living
costs. Also Feinstein finds modest gains in real wages already in the first
two decades of the nineteenth century, gains which are absent here for
agricultural workers.24
200
180
160
140
120
100
80
North and
midlands
South east and
south west
60
40
20
0
1670 1690 1710 1730 1750 1770 1790 1810 1830 1850 1870
Year
Figure 9. Agricultural real wages by region, 1670-1869
Note: The nominal wage in each of the two areas was calculated as a weighted average of the nominal wage in
each region within the area, using weights derived from the 1851 occupational census.
Source: tab. 9
Real wage trends were very different in the north than in the south of
the country. Figure 9 shows estimated real wages relative to 1770-9 for
the south east and south west combined, and for the north and midlands
combined. In the south real wages do not rise above the average level of
1670-1769 until the 1840s. Even by the 1860s real wages are only about
15 per cent above their level of the century before 1770. Further, there
is a clear decline in real wages over the years 1750-1819 compared with
the earlier years. In the north and midlands in contrast, after a period
of flat real wages from 1670 to 1809, a sustained increase begins in the
24 If readers prefer the Feinstein approach to the cost of living they can substitute his real wage
index for 1770-1869 and calculate a revised real wage series that will increase to a lesser extent in
the decades after 1820.
 Economic History Society 2001

498
gregory clark
1810s, and by the 1860s real wages are 85 per cent above their level of
the years before 1770.
V
In industrializing economies, the poorest workers are those in agriculture.
Since urban areas are providing an increasing proportion of employment
there is migration from the country to the city. To induce this rural to
urban migration, urban living standards have to rise relative to those in
rural areas. In England there was certainly a widening of the gap between
nominal agricultural wages and nominal urban wages in the years 1770
to 1869. From the 1770s to the 1860s the day wage of agricultural
workers in England increased by 90 per cent, while Feinstein’s general
wage series for Great Britain shows increases of 118 per cent. Given the
migration flows, in an industrializing economy the movement of farm
workers’ real wages will set a lower bound on the movement of real
wages as a whole. If their real wages rise then we can be sure that real
wages in the cities are rising by as much or more. Otherwise the increased
migration flow to the cities would not be occurring. What makes the
wages of male farm workers attractive also as a standard for real wages
is that their jobs and living conditions changed little in the years 1670
to 1869. The same tasks were being performed at the same times of year
from 1670 until at least the 1850s. Workers were living in the same
village settings. Real wage measures for urban workers show just what
private goods the wage can purchase, but give no measure of the costs
of urban disamenities such as higher infant mortality, pollution, conges-
tion, and unemployment risks. Nor do they measure the costs of changed
work conditions as in the move from domestic industry to the factory.
In table 9 we see that from the 1820s onwards real male farm wages
in all regions of England rise steadily from their levels of the 1770s. The
national rise is greater than Feinstein recently calculated for workers as
a whole for all of the decades from the 1820s. The reason, as we saw
above, is that while nominal male farm wages increase more slowly than
Feinstein (and earlier Williamson) found for workers as a whole, the cost
of living series used here shows a consistently lower cost of living in all
these decades. Thus Feinstein’s recent pessimism with regard to living
standards is probably too great. If farm day labourers were improving
their living standards in these years, even in the south of England, it
is hard to conceive that living standards were not also improving in
urban areas.
VI
From the farmer’s perspective the real wage that matters is the cost of
agricultural labour relative to the price of the output he sells. This will
differ from the real wages given above because the bundle of goods
produced by English agriculture was different from the bundle of goods
 Economic History Society 2001

english farm wages, 1670-1869
499
Figure 10. Indices of wages etc. measured in product prices, 1670-1869
Sources: Capital rentals: Clark, ‘Land hunger’. Land rents: idem, ‘Renting the revolution’; idem, ‘Land rental
values’. Agricultural output prices: idem, ‘Land rental values’
consumed by the workers. Figure 10 shows the day wage of agricultural
workers from 1670 to 1869 measured in terms of the price of farm
output. Farm wages on this measure rise even less during the industrial
revolution than does the general purchasing power of wages. By the
1860s they are only about 28 per cent higher than in the 1770s.
The overall productivity of agriculture—the average output per unit of
inputs—can be measured as a weighted average of the cost of the inputs—
wages, land, and capital—relative to the price of output.25 Thus, suppose
that overall productivity doubles. Then the average payment to inputs
measured in the price of the products would also have to double. Since
wages increased so little and constituted about two-fifths of production
costs in the nineteenth century, for there to be significant productivity
advance other input costs measured in product prices would have to have
risen by large amounts. Real rentals on land did rise more than for
labour, as figure 10 shows, but the rental cost of capital goods actually
fell slightly, if it is assumed that farmer’s capital had the same price as
output. The net effect is a very modest gain in overall productivity, which
appears only in the years after 1820.
Thus the wage series developed here, if it can be taken as measuring
the labour cost to farmers, implies, in combination with our knowledge
of the rental value of land and the return on capital, that the concept
that an agricultural revolution occurred at any time in the years 1670 to
1869 is mistaken, and that such advance in productivity as did occur
was concentrated in the early nineteenth century.
University of California at Davis
25 This method is explained in Clark, ‘Too much revolution’.
 Economic History Society 2001

500
gregory clark
APPENDIX: Details of wages and the cost of living
Manuscript sources on day wages
Beveridge Collection, Robbins Library, London School of Economics. Brooke, Isle of
Wight (Box I11). Delisle Accounts (Box W2). Eton (Box 119). Pelham Papers
(Box H12). Stowe Papers (Box H2). St Bartholomew’s Hospital, Sandwich (Box
E9). Winchester College (Box W4).
Bedford Record Office. Chester, BC 482, CH 938. Orlebar, OR 1372-4.
Berkshire Record Office. Craven, D/E C A3. Throckmorton, D/E We A3, A5, A13.
Buckinghamshire Record Office. Chester, D/C/2/45, D/C/2/52. Drake, D/DR/2/48,
D/DR/2/51, D/DR/2/89, D/DR/2/164, D/DR/2/166, D/DR/2/168. Hampden,
D/MH/30/5,
D/MH/30/13,
D/MH/30/15,
D/MH/30/19,
D/MH/30/21,
D/MH/33/1-28.
Cambridge Record Office. Cotton, 588/A2, A4.
Chester Record Office. Tollemarche, DTW/2343/A/3/1. Chester Treasurer’s
Accounts, TAB/7.
Cumbria Record Office (Carlisle). Curwen, D/LONS, D/LONS/W3/14,
D/LONS/L3/5/48-56,
D/PENN/202-4.
90.
Fleming,
D/S/FLEMING/21.
Pennington,
Cumbria Record Office (Kendal). Browne, WD/TE/Box11/3.
Devon Record Office. Courtenay, 1508 M/V10, 1508 M/V31, 1508 M/V37. Crosse
1160M/Accounts A9(2) (Brampton). Drake 346 M/E2-11.
Dorset Record Office. Bastard, 5339. Bragge, D 83/22. Damer, D188. Digby, KG
1229-40. Larder, PE/WCH/MI/1. Syndercombe, PE/SYM 2/5-6, KW 11A.
Durham Record Office. Salvin, D/Sa/E167-177, D/Sa/E763-5.
East Suffolk Record Office. Crowley, HA 1/GB3/2/1. Hill Farm, HA 2/B2/1.
Labour Accounts, HA 2/B7/2. HA 10/50/18/14, HA 30/369/41-43, HA
67/461/454.
East Sussex Record Office. Ashburnham, ASH/1630. Shiffner, 1998, 3510. Stapley
of Hickstead, HIC/467. Trevors of Glynde Place, GLY 2932-4.
Essex Record Office. Bentall-Linnett, D/DE/A1-2. Cook’s Farm, D/DU 441/52.
Coopersale Hall, D/DU 363/4. Farm Accounts, D/DU 441/51-4. Gidea Hall,
D/DBe/A1. Kempton, D/NM 1/1. Lennard, D/DL/E1, D/DL/E3. Petre,
D/DP/A18-20, D/DP/A22, D/DP/A55, D/DP/A57, D/DP/A59, D/DP/A214.
Tabor Family, D/DTa/A1. Tylers Hall, D/DM/A20, D/DM/A22. Ulnes Farm,
D/Dop/E15. Watkinson, D/DU 224/2.
Hampshire Record Office. Russell, 149M89/R5/6103, 149M89/R5/4613.
Wheeler, 3m51/605.
Hertford Record Office. Broadfield Hall, 70474A. Labour Accounts, D/ER96.
Hatfield, 26294. Rolt, D/EAS/21710. Sebright, 18104. Wilshere, 61589.
Huntingdon Record Office. Bernard, ddM5/4/1. Brampton, ddM5/5,
ddMM/44D/7. Houghton, ddM/44D.
Kent Record Office. Best, U480/E1-2. Croft, U709/A1. Darell, U386/A1. Gambia,
U194/A8. Rockingham, U471/A18. Sackville, U269/A69/2, U269/A70/2,
U269/A89. Tylden, U593/A7.
Lancashire Record Office. Clifton, DDCI 399. Farington, DDF 31. Hesketh,
DDHe/62/14-15, DDHe/62/25-29. Molyneaux, DDM/1/141.
Leicester Record Office. Braye, Box III. Croxton Lodge Farm, DE 814/9. Ferrers,
26D53/2335, 26D53/2465-6. Finch, DG7/1/37, DG7/1/43A. Herrick, DG9/2054.
Rectory Payments. DE667/70. Rothley, 2D31/241.
 Economic History Society 2001

english farm wages, 1670-1869
501
Lincoln Record Office. Ancaster, 2 ANC 7/7/53, 3 ANC 6/24-25, 3 ANC 9/15/165.
Brace, 23/8/1. Monson, 10/4B/13-20.
Northampton Record Office. Dryden, D(CA) 307-8, D(CA) 312, D(CA) 323.
Fitzwilliam Misc. Vols. 6, 8, 23, 50, 74, 106-7, 157, 189, 191, 239, 790. Howe,
YZ 997. Isham, IL 3945. Maxwell. Tryon, VII/619a-623. Wakefield, G2580/6.
Nottingham Record Office. Arundel Castle, W114/8. Edge, DDE1/1-37. Franklin,
DDF1/22, DDF1/122. Galway, 12375/385. Middleton, MIAV/226. Nevile,
DDN213/5. Newcastle, NEA444/1. Portland, DD4P/58/78, DD5P/1-150,
DD5P/2/6-8,
DD5P/4/1.
Savile,
DDSR/206/2,
DDSR/211/257/1-22,
DDSR/A4/49/1.
Northumberland Record Office (Berwick). Haggeston, ZHG IV/3. Simpson, ZS1/1-
4, ZS1/56-7, ZS1/94.
Northumberland Record Office (Newcastle). Allgood, ZAL Box 44/1, Box 44/10,
Box 60. Blackett, ZBH 273/2, ZBL 283/1-38. Clark, ZCL.A. Hope-Wallace,
ZHW/4/23. Swinburne, ZSW.
Reading University Library. Wyche of Hockwold, NORF, 14/1/1.
Sheffield City Library. Oakes, 1518.
Somerset Record Office. Carew, DD/TB/Boxes 12-13, 14/6, 14/9, 14/11-12. Hylton
DD/HY Box 12. Parsonage, DD/X/REE/C/1308. Popham DD/PO/32/2-97. Wil-
loughby DD/WO/Box 49/10/pt 2.
Stafford Record Office. Aqualate, D(W) 1788/V/143-168. Bagot, D1721/F/61-66.
Bradford, TP594. Gifford, D590/645-50. Hatherton, D260/M/E/86-97.
D1788/143. Leveson-Gower, D593/F/2/18-35, D593/F/3/1/1-4, D593/F/3/25-27.
Shrewsbury, D240/E/F/4/1-5, 7, D240/E/F/5/1-11. Sutherland, D593/L/6. Ver-
non, D1826/41.
Surrey Record Office (Guildford). Howard, 1/53/4, 1/53/7-8. More-Molyneux, LM
1087/1/8, LM 1087/2/5/1, LM 1087/2/16. Nicholas, 22/1/2. Shallet, 121/1/11/3.
Wyatt, LM 1087/2/8, LM 1087/3/9, LM 1087/5/1-4.
Warwick Record Office. Aylesford. Barker, CR233. Chesterfield, CR229/71.
Conway, CR114A/292, CR114A/353, CR114A/357. Leigh, DR18/385.
Northampton, CR556/275-6. Pleydell-Bouverie, 5476. Seymour, CR114A/202.
Shirley, CR229/64.
West Suffolk Record Office. Stanton Accounts, E1/11/2. Warner, 1341/5/2.
West Sussex Record Office. Goodwood, E5422-5424, E5530-5531.
Wiltshire Record Office. Ashe, 118/140B, 118/141. Ballard, 1195/22. Blandy,
116/21. Burdett, 1883/192/1. Chippenham, 811/207. Duke of Somerset,
1332/Box 10. Enford, 415/86.
Yorkshire Record Office (Sheepscar). Harewood, III 251. Ingram/Irwin,
TN/EA/12/11. Robinson/Weddell, NH 2187.
The appendix tables begin overleaf
 Economic History Society 2001

502
gregory clark
Appendix table A1. Winter farm wages and living costs by year,
1670-1869 (d./day)
Year
Observations Wage Cost of living
Year
Observations Wage Cost of living
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
11
10
9
10
13
11
11
10
9
11
9
11
10
6
8
5
6
4
4
5
9
10.5
10.5
11.5
11.1
10.6
11.1
11.2
10.7
10.5
10.6
11.1
10.6
10.4
11.2
10.6
10.6
10.1
9.8
80
77
76
89
94
80
77
82
83
78
81
80
78
76
83
75
77
68
66
73
70
77
86
95
80
95
95
101
100
91
77
72
69
75
71
67
67
73
89
100
87
89
79
86
76
79
78
76
72
77
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
12
8
9.9
9.8
10.5
10.8
10.3
9.6
76
73
75
73
76
83
79
90
92
77
72
67
66
70
75
77
74
72
70
80
89
78
69
63
65
72
72
71
75
73
73
79
81
77
78
86
100
93
78
75
71
73
76
82
91
94
104
100
91
89
13
11
13
11
11
16
12
14
11
13
11
12
16
15
15
10
8
15
16
12
16
15
19
19
22
23
16
15
20
18
19
18
17
17
15
16
19
16
19
20
18
21
20
19
19
22
21
21
10.0
10.3
10.3
10.6
10.4
10.6
10.7
11.0
11.1
10.9
11.0
10.3
11.0
11.4
10.7
10.1
11.0
10.8
10.7
10.7
10.5
10.6
10.5
10.4
10.4
10.8
10.8
10.5
10.6
10.6
10.3
10.5
11.0
10.9
11.2
11.1
10.9
11.2
11.1
11.1
11.3
11.4
11.3
11.5
11.3
10.0
10.1
9.8
12
12
7
9.9
10.1
10.4
9.7
9.9
9.8
9.8
9.8
9.7
9.8
10.2
9.8
9.9
9.5
10.1
10.7
10.9
10.5
9.8
9.9
9.4
13
10
13
11
13
13
13
13
14
17
14
16
19
18
15
13
15
15
14
11
16
16
11
11
12
12
9.6
10.1
10.1
10.6
10.5
10.8
10.3
 Economic History Society 2001

english farm wages, 1670-1869
503
Appendix table A1. Continued
Year
Observations Wage Cost of living
Year
Observations Wage Cost of living
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
16
18
16
15
11
14
14
14
13
13
15
11
12
12
10
11
11
12
13
13
16
15
13
14
13
18
17
16
19
19
21
19
20
20
22
22
24
23
24
27
28
26
26
23
23
26
28
28
26
23
11.5
12.0
12.3
12.1
12.5
12.6
13.1
12.7
12.6
13.1
13.1
13.3
13.4
13.4
13.6
13.4
13.2
13.5
13.9
13.7
13.7
13.8
14.3
14.2
14.6
15.0
16.1
16.6
17.1
17.3
18.3
18.2
18.2
19.1
19.1
19.7
20.0
20.0
20.6
21.3
22.1
21.8
23.5
24.5
24.0
22.8
22.2
22.4
22.4
22.2
96
101
105
107
107
102
93
100
99
90
91
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
24
26
28
27
23
27
25
23
24
24
23
26
25
26
29
25
23
29
31
28
27
24
21
22
21
17
16
17
15
12
15
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
22.0
21.0
19.2
18.7
20.1
21.0
20.5
20.5
20.7
20.7
20.8
21.0
20.7
20.5
20.2
19.5
19.3
19.9
20.3
21.2
20.2
20.9
21.1
20.3
20.5
20.5
21.1
21.7
21.0
20.4
19.3
19.1
19.2
20.9
22.6
23.2
23.4
23.1
22.5
22.4
22.9
23.2
23.0
22.8
22.9
23.3
24.1
24.8
25.1
24.6
144
129
114
120
135
142
134
134
126
132
130
132
121
118
115
106
117
125
128
140
135
131
124
110
115
112
120
141
112
108
101
99
100
117
138
138
133
128
114
113
129
129
128
118
113
117
126
142
138
122
99
102
109
105
98
97
99
101
106
110
106
102
109
111
134
138
119
120
139
180
187
142
134
141
168
158
154
165
180
188
179
212
203
170
153
156
171
168
172
Sources: The manuscript sources listed above, as well as Bacon, Report; Brassley, Accounts; Eccleston, ‘Survey’;
Fox, ‘Agricultural wages’; Gilboy, ‘Labour’; John, ‘Statistical appendix’, pp. 1089-98; Richardson, ‘Agricultural
labourer’s standard of living’; idem, ‘Standard of living’; Thorold Rogers, History, pp. 646-703.
 Economic History Society 2001

504
gregory clark
Appendix table A2. The rural cost of living, 1670-1869
Decade
Grain Beer Meat Bacon Dairy Sugar Tea Fuel Candles Rent Clothing All
and soap
1670-9
1680-9
1690-9
1700-9
1710-19
1720-9
1730-9
1740-9
1750-9
1760-9
1770-9
1780-9
1790-9
1800-9
1810-19
1820-9
1830-9
1840-9
1850-9
1860-9
75
66
87
68
73
75
65
64
79
84
74
72
76 108
72
74
73
69
74
75
84
90
91
79 115
76 108
82 133
72 114
71 105
115
114
109
110
100
97
97
97
98
95
84
76
86
78
89
83
74
92
87
100
100
106
115
150
165
101
102
110
122
113
94
97
94
82
76
90
77
80
80
73
73
82
102
102
102
102
80
80
63
75
77
79
83
83
83
83
87
88
86
90
90
90
91
106
109
109
104
104
103
103
100
101
98
103
105
90
74
69
75
73
96
87
74 104
83 106
63
96
83
90 100
87
100 100 100 100 100 100 100 100
99 119 102
100
111
112
167
212
213
215
196
198
247
100
100
118
160
174
130
123
120
118
126
98 100 104
85 100
61 122
86 155
89 154
70 128
56 103
124 130 127 101 120 152
181 160 189 130 163 103
194 165 198 141 188 127
130 149 162 125 161
123 125 136 104 154
121 114 137 105 153
117 139 149 110 142
114 138 171 123 159
80
74
79
63
59
46
50
58
92
88
89
63
81
Notes: The index for each commodity and overall is set to 100 for 1770-9. The commodities and the weights
used for each category of good where more than one was used were: Grain, flour and bread: before 1820, wheat
(0.80), barley (0.05), oats (0.15), after 1820, wheat (0.90), barley (0.033), oats (0.067). Meat—beef (0.5), mutton
(0.5). Dairy: before 1820, cheese, (0.31), butter, (0.34), milk (0.35), after 1820, cheese, (0.29), butter, (0.35),
milk (0.36). Fuel: faggots, 1670-1830 (0.5), coal, 1770-1869 (0.5). Light and soap: tallow candles, 1670-1830
(0.5), tallow, 1670-1869 (0.5). Cottage rents: charity owned housing outside towns and cities, 1680-1869. Clothing:
before 1770, wool cloth, 1670-1769 (0.5), linen cloth, 1670-1769 (0.2), stockings, 1710-69 (0.1), shoes, 1670-
1769 (0.2); after 1770, from Feinstein.
Sources: Bowden, ‘Statistical appendix’, pp. 828-31, 843-6. Beveridge, Prices and wages, pp. 85-90, 143-8, 193-6,
236-40, 292-5, 313, 434-7, 457-8. G. Clark, ‘Shelter from the storm: housing and the industrial revolution’
(Working paper, Univ. of California, Davis, 2000; http://www.econ.ucdavis.edu/faculty/g.clark). John, ‘Statistical
appendix’. Feinstein, ‘Pessimism perpetuated’, p. 640. Sauerbeck, ‘Prices of commodities’. House of Commons,
Report on Wholesale and Retail Prices.
Footnote references
Allen, R. C., Enclosure and the yeoman (Oxford, 1992).
Bacon, R. N., The report on the agriculture of Norfolk (1844).
Beveridge, Lord, Prices and wages in England, 1: the mercantilist era (1939).
Bowden, P. J., ‘Statistical appendix’, in J. Thirsk, ed., The agrarian history of England and Wales,
V.II (Cambridge, 1985), pp. 827-902.
Bowley, A. L., ‘The statistics of wages in the United Kingdom during the last hundred years, I:
agricultural wages’, J. Roy. Stat. Soc., 61 (1898), pp. 702-22.
Brassley, P. Lambert, A., and Saunders, P., eds., Accounts of the Reverend John Crakanthorp of
Fowlmere, 1682-1710 (Cambridge, 1988).
Clark, G., ‘Land hunger: land as a commodity and as a status good in England, 1500-1910’, Exp.
Econ. Hist., 35 (1998), pp. 59-82.
Clark, G., ‘Renting the revolution’, J. Econ. Hist., LVIII (1998), pp. 206-10.
Clark, G., ‘Too much revolution: agriculture and the industrial revolution, 1700-1860’, in J. Mokyr,
ed., The British industrial revolution: an economic assessment, (Boulder, Col., 1999), pp. 206-40.
Clark, G., ‘Land rental values and the agrarian economy: England and Wales, 1500-1912’, Eur.
Rev. Econ. Hist. (forthcoming).
Clark, G. and van der Werf, Y., ‘Work in progress: the industrious revolution?’, J. Econ. Hist.,
LVIII (1998), pp. 830-43.
Eccleston, B., ‘A survey of wage rates in five midland counties, 1750-1834’, (unpub. Ph.D. thesis,
Univ. of Leicester, 1976).
Feinstein, C., ‘Changes in nominal wages, the cost of living and real wages in the United Kingdom
over two centuries’, in P. Scholliers and V. Zamagni, eds., Labour’s reward: real wages and economic
change in nineteenth- and twentieth-century Europe (Aldershot, 1995), pp. 3-36.
 Economic History Society 2001

english farm wages, 1670-1869
505
Feinstein, C., ‘Pessimism perpetuated: real wages and the standard of living in Britain during and
after the industrial revolution’, J. Econ. Hist., LVIII (1998), pp. 625-58.
Fox, A. Wilson, ‘Agricultural wages in England and Wales during the last half century’, J. Roy.
Stat. Soc., XLVI (1903), pp. 273-348.
Gilboy, E. W., ‘Labour at Thornborough: an eighteenth-century estate’, Econ. Hist. Rev., III (1932),
pp. 388-98.
Gilboy, E. W., Wages in eighteenth century England (Cambridge, Mass., 1934).
Hobsbawm, E. J. and Rude´, G., Captain Swing (1969).
Horrell, S., ‘Home demand and British industrialization’, J. Econ. Hist., LVI (1996), pp. 561-604.
John, A. H., ‘Statistical appendix’, in G. E. Mingay, ed., The agrarian history of England and Wales,
VI: 1750-1850 (Cambridge, 1989), pp. 1089-177.
Lindert, P. H. and Williamson, J. G., ‘English workers’ real wages: reply to Crafts’, J. Econ. Hist.,
XLV (1985), pp. 145-53.
Richardson, T. L., ‘The agricultural labourer’s standard of living in Kent, 1790-1840’, in D. Oddy
and D. Miller, eds., The making of the modern British diet (1976), pp. 103-16.
Richardson, T. L., ‘The standard of living controversy, 1790-1840, with special reference to
agricultural labourers in seven English counties’ (unpub. Ph.D. thesis, Univ. of Hull, 1977).
Rogers, J. E. Thorold, A history of agriculture and prices in England, 7.i (Oxford, 1902).
Sauerbeck, A., ‘Prices of commodities and the precious metals’, J. Stat. Soc. London, 49 (1886),
pp. 581-648.
Young, A., A six months’ tour through the north of England (1771).
Young, A., A farmer’s tour through the east of England (1771).
Young, A., A six weeks’ tour through the southern counties of England and Wales (1771).
Webb, S. and Webb, B., ‘The Assize of Bread’, Econ. J., 14 (1904), pp. 196-218.
Official publications
House of Commons, Report on the Poor Laws in England and Wales, Appendix (B), Part 1. Sessional
Papers, XXX (1834).
House of Lords, On the Burdens Affecting Real Property, Appendix No. 2 (1846, XXIII).
House of Commons, Report on Wholesale and Retail Prices in the United Kingdom. Sessional Papers,
LXVII (1903).
 Economic History Society 2001