Paper for the work-shop The impact of markets on the management of rural land,
Zaragoza, Spain, September 2006, arranged by COST Action A-35
Programme for the Study of European Rural Societies.
Revised version 2007-03-29
The commercialisation of the peasant economy – markets and
agricultural production in southern Sweden 1711–18601
Mats Olsson and Patrick Svensson
Department of Economic History
Lund University
Sweden
Mats Olsson/Patrick Svensson
Dept. of Economic History
P. O. Box 7083
S – 220 07 Lund
Sweden
[email protected]
[email protected]
1
This work has been done within the project “Economic Development and Social Dynamics. Swedish
Agricultural Transformation in European Perspective” financed by the Swedish Research Council. We are
grateful to Kirk Scott for help with the statistical programming. We are also grateful to Michael Kopsidis and
other participants at the COST-workshop in Zaragoza 2006 for valuable comments and suggestions.
1
Introduction: Peasants and markets
In pre-industrial Sweden, the vast majority of the land was managed by peasants, either as
owners or as tenants. A commercialisation of the agrarian sector therefore implied a
commercialisation of the peasant economy. This theme of commercialisation of the peasant
economy has over time preoccupied many researchers. There are several questions discussed
in this context: Were peasants voluntarily taking part in markets or were they forced into the
markets? What incentives were important for the peasant household? How did markets
change the organisation and production in agriculture, i.e. the management of land?
Peasants were almost always to some extent commercialised; to be able to pay taxes and buy
necessities, as salt, the peasant household had to sell some of their production. Some kind of a
consensual on traditional “peasantries” and commercialisation in Europe was formulated by
Frank Ellis in his synthesis stating that “Peasants are….fundamentally characterized by partial
engagement in markets which tend to function with a high degree of imperfection.” (Ellis
1988, p.12, see also Scott 1998, pp. 7–9) This fundamental degree of commercialisation is,
however, not the one we are primarily interested in. Instead, it is the transformation from the
self-sufficient household with a small marketed surplus to the specialised, commercialised
household that is of interest (see de Vries 1975:207). This transformation includes
reallocation of labour time in the household as well as changes in organisation and
management of land itself.
Different incentives encouraging the peasant household to take part in markets have been
proposed. One such incentive is an increase in prices for products produced by the peasants. A
sustained increase in, for example, grain prices lead to land reclamation and rising prices on
land (Slicher van Bath 1963). Investments in agriculture increases too, due to as well rising
income among the peasants as through an increased amount of borrowing. Moreover, by land
reclamation the cultivated area increases and production is rising. Due to diminishing returns
on land, and the limits of arable land available, sustained price increases imply more intensive
production as well. New crop rotation and other organisational innovations are introduced
(Dickler 1975).
The extent to which prices affect production is depending on the way prices are set. If prices
are wholly dependent on harvest, total revenue for the peasants would change to a lesser
2
extent than if prices were set exogenously; in a non-integrated market good harvests could
lead to lower total revenue for the farmer as compared to revenue at normal harvests. This
occurs if prices fall so much that it counteracts the increase in the amount of grain sold. Thus,
a more open market, leading to less volatile prices, is preferential not only for consumers but
also for producers (Persson 1999). Falling transportation costs and progress in information
systems during the 17th and 18th centuries paved the way for gradual market integration in
many regions of Europe. The result of market integration was less severe price fluctuations
resulting in possibilities of liberal trade reforms substituting regulation of grain markets.
Together this led to improved performance of grain markets and thereby increased
investments in grain production and further commercialisation (Persson 1999).
However, some researchers have argued that price increases or, for that matter, price stability
are not sufficient conditions to transform self-sufficient farms to commercialised farms. A
necessary incentive for commercialisation is instead brought about through changing
consumption patterns among the peasants (de Vries 1975). When superior goods was
provided from the outside, peasants desired these goods and specialised in agricultural
production in order to be able to buy them. This was done through the devotion of less time
on non-agricultural work and more on agricultural work. The non-agricultural products are
bought from markets in return for agricultural products sold. The most common example of
such a product is textiles (de Vries 1975; Schön 1997). One pre-condition for this model of
commercialisation is the existence of a sector producing items demanded by the peasants.
We must bear in mind that not all researchers are inclined to believe that peasants reacted to
prices and to supply of products from outside the peasant economy. This group of researchers
ranges from ethnologists claiming that peasants were conservative, inward-looking, attached
to traditional values and kinship (e.g. Wolf 1966) to others emphasising the risk-averse
behaviour among peasants (Brenner 1976; Scott 1976). According to these researchers, the
commercialisation of agriculture came through peasants being forced into markets, not to
them responding to economic incentives. On the other hand, some researchers argue that
peasants did react to economic incentives (e.g. Bjørn 1988; Hoffman 1996; de Vries 2001).
The commercialisation of the peasant economy came through economic and institutional
changes which peasants exploited in order to improve their economic situation. Market
involvement was not a more risky business than staying outside markets, according to de
Vries (2001); crop yields tended to be more volatile than markets. It was the efficiency of the
3
market that was decisive for peasant participation. Our point of departure is that peasants did
react to institutional and economic change. This assumption is based on previous research as
well in Sweden as in other countries (Grantham 1975; Hoffman 1996; Popkin 1980; Svensson
2001, 2006; de Vries 2001).
Conclusively, increase in prices or price stability are potential forces affecting investments in
production and commercialisation as is the emergence of a non-agricultural production
directed towards the peasants. Our aim is to study the commercialisation of the peasant
economy and its repercussions on the management of land. This is done by estimating the
degree of commercialisation on individual farms and to relate this to changes in prices and to
organisational and institutional changes. We study a large sample of parishes in the province
of Scania in southern Sweden during the 18th and 19th century representing all types of
peasant households present in this region.
Commercialisation of agriculture in Sweden 17th to 19th century – obstacles
and possibilities
Commercialisation and market extension is associated with changes in demand and supply.
For peasants in southern Sweden, producing grain and animals, demand came basically from
three different sources. First, and foremost, local demand was important. In the early 18th
century the vast majority of the population possessed farms and produced their own food
supply, but this changed over time. The share of landless in the population reached almost
fifty per cent already in the second half of the 18th century and this group was dependent on
food from the local farms (Lundh 1999). The landless population was mainly working as farm
labourers and as such some of the payment could come as free meals at the farm. However,
most of them had their own households and, thus, had to buy their food. Sweden experienced
a population growth where population doubled during the period 1750 to 1860. At the same
time the number of peasants increased by 25 per cent, while the number of landless more than
quadrupled (Wohlin 1909).
Second, local natural conditions and traditions led to differences in production mixes. In the
more wooded areas soil type and fertility favoured animal and wood production while on the
4
plains farms tended to specialise in grain. The local specialisation made trade of animals and
grain between these areas a necessity (Hanssen 1952).
Earlier research has found that other factors than harvest was important for the grain price in
19th century Scania, e.g. trade and harvest outside the area (e.g. Dribe 2000). Additionally,
turning to comparisons with other prices we find that correlation coefficients 1711–1860
between Scanian rye prices and Amsterdam and Stockholm were 0.74 and 0.96 respectively.
When all trends are eliminated, using first difference, the same correlations were 0.36 and
0.67 respectively. 2 Furthermore, the short term variations in price and production are only one
aspect of commercialisation. Prices in the long run are both a reflection of changing economic
and demographic conditions and important incentives for producers. Grain prices in the area
studied were certainly rising during the 18th and 19th century. Particularly important periods
were the periods around the Napoleonic wars and the late 1840s and beginning of the 1850s
when Swedish grain exports exploded (Fridlizius 1957; Schön 2000). Since prices only partly
were dependent on harvests, this means that peasants who produced a surplus faced increasing
income during these periods. This is even more so since real wages for the agricultural
labourers were falling at the same time.
Hence, and third, there was demand coming from outside the regional rural areas. Cities and
towns in Sweden were small and for most of them, except Stockholm, it was only the very
close rural areas that were directly involved in supplying the urban population with food.
More important was the national interior market where deficit areas bought food from
surplus-producing areas. One of the aims for the expansion of the kingdom of Sweden during
the 17th century was to be nationally self-subsistent with food; the conquering of the provinces
of Scania and of Livonia at the Baltic sea, both grain surplus-producing areas, was in line with
this idea. A further consequence of this aim was that grain exports were prohibited, but even
so, Sweden was a net-importer of grain until the first decades of the 19th century. From the
1820s exports of grain started to be of significance and by 1850 it had become one of the most
important export goods in the Swedish economy (Schön 2000). For southern Sweden, and
Scania in particular, this meant that when the province was conquered from Denmark in 1658
earlier export possibilities were banned and Scanian grain was sold within Sweden until new
possibilities emerged in the 1820s.
2
Netherland prices from International Institute of Social History.
5
From establishing that there was a demand and that it increased over time it follows that if a
market and commercialisation were to emerge it required a surplus production of grain and
animals. Land itself poses the first factor influencing the size of the supply and the possibility
of increased supply. Scania consisted of two major types of natural conditions: Part of the
landscape hade excellent soil conditions and extensive plains, and part of the landscape had
poorer soils and was forested and hilly. In plain-land 18th century southern Sweden the openfield system with a three field rotation dominated. Major staple crops were rye and barley,
sometimes supplemented by oats; less frequently wheat was cultivated. On the plains more
than fifty percent of the land was arable in the early 1700s (Gadd 2000). The arable land was
supplemented by meadows and wastelands providing fodder and pasture for the animals.
Extending the size of the arable land was thus possible but would ultimately result in less
fodder and grazing for the animals, reducing the possible number of draught animals and,
hence, reducing manure for the fields. This dilemma could only be solved by new methods of
organisation including the cultivation of fodder crops on the arable land or by improved
technology decreasing the necessary number of draught animals (Gadd 1983).
By contrast, forests and wastelands constituted the majority of the land in the rest of Scania.
In these more forested areas the farms in the villages used the open-field system but with
two–field or no rotation. Here also single farms existed, often on more remote places, having
their land isolated from other farms’ land. Rye and, particularly, barley was cultivated
alongside other crops of minor importance such as oats, buckwheat, beans, turnips and peas
(Enclosure Acts). Non-arable land, potentially ready for conversion to arable land, existed to a
higher degree here than on the plains but at the same time the fertility of this land was
considerably lower. Since this land most often consisted of forests, swamps, and rocky and
stony land this conversion also was costly in terms of labour time.
Another factor influencing supply is the organisation of production and the possibilities of
changing this organisation. The open-field systems were managed on the village level by a
village council. This implied collective decision making regarding what crops to sow, when to
sow and when to harvest. There were also rules governing when animals were let out in the
fields and restrictions on the number of animals (e.g. Dahlman 1980; Overton 1996). The
introduction of new crops or new types of rotations was, thus, not an individual but a
collective decision.
6
A first wave of enclosures took place in Scania during the second half of the 18th century.
These enclosures (Storskiften) were rearrangements within the existing open-field system
decreasing the number of strips for each farm. Often these rearrangements implied some kind
of progress in farming as well, e.g. new or improved ditching. More radical enclosures,
breaking the village organisation and the open-field system, were introduced with the
enclosure acts of 1803 and 1827. From one year to another the old villages, with their dense
settlements and open field systems, ceased to exist. The modern landscape of the lonely and
efficient farmer was formed. Soon the natural meadows started to decrease and the fodder
production was integrated in new crop rotation systems (Olsson 2005).
A third factor was socio-economic differences existing within the peasant group. In Sweden
the peasantry was dominated by yeomen freeholders and Crown tenants since the Middle
Ages. They held a unique economic and political position, displayed by the fact that they
continuously held their own stand in Swedish parliament, even during the aristocracy’s wars
of conquest in the 17th century.
The province of Scania was divided in terms of property rights; half of the landscape was
dominated by the nobility, and half of the landscape was dominated by the peasantry. Noble
land was managed partly direct under the manor and partly by tenants under the nobility.
Peasant land was managed by freeholders (self-owners) and tenants under the Crown; the
situation of these latter groups was in many respects the same.
The nobility expanded manorial production, by eviction of tenants, in two major waves. The
first was in the 16th and the beginning of the 17th centuries; the second started in the end of the
18th century and lasted through the 19th century. The manorial part of Skåne completed its
way towards Gutsherrschaft during the agricultural revolution. This meant, by the year 1850
that in many districts half of the acreage was tilled directly under the manor and corvée duties
had often increased to 300–400 days per year for each remaining peasant, implying that the
tenants had to hire men to work for them on the demesne (Olsson 2002, 2006). For the
peasants on noble land, being tenants under a manorial landlord, this development implied
uncertain rights of possession of land and heavy work obligations on the demesne.
7
The half of Skåne that was dominated by Crown tenants and freeholders is in glaring contrast
to this development. By the end of the 17th century a new and rigid taxation system was
imposed, that in practice petrified the taxation level for these peasants, over time creating a
prosperous development in contrast to the one experienced by the tenants under the nobility
(Olsson 2005). Furthermore, most of the Crown tenants purchased their farmsteads from the
Crown already before 1850. These peasants were active in the agricultural transformation and
the enclosure movement, and the social and economic differentiation started “from below” in
the commercialisation process (Svensson 2001). The possibility of increasing production
might therefore have differed between peasants with different property rights.
From the middle of the 18th century the Swedish state became more active in promoting
agriculture. Besides issuing enclosure acts, other laws enhancing commercialisation and
investments were passed. In 1747 it became legal to split farms into smaller units, in 1775
internal trade was liberalised and in 1789 property rights for freeholders and Crown tenants
were strengthened (Magnusson 1997; Persson 1999). The deregulation of the grain market
continued during the first half of the 19th century with the abolishment of city tolls in 1810
and with the eradication of the export prohibition in the 1820s.
To summarise, natural conditions, organisation of management of land, property rights and
institutional changes affected the possibilities of increasing supply, answering to growing
demand. These different obstacles to and possibilities for production will be explored in detail
in the empirical analysis below.
Previous estimations of production and productivity
In order to be able to fulfil our aim we must estimate changes in production over time. This
puts us in the middle of an ongoing research tradition dealing with production and
productivity in historic Europe. This research is particularly strong with regards to the
development of British agriculture (e.g. Allen 2000; Clark 1991, 2006; Overton and Campbell
1996; Turner et al 2001). Here, a number of sources and methods have been used to estimate
production, productivity and real wages.
A first approach has used more “direct” sources on production, such as farm accounts (mainly
from manorial demesnes), tithes, probate inventories, and different sources containing
8
information on the size of arable land. In this tradition grain yields, crop mix, total area and
land use are important factors when estimating production development. This approach,
aggregating local micro series and combining them with macro series on land, gives rise to
specific problems; different sources exists for different periods of time, and, of course, the
problem of how representative the local sources are for national estimates (see Overton and
Campbell 1996). A second approach, moving from the top-down, estimates productivity (total
factor productivity, labour productivity etc.) and real rents using information on prices,
wages, income and population (see Allen 2000). Sometimes these two approaches are in
different forms used together in order to get a consistent picture of the development.
So far, no coherent picture of the production and productivity development in Britain has
emerged. Differences in estimation are largest for the Middle Ages while estimates on the 19th
century are more similar. 3 How large the difference in production and productivity between,
for example, the 14th and 18th century was and when, or if, growth took place during this
period is still under debate. Also for other parts of Europe estimates of GDP per capita in the
long run has generated results for different countries on agricultural production development
(e.g. van Zanden 1999).
For Sweden, the possibilities of estimating the production are less, at least for the Middle
Ages, due to lack of sources. For the early modern period, there is to a higher degree
similarities with the British and Continental possibilities. Grain yields and acreage arable land
are found in manorial accounts, in probate inventories and in enclosure acts. This has been
used in order to estimate the development of agricultural production from the middle of the
18th century to the middle of the 19th century, but only on a local level (Palm 1997; Gadd
2000). Wages, prices, and population numbers for Sweden from the 18th and 19th century have
made it possible to estimate changes in consumption and, thus, in production (Schön 1995).
Official statistics that are at least satisfactory reliable exists from 1855. Together these
estimates give us a picture, although based on several assumptions on an aggregated level, of
the agricultural development in Sweden from the beginning of the 19th century. However, the
research on this matter is still in its initial stage and the production development during large
parts of the period of the agricultural transformation is still unknown.
3
However, there is still an ongoing debate on the development in the 18th century, particularly 1750 to 1800, see
Allen 1992; Overton 1996; Turner et al 2001.
9
Data
Our case rests on the assumption that what is produced on each farm is either consumed or
marketed. Thus, to be able to calculate changes in the degree of commercialisation among the
peasants over time we must do two things: first, establish a measure of the yearly production
on each farm and, second, estimate the level of self-sufficiency, i.e. what is not available for
the market.
The first task is executed through the use of flexible tithes. 4 In the southernmost province of
Sweden, Scania, grain tithes were divided into three distinct parts, with three distinct
recipients: the Crown, the church and the local clergy. The two former parts of the tithes were
regulated to a fixed yearly amount per farmstead in 1683, and stayed unaltered until the
abolition of tithes in 1904.
However, the clergy’s grain tithes in many Scanian parishes remained a flexible production
tax until the middle of the 19th century. The same goes for the animal tithes, which in its
entirety were reserved for the parish priest, together with some other, smaller obligations. The
priest held close and elaborate account on each payment and current account on each
peasant’s animal breeding. This was made possible by the fact that the priest himself was an
active farm manager living next door to the peasants paying the tithes.
The data set contains 21 parishes with tithe rolls exceeding 25 continuous years with more
then 1 400 decimants. Together they cover the period 1711 to 1860 reaching a total of over
55 000 farm tithe payments, an average of 360 farms per year. The material represents a broad
selection of the province’s geographical and socio-economic conditions.
4
The 1960s and 1970s saw an extensive research on tithes in many European countries. The questions at issue
were typically agricultural production output and its interaction with demographical change, from the Middle
Ages until the beginning of the 19th century. In many countries there are excellent tithe series. On individual
farm level there are e.g. in Hungary series from the 16th century to the abolition of tithes 1848. However, often
these series do not reflect the total farm production, since part of the land was exempted from tithes and the
peasants were creative in finding ways to avoid payment. The results of this research has mainly concerned the
period before 1800. They indicate an increase in agricultural production in the 18th century for major parts of
western and central Europe. This development, however, was not in any way uniform (see Le Roy Ladurie &
Goy 1982). For Scandinavia aggregated tithes in Swedish counties has been used in a study on agricultural
development in the 16th and 17th centuries. Uncertainties about what really is included in these tithe series
(number of decimants, change in collection premises, etc.) have led to the conclusion that they can not be used
for simple production output estimates (Leijonhufvud 2001). Tithe rolls on individual farm level has, however,
been used by Swedish researchers to estimate grain production, Helmfrid 1949, Olsson 2005 and Berg, work in
progress.
10
For each farm in the sample the yearly tithe payments have been registered into the database. 5
The period covered is 1711 to 1860 but no individual farm is present during the whole period.
The differences in length of observation vary because the period for which tithe accounts are
preserved varies for different parishes and because some farms converted flexible tithes to
fixed tithes during our period of investigation.
The tithe accounts contain information on individual crops, for most farms rye, barley and
oats, but for some also wheat, buckwheat, peas or beans, displaying the differences in crops
produced in different areas of Scania. Every thirtieth sheaf from each peasant was annually
collected directly from the field to the clergy’s barn, and every tenth calf, foal, piglet, lamb
and gosling also found its way to the parsonage.
Since a sheaf is neither a weight nor an exact volume measure, we must elaborate it. For this
purpose we have calculated how much grain was threshed out of the collected sheaves. We
use the clergymen’s threshing accounts from one plain-land parish and one parish situated in
the more wooded part of Scania. From these accounts two things are evident: Output differed
as well between crops as between regional areas (see table 1). First, in both areas, output from
one sheaf of barley, mixed barley and oats was 1.5 to 2 times more in volume than from one
sheaf of rye or wheat. However, the weight of the grain differed as well, which is reflected in
the prices of the specific crops. Hence, in order to calculate the total grain tithes we have
multiplied each crop’s number of sheaves with their relative output (1 for rye and wheat, 1.5
for barley and mixed barley and 2 for oats) and then multiplied this sum with the relative price
between the different crops for each year.
Table 1. Average grain output per sheaf (in barrels, one barrel is 165 litres) for different crops
in two parishes in Scania 1783–1827
Billinge 1783–1814
Halmstad
1813–
1818
Rye
Wheat
Barley
0.15
0.23
0.15
0.28
0.22
0.47
Mixed
barley
0.26
n.a.
Oats
0.29
0.54
Sources: Tithe rolls for Billinge and Halmstad parishes, Regional Archives in Lund.
5
The data is registered in the Scanian Database of Agricultural History, Department of Economic History, Lund
University, Sweden.
11
Second, the output per sheaf was, for all crops, in the plain-land parish almost double that of
the wooded parish. This finding is corroborated by earlier findings on differences in sheaf
output for different regions of Scania (Weibull 1952). In order to correct for this, the sheaf
sums must be halved for the parishes in the wooded parts of Scania. This measure, the
weighted sum of all grain tithes for each individual farm and year corrected for output
differences, constitutes the dependent variable in the grain analyses.
In order to calculate the degree of commercialisation we must put this in relation to a selfsubsistence level. In an imperfect market economy each household had to produce as much
grain as to fill its own consumption plus the amount necessary to pay taxes and farmhand
wages and to purchase necessities such as salt and iron. However, for some parts of the region
farms did not produce enough grain to cover all these expenses. Following land surveyors’
descriptions of villages in different parishes during mid-eighteenth century it is evident that
taxes was paid by the surplus produced on the farms and that in certain regions this was
animals or wood but not grain; it is explicitly stated that these parishes produced no grain
surplus (Enclosure Acts). Using this information we can establish a self-subsistence level of
grain production. The grain tithes paid in these villages around the middle of the 18th century
can thus be used as a minimum level of production also on farms where they paid tax in grain.
Everything produced above this level was potential goods for the market.
Finally, the level of self-subsistence is dependent on the existing household size at that
particular point of time, mid-eighteenth century. With increasing production it is likely that
more labour were needed and that some of this labour consumed within the household.
Information on household size and composition is found in the poll tax registers. We have
used yearly poll tax registers to establish household sizes. Correcting the number of persons in
each household with consumption by age (1 for adult males, 0.7 for adult females and 0.5 for
children and old people, see Myrdal 1933) generates the consumption units per household. As
we can see from table 2 the average consumption units differed by the size of the farm. The
level remained fairly stable over time until the 1810s and 1820s when a significant increase in
household size occurred. We correct for this by increasing the self-subsistence level for this
period and onwards by the proportional change in household size.
12
Table 2. Average consumption units per household by size of farm, five parishes in Scania.
Parish
Hög
S:t Ibb
Halmstad, Hög, Kävlinge
and Sireköpinge*
Harjager’s district **
Year
Small farm
1767
1800
1830
1772–1833
1770
1780
1790
1800
1810
1820
1830
1830–1859
3.2
3.5
3.5
3.3
1822
3.5
Mediumsized farm
3.5
4.2
4.1
3.7
3.0
2.8
3.7
3.2
2.9
4.4
4.0
Large farm
5.0
5.8
3.7
4.6
5.2
5.5
4.6
4.4
4.3
4.5
3.8
4.6
6.1
6.2
6.2
Sources: Hög parish, Scanian Demographic Database; St Ibb parish, Poll tax registers, Regional
Archives in Lund; *Dribe (2000), **Andersson (2007).
When it comes to animal production we have restricted ourselves to horses, cows and oxen.
Every tenth foal and calf was paid in tithe so here we have absolute numbers of animals born
every year on each farm. We have created a variable measuring beast production by adding
foals and calves, corrected for price differences; cows commanded about two thirds of the
price for horses.
The self-subsistence level for animals varied according to different needs for draught animals
in ploughing. On the stiff clay-soils three to four pairs of beasts were needed, while on the
sandier soils it was sufficient with one pair of beasts per farm. We use accounts from land
surveyors for different villages to classify them according to the ratio needed on average.
This leaves us with a variable measuring the commercialisation level for animal production
corrected for number of draught animals needed.
The explanatory variables are collected through the use of a multitude of sources. Poll tax
registers are used to obtain information on ownership of farms. In the analyses we group
ownership according to the discussion earlier: Freehold land, Crown land and noble land
bought by peasants constitute one group and all manorial land constitutes the other group. The
13
poll tax registers are also used to supplement the tithe registers when it comes to farm sizes,
measured in mantal. 6
Table 3. Descriptive statistics of the total sample
Variable
Covariate
Per cent of observations:
Type of land
Noble land
Freehold/Crown land
1711–1860
1711–1789
1790–1860
40
60
35
65
43
57
Manager
Peasant
Tenant to peasant
Demesne
Servants
Person of rank
94
3
0.6
0.7
2
92
4
0.7
0.4
3
95
3
0.6
0.8
0.9
Early enclosures
Not enclosed
Enclosed
Re-enclosed
66
29
5
84
15
1
56
36
7
Radical enclosures
Solitary unit
Not enclosed
Enclosed
9
77
14
6
94
–
11
67
22
0.29
6.11
0.32
5.96
0.28
6.20
Mean values:
Farm size
Soil conditions
Mantal
Land fertility (village level)
Enclosures occurred at different times in different villages. They were only carried out if at
least one landowner in a village applied for enclosure. Time of enclosures for the different
villages is found in the Land Surveyors office and all enclosures have been registered in the
database. We use three different variables measuring the effect of enclosures. For the early,
less radical, enclosures, starting in mid-eighteenth century, we estimate the effect of enclosure
using a dummy variable. However, in some villages early enclosures occurred more than
once, e.g. further reductions in the number of plots, why another dummy variable measures
the effect of re-enclosure. For the radical enclosures, implying the abolishment of the openfield system and the introduction of individual management of land, we estimate the effect of
this using a dummy variable. Finally, some farms were isolated and had all their land in one
6
Mantal is a tax assessment unit originally supposed to reflect the production capacity of each farm: the higher
the mantal the larger the farm’s production capacity. Since the mantal was a rigid unit, the correspondence
between the actual production capacity or the size of the land and the mantal lost relevance over time. However,
estimations show that there still was a correspondence between mantal and size in the early 1800s (Olai 1987;
Svensson 2001).
14
unit already from the start why they had no need for a radical enclosure. The effect of being a
solitary unit, as compared to being in the open-field system or being enclosed, is also
measured using a dummy variable.
Natural conditions affecting agriculture varied quite much in Scania. The database contains
information on soil fertility on village level emanating from studies made in the 1970s. The
study classifies land from 1 to 10 where 10 indicate the best fertility, and our variable reflects
the mean fertility value for each village (see Bohman 2007 for a more extensive description).
Institutional period factors affecting possibilities of trade such as the abolition of trade
restrictions for the rural countryside in 1775, the abolishment of the interior toll in 1810 and
the abolishment of grain export prohibition in 1825 are used as dummy variables measuring
commercialisation before and after these changes.
Finally, nominal prices of agricultural products are used as explanatory variables. For grain
the correlation between the staple crops are very high (see figure 1) so we use rye prices as
indicators of grain prices. In the analysis we estimate two different potential effects of prices.
First we study the price level, following assumptions on price increases as important for
commercialisation. Here we use the mean price for the seven years preceding the tithe
payment (mean (pt-1 + pt-2 + pt-3 +…pt-7), where p is the nominal price of rye and t is the year
of production). Studying the effect of price stability we calculate the variation coefficient for
the price of rye for five years proceeding the year of production (t-1 to t-5) and from this value
we subtract the variation coefficient for the years t-6 to t-10; a small value indicates higher
price stability than a high value. When focusing on the animal production we use the relative
price of cows and rye in order to estimate if the peasants changed their production mix
according to price changes.
15
Figure 1. Nominal grain prices in Malmöhus County, Scania, 1711–1860
14
12
10
8
rye
6
barley
4
2
oats
1858
1851
1844
1837
1830
1823
1816
1809
1802
1795
1788
1781
1774
1767
1760
1753
1746
1739
1732
1725
1718
1711
0
Sources: Jörberg (1972), Bengtsson and Dribe (1997)
Production and commercialisation in southern Sweden
In order to value the determinants of commercialisation in 18th and 19th century’s agriculture
we will use multivariate analyses on individual household level, both for grain and animal
production. But before then: Let us approach the grain production, from some different
angles, on parish level.
The long term development of grain production output is estimated in figure 2 and table 4.
This is in fact also a total area productivity estimate, since the acreage contained in each
mantal was constant over time, although farms could be divided and the land use could
change over time, generally from pastures to arable. Since all figures are expressed in grain
sheaves per mantal, and all grains are equalized with rye (see data section above), the
estimates has been done by multiplying the parishes’ average number of sheaves with the
threshing average for rye (0.38 hl/sheaf), and then multiplying by 0.68, which was the average
volume/weight ratio for rye. 7 This figure is multiplied by 0.29, which was the average farm
7
This is measured from the 1865–1873 official statistics (BiSOS) for the county of Halland, which is the earliest
point in time when grain volumes are properly reported in weights.
16
size, measured in mantal, and finally we must multiply by 30, to convert the share of the
harvest that was delivered to the parish priest to farm gross production.
The initial rise in grain production, almost by 30 percent until the middle of the 18th century,
must probably be seen as a recovery after the Great Nordic War (1700–1721). After that the
trend series reveal some 25 stagnant years before production started to rise again in 1775.
From the 1790s and onwards we can see a more marked production increase, only shortly
interrupted in the mid 1820s and early 1850s. Total production almost increased by 81 percent
in the 18th century and then doubled 1800–1840.
Figure 2. Total grain production estimates 1711–1860, 1 000 kg per farm
with trend (11-years moving average).
11
10
9
8
7
6
5
4
3
2
1
1855
1849
1843
1837
1831
1825
1819
1813
1807
1801
1795
1789
1783
1777
1771
1765
1759
1753
1747
1741
1735
1729
1723
1717
1711
0
Source: Scanian Database of Agricultural History
The annual short term fluctuations were of course considerable, above all due to temperature
and precipitation. Production volatility actually increased over time, especially after the
radical enclosures in the first half of the 19th century. This confirms the theory of the open
field system as a rational behaviour towards risk (McCloskey 1975). With the scattered strips
of the open fields each peasant possessed land suited for different climates – hot and cold, dry
and wet. Eventually this risk-aversion system was abandoned for individual farming, with all
land in one plot. At the same time as the peasants could increase their production output, they
17
became more vulnerable to annual climate fluctuations. Thus, the substantial rise in
production varied from year to year in pre-industrial commercial farming.
Noticeable is also the high degree of correlation between groups of parishes in yearly
production fluctuations. The correlation coefficient between the two property right groups
presented in figure 4 is 0.94, and correlations between the three soil condition groups in figure
3 are 0.91, 0.90 and 0.83 (the lowest correlation between the most diverging soil-types).
Between individual parishes in the sample, production correlations are typically 0.7 to 0.9.
But on individual farmstead level the correlations, even within the same parishes, are often
weaker. This indicates two things. First, that individual farm output to a great extent was
dependent on the farmer, but when individual farm output is aggregated they tend to correlate
with weather changes. Second, the high degree of correlation between remote parishes
confirms the reliability of the local priest tithes, as a source for estimating production output.
Figure 3. Grain production estimates 1711–1860, 1 000 kg per farm, in parish groups with
soil conditions 3–5, 6–7 and 8–10, respectively
14
13
12
11
fert 8-10
10
fert 6-7
9
8
7
6
5
fert 3-5
4
3
2
1
1855
1849
1843
1837
1831
1825
1819
1813
1807
1801
1795
1789
1783
1777
1771
1765
1759
1753
1747
1741
1735
1729
1723
1717
1711
0
Source: Scanian Database of Agricultural History
In figure 3, the effect of natural conditions is measured using modern soil fertility
classifications. 8 We classify the parishes in three groups, where a higher number in soil
8
For explanation, see above and Bohman 2007.
18
condition represents higher fertility. This investigation reveals no pronounced outcome
differences in the 18th century. From 1800 and onwards, however, it seems like if peasants’
ability to increase production to a great extent were dependent on soil conditions. From the
1730s to the 1830s mean production per decade increased by 77 percent in the parishes with
the worst soil, by 164 percent for the middle group, and by 274 percent on the best soil.
Figure 4. Grain production estimates 1711–1860, 1 000 kg per farm, parishes grouped by
property rights
8
7
6
Freeholders and
Crown's tenants
5
4
3
Nobility's
tenants
2
1
1836
1831
1826
1821
1816
1811
1806
1801
1796
1791
1786
1781
1776
1771
1766
1761
1756
1751
1746
1741
1736
1731
1726
1721
1716
1711
0
Source: Scanian Database of Agricultural History.
Note: All parishes could not be used in this estimation, since some of them had a mix of property
rights. After 1838 there are too few parish observations for comparisons.
In the same way property rights did not seem to have a great influence on production
outcomes in the 18th century. From the results shown in figure 4 we find that there were no
major differences between parishes dominated by freeholders and Crown tenants, and
parishes dominated by the nobility. But from the early 19th century and onwards the rate of
growth for the tenants of the nobility is slower than the one performed by freeholders and
Crown tenants. This pattern has been revealed in earlier research, and is explained by weaker
property rights for tenants under the nobility (Olsson 2005). Especially in the 19th century the
contrast became sharp between these groups. A vast majority of the Crown tenants became
freeholders when they bought their land and the land rents and taxation systems was
19
favourable and stable for them, but not for tenants of the nobility. Furthermore, the latter’s
incentives for investments in land were certainly not promoted by the growing risk of being
evicted, when the landlords increased demesne production in the 19th century. In the six most
pronounced manorial parishes in our sample almost one third of the peasants were evicted in
the period 1825–1860.
Figure 5. Grain production estimates 1711–1860, 1 000 kg per farm, for early surplus
parishes (plains) and non-surplus parishes (forest), compared to average household
subsistence level
28
26
24
22
20
18
16
14
12
10
8
6
4
2
Subsistence level
1858
1851
1844
1837
1830
1823
1816
1809
1802
1795
1788
1781
1774
1767
1760
1753
1746
1739
1732
1725
1718
1711
0
Source: Scanian Database of Agricultural History
Until now we have looked at pure production estimates. 9 Now we must turn to the share of
the output that could reach the market. From the land surveyors’ reports from the first half of
the 18th century we have identified some parishes with surplus production of grain, but also
parishes where the peasants “in normal years” only reached subsistence level, and “could not
sell any grain for their needs of cash” (Land surveyors’ accounts, e.g. Bäringe 1743, L15–
3:2). The dotted line in figure 5 represents this subsistence level, which we define as the mean
production level 1730–1742 for the early non-surplus parishes. The initial subsistence level is
in this estimate 2 623 kg grain per year and average farm household. The 19th century
9
The multivariate analysis with ”pure” grain production as dependent variable is, for comparison, shown in table
7, Appendix 1.
20
subsistence level is altered with increasing household sizes (see data section), and reaches its
maximum in 1845, with 3 833 kg.
In the first half of the 18th century the mean decade share of subsistence in the grain-poorer
parishes varied between 88 and 105 percent (see table 4). Already by that time, the parishes in
the plains could sell about half of their grain production. After 1750 most parishes, at least to
some extent, could produce for the market, but this share was still modest in the wooded
areas. In the 19th century there was a substantial grain surplus. The peasants in the plain-lands
could sell four out of five produced barrels of grain in the 1830s and 1840s, and by 1850 even
the peasants in the now less forested areas could sell every second barrel.
Table 4. Grain production, 1 000 kg per farm, mean per decade,
and household productions’ share of subsistence.
Production
mean, all
parishes
1710
1720
1730
1740
1750
1760
1770
1780
1790
1800
1810
1820
1830
1840
1850
2,4
2,3
2,7
2,8
3,1
3,1
3,2
3,3
4,0
4,3
5,0
6,0
6,5
8,5
7,8
Share of subsistence
Parishes in
the forest
91 %
88 %
105 %
106 %*
108 %
124 %
124 %
122 %
153 %
160 %
169 %
171 %
142 %
156 %
211 %
Parishes in
the plains
193 %*
188 %*
223 %
232 %
246 %
241 %
251 %
271 %
324 %
356 %
411 %
464 %
536 %
554 %
–
Source: Scanian Database of Agricultural History
* = extrapolated
– = missing value, too uncertain to extrapolate
The panel data set is well fitted for general least square regressions with random effects,
which measure variations both within and between the 1 397 individual farmsteads in the
database. In table 5 the regression for commercialisation of grain production is estimated. The
average observation for the farms is 35 years and the maximum is 124 years of observation.
21
The overall r-square value for the regression is 0.41, which means that the independent
variables explain 41 percent of the changes in the dependent variable. However, the value of
explanation is much higher between the individual farms than within them, 0.50 compared to
0.13. This is due to the production volatility discussed above, but it is not a problem since our
aim is not to explain the short term correlations between weather and agricultural output.
The coefficients in table 5 can be valued against the estimates of the early levels of selfsufficiency, which has been estimated to 2.6 metric ton of grain for an average household.
The strongest impact on grain production has the size of the farmstead, measured in mantal.
The average size of the farmsteads, for the whole period, was 0.29 mantal. The coefficient 6.9
can be interpreted as an indication that when the farm size, for example, doubled from 0.29 to
0.58 mantal the average effect on commercialised production was 2.0 (6.9*0.29), in this
example not far from the estimated output for self-sufficiency in grains. This is a reasonable
result, since in the late 17th century a quarter of a mantal was seen as a minimum farm size in
fiscal legislation (Sommarin 1939, p. 35).
On parish level we observed lower production outcomes in the 19th century for tenants on
noble land compared to other peasants. This is confirmed with a significant positive effect on
commercialisation, being a freeholder or a Crown tenant.
The “Man. by:”-variables in the regression table express different situations when the farms
were managed by other cultivators than traditional peasant household. As we have seen in
table 3 these observations are proportionately few in the sample. No significantly different
effect is observed for farms managed by persons of rank, most often officers or state officials,
as compared to farms managed by peasants. The effect of tenancies of peasants’ farmstead, a
way of management taking place when the owners for some reasons could not manage the
farm themselves, was to some extent positive, but not entirely significant. The same goes for
managements directly and only by servants, when no possessor lived on the farm. For these
groups we must also consider that their households normally were smaller, and, thus, the
commercialisation effect was probably even stronger than the figures show.
For the farms where peasant cultivators were evicted and the management was taken over by
the demesne directly, we see a quite strong and significant negative effect on
commercialisation. However, this effect is probably exaggerated since these farms were
22
managed by workers employed by the manor. This implies that no household was present and,
thus, that no consumption took place; all production went to the market in accordance with
earlier findings on demesne production as almost entirely commercialised in the 19th century
(Jonsson 1980, Olsson 2006). There is also a selection problem present due to the fact that
when farms were taken over by the demesne the priest tithes were normally transformed to a
fixed annual amount, the very same year or after only one or two years, and thereby they
disappear from our sample. So, what the result really indicates is that production went down
the first few years after an eviction.
Table 5. GLS regression, impacts on grain commercialisation 1711–1860
(Random-effects, Gaussian)
Mantal
Freeholders & Crown tenants
Man. by: Tenant to peasant
Man. by: Demesne
Man. by: Servants solo
Man. by: Person of rank
Early enclosures 1
Early enclosures 2
Radical enclosures
Initially solitary
Grain price 7 years av.
Grain price change vari.coff.
Soil conditions
Trade deregulation 1775
Trade deregulation 1810
Trade deregulation 1825
Constant
Group variable: farms
Number of observations
Number of groups
R-sq:
within
between
overall
Wald chi2(16)
Prob > chi2
Coef.
6.924797
.1968287
.0973194
-.3334623
.1469585
.0041971
.2378536
.1490427
.9858308
-.3668776
.1864874
-.2522298
.2874590
.1683778
-.0918919
-.0709525
-3.667240
Std. Err.
.2260816
.0754153
.0607566
.0962411
.0894200
.0648999
.0235902
.0412800
.0282670
.1748928
.0063052
.0369390
.0255959
.0183355
.0238099
.0191755
.1766406
P>|z|
0.000
0.009
0.109
0.001
0.100
0.948
0.000
0.000
0.000
0.036
0.000
0.000
0.000
0.000
0.000
0.000
0.000
49 141
1 397
0.1343
0.4991
0.4084
8868.58
0.0000
Some of the strongest effects on grain commercialisation are found in the enclosure variables.
Early redistributions of land in the village communities, most often in the plain-lands in the
23
18th century, could certainly increase production, but an even stronger impact had the radical
enclosures in the 19th century. The interpretation of the coefficient is that enclosing a farm on
average affected grain output with about half of a farm household total consumption. As
expected, the initially solitary farmsteads, most often in the forests, had a lower degree of
commercialised grain production.
Prices had an effect on production and commercialisation, both in terms of levels and
stability. The peasants reacted to price increases with more market oriented production, and to
increases in volatility with withdrawals from the market. The latter can be interpreted as a
response to market uncertainties by exits.
As we have seen, soil conditions had an impact on market possibilities on parish level in the
19th century. This is confirmed in the individual household analysis, with a quite strong and
significant positive impact of better soil conditions. Going back to figure 3, this effect is
above all present after the radical enclosures.
Finally, we see a somewhat contradictory impact of trade deregulation. The first deregulation
in 1775, which meant that trade activities outside the towns were allowed, seems to have had
a positive effect on commercialisation. But the other two, the abolishment of the interior toll
in 1810 and the abolishment of grain export prohibition in 1825, seems to have no, or even a
small negative effect. However, if we use pure production outputs as dependent variable, as in
Appendix 1, the deregulation of 1825 displays a quite strong positive effect. This indicates
that the effect of this institutional change is crowded out by the way the increase in household
sizes is estimated.
In table 6 we estimate the effect of the explanatory variables on animal commercialisation
using the same model as for the grain commercialisation. The number of farms in this sample
is lower than in the grain estimation. This is due to the fact that we omit two large parishes
where calf production is missing. Moreover, for some of the remaining parishes periods exist
where the animal production is somewhat uncertain due to years with missing values or
periods of very uncertain registration. These periods have also been omitted in the analysis.
The average observation for the farms is 30 years and the maximum is 115 years of
observation.
24
Only about 12 percent of the variation is explained by the explanatory variables. This is
mainly due to a very low explanation of the within farm variation. Using a panel data
approach increases variation by 26 per cent, but considering the fact that on most farms on
average one or two animals were born per year and that this generates a lot of years with no
foal or calf born, the low explanatory power of the within variation is not surprising. It is
simply not possible to estimate year to year variation when looking at animal production on a
single farm.
Table 6. GLS regression, impacts on animal commercialisation 1711–1860
(Random-effects, Gaussian)
Mantal
Freeholders & Crown tenants
Man. by: Tenant to peasant
Man. by: Demesne
Man. by: Servants solo
Man. by: Person of rank
Early enclosures 1
Early enclosures 2
Radical enclosures
Initially solitary
Cowprice/ryeprice, trend
Soil conditions
Constant
Group variable: farms
Number of observations
Number of groups
R-sq:
within
between
overall
Wald chi2(12)
Prob > chi2
Coef.
2.154919
.1007716
-.0770641
-.4609550
-.4622129
-.9152256
-.3027005
-.2862783
.1954064
.3508666
-.2271560
-.1604838
2.627477
Std. Err.
.135668
.0461546
.0683607
.125454
.1186251
.100932
.0282314
.0532559
.0282069
.0873806
.0172333
.0147726
.1133281
P>|z|
0.000
0.029
0.260
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
33 639
1 130
0.0125
0.2658
0.1185
855.16
0.0000
From the results it is clear that farm size had an effect on the possibilities of surplus
production of animals, the larger the farm the higher the commercialisation. It is also apparent
that peasants invested more in animal production than persons of rank, or other constellations
connected to higher social groups. Among the peasants, being a freeholder or a Crown tenant
had a positive effect on commercialisation in animals compared to being a tenant under the
nobility.
25
The effects of enclosure of farms are different for the different forms of enclosure. Early
enclosures have a negative effect on animal production, while the more radical enclosures
have a positive effect. The explanation for this pattern is not straightforward; it is possible that
the early enclosures implied investments in arable land pushing the animal production aside.
We know from earlier research that eighteenth century enclosures often implied land
reclamation, converting wastelands into arable land (e.g. Olai 1987). This would decrease the
amount of animals, due to less fodder and smaller grazing areas, closer to the minimum level
of draught animals. For the radical enclosures we find a positive effect on commercialisation.
This would be explained by the new crop rotations introduced together with more arable land
to plough, making it necessary, and possible, to increase the number of draught animals. On
the other hand, this effect would be counteracted by the more efficient ploughs that were
introduced in the first part of the 19th century, reducing the need for draught animals
substantially (Olsson 2005).
The effect found for solitary farms is connected to the fact that these farms were principally
situated in more remote forested areas. The results show that even if we control for soil
quality, where better conditions meant lower commercialisation in animals, there is an
additional effect of being a solitary farm, probably not related to enclosures or the open-field
system but to its geographical location. The effect of the soil conditions confirms previous
findings on specialisation according to natural conditions (Hanssen 1952).
However, together these effects point to the conclusion that the production of animals above
all was connected to the need for beasts on the farm, not in the first place to market demand.
This is even more evident when focusing on the price effect. Controlling for all other
variables we find a negative effect of the relative price of cows and rye. This might seem
contra-intuitive. However, if we look at the price trends for rye and cows we find that the
largest increase in the relative price of cows occurs from the beginning of the 19th century and
onwards. This period is also a period of rising prices on grain, even if the increase in the price
of cows is faster, and, as we have seen, large investments in grain production. Hence, the
negative effect is most probably due to a lack of commercialisation on animals, at least
compared to the one that occurred in grain.
Conclusions
26
The province of Scania is called “the granary of Sweden” and the commercialisation of the
peasant economy in Scania gradually emerged already in the 18th century. During the second
half of this century farmers produced a substantial surplus of grain available for the market.
This was first, and foremost, the case when it comes to farms on the plains. However, the
largest increase in commercialisation occurred during the first half of the 19th century, and
from this time on even farms in less ideal conditions began producing grain for the market.
Looking at production development, the first half of the 18th century consisted of a recovery
from a time of great warfare. The Great Nordic war ended in 1721, but relatively shortly
thereafter grain production levelled out on a subsistence level in the forested areas and a bit
higher on the plains. Farms in the plains produced grain enough to pay taxes and sell to the
market while the former most often paid their taxes in animals or by wood products. Early
enclosures and trade deregulation in the second half of the century opened up for increased
commercialisation but there were no marked differences between farms with different types
of property rights or different soil conditions in the rate of increase in commercialisation
during this period.
All this was to be changed by the radical enclosures. They brought an end to the open-field
system and led to the introduction of new crop rotations and to conversion of meadows into
arable land. The growth in commercialisation was from then on higher on freehold and Crown
land, compared to on noble land managed by tenants. This was due to strengthened property
rights on freehold land combined with rigid taxes. On noble land peasants faced increased
land rents and uncertain rights of possession due to evictions of tenants following expansions
of demesnes. Moreover, the introduction of individual management of land, and the fact that
the land was consolidated in one unit per farmer, had effects as well on total production as on
the volatility in production outputs. The peasant could now exploit the fertility of the land to a
much higher degree, but at the same time they were more exposed to weather conditions.
We also find that peasants responded to prices. Periods of higher prices led to increased
commercialisation and so did periods of stable prices; volatile prices were negatively
correlated with commercialisation.
The commercialisation in grain had no equivalence when it comes to animals. The production
of animals in Scania was above all connected to the needs of the farm and to tax payments.
27
On the plains peasants had to produce a relatively large number of oxen and horses in order to
be able to plough the stiff clay soils, while in the forested areas less draught animals were
needed so part of the production could be used as tax payments. However, we find no respond
to increasing relative prices for cows during the 19th century which indicates that even if the
production increased the market was still fairly limited when it comes to animals.
Conclusively, the southern Swedish case reveals that peasants reacted to prices and trade
opportunities. Commercialisation and increase in grain production was possible within the
open-field system, but with individual management of land those who had the best soil
conditions and who were self-owners increased their production at a much faster rate than
those who had worse conditions or had less independence. The increase in production turned
Sweden from being grain-importing to becoming a large exporter at the same time as the rise
in income for the peasants created a market for industrial goods. The early expansion of the
textile industry during the first half of the 19th century is one distinct mark of this.
28
References
Sources
Enclosure Acts, Lantmäteriet
International Institute of Social History. Prices and wages, digital database published by J. L.
van Zanden, available at www.iisg.nl/hpw/data.html.
Land Surveyors’ accounts, Lantmäteriet.
Poll tax registers, Regional Archives of Lund.
Scanian Database of Agricultural History, Dept of Economic History, Lund University
Scanian Demographic Database, www.ekh.lu.se/ed/databaser/sdd.asp
Tithe rolls, Regional Archives of Lund
Litterature
Allen, R. C. (1992). Enclosure and the Yeoman. Oxford: Clarendon Press.
Allen, R. C. (2000). Economic structure and agricultural productivity in Europe, 1300–1800.
European Review of Economic History 3, 1–25.
Andersson, C. (2007) Tjänstefolkssystemet i Skåne 1822, unpublished essay, Department of
Economic History, Lund University.
Bengtsson, T. and Dribe, M. (1997). Economy and Demography in Western Scania, Sweden,
1650–1900. EurAsian Project on Population and Family History, Working paper series,
no. 10.
BiSOS, Bidrag till Sveriges officiella statistik, del N: Jordbruk och boskapsskötsel.
Hushållningssällskapens berättelser 1866–1873.
Bjørn, C. (1988). Det danske landbrugs historie, 1536–1810. Odense: Landbohistorisk
selskab.
Bohman, M. (2007) A matter of geography? Agricultural specialization in southern Sweden
(Scania) c. 1700–1850. Paper presented at the COST-workshop in Rennes, June 2007.
Brenner, R. (1976). Agrarian class structure and economic development in pre-industrial
Europe. Past and Present, 70, 30–75.
Clark, G. (1991). Labour Productivity in English Agriculture, 1250–1860: Evidence from
Labour Inputs. In Campbell, B. M. S. & M. Overton (eds.) Land, Labour and Livestock:
Historical Studies in European Agricultural Productivity. Manchester: Manchester
University Press, 211–235.
29
Clark, G. (2006) The Long March of History: Farm Wages, Population and Economic
Growth, England 1209–1869. Paper at the XIV International Economic History
Congress, Helsinki, 2006.
Dahlman, C. J. (1980). The open field system and beyond: a property rights analysis of an
economic institution. Cambridge: Cambridge University Press.
Dickler, R. A.(1975). Organization and Change in Productivity in Eastern Prussia. In Parker,
W. N. & E. L. Jones (eds.) European Peasants and Their Markets. Essays in Agrarian
Economic History. Princeton: Princeton University Press.
Dribe, M. (2000). Leaving Home in a Peasant Society. Economic Fluctuations, Household
Dynamics and Youth Migration in Southern Sweden, 1829–1866. Lund: Almqvist &
Wiksell International.
Ellis, F. (1988). Peasant Economics: Farm Households and Agrarian Development.
Cambridge: Cambridge University Press.
Fridlizius, G. (1957). Swedish corn export in the free trade era: patterns in the oats trade
1850–1880. Lund: Gleerups förlag.
Gadd, C-J. (1983). Järn och potatis. Jordbruk, teknik och social omvandling i Skaraborgs län
1750–1860. Göteborg: Ekonomisk-historiska institutionen.
Gadd, C-J. (2000). Den agrara revolutionen, 1700–1870. Det svenska jordbrukets historia,
del 3: Natur och Kultur/LT:s förlag.
Grantham, G. (1975) Scale and Organization in French Farming, 1840–1880. In Parker, W. N.
& E. L. Jones (eds.) European Peasants and Their Markets. Essays in Agrarian
Economic History. Princeton: Princeton University Press.
Hanssen, B. (1952). Österlen. Allmoge, köpstafolk & kultursammanhang vid slutet av 1700talet i sydöstra Skåne . Ystad: Nordiska Museet.
Helmfrid, B. (1949). Tiondelängderna som källa till ett byalags ekonomiska historia 1555–
1753. Stockholm.
Hoffman, P. T. (1996). Growth in a Traditional Society. The French Countryside 1450–1815.
New Jersey: Princeton University Press.
Jonsson, U. (1980). Jordmagnater, landbönder och torpare i sydöstra Södermanland 1800–
1880. Stockholm: Almqvist & Wiksell International.
Jörberg, L. (1972). A History of Prices in Sweden 1732–1914. Lund: Gleerups förlag.
Leijonhufvud, L. (2001). Grain tithes and manorial yields in early modern Sweden : trends
and patterns of production and productivity c. 1540–1680. Uppsala: SLU.
30
Le Roy Ladurie, E. and Goy, J. (1982) Tithe and Agrarian History from the Fourteenth to the
Nineteenth Centuries. An essay in comparative history. Cambridge: Cambridge
University Press.
Lundh, C. (1999). The social mobility of servants in rural Sweden, 1740–1894. Continuity
and Change 14, 57–89.
Magnusson, L. (1997) Sveriges ekonomiska historia. Stockholm: Rabén Prisma.
McCloskey, D. (1975). The Persistence of English Common Fields. In Parker, W. N. & E. L.
Jones (eds.) European Peasants and Their Markets. Essays in Agrarian Economic
History. Princeton: Princeton University Press.
Myrdal, G. (1933). Wages, cost of living and national income in Sweden 1860–1930. Vol. 1.
Stockholm Economic Studies. London: P. S. King.
Olai, B. (1987). “…till vinnande af ett redigt Storskifte…” En komparativ studie av storskiftet
i fem härader. Uppsala: Almqvist & Wiksell International.
Olsson, M. (2002). Storgodsdrift. Godsekonomi och arbetsorganisation i Skåne från dansk tid
till mitten av 1800-talet, Lund: Almqvist & Wiksell International.
Olsson, M. (2005). Skatta dig lycklig. Jordränta och jordbruk i Skåne 1660–1900, Hedemora:
Gidlunds förlag.
Olsson, M. (2006) Manorial Economy and Corvée Labour in southern Sweden 1650–1850,
Economic History Review, LIX:3, 481–497.
Overton, M. (1996). Agricultural revolution in England. The transformation of the agrarian
economy 1500–1850. Cambridge: Cambridge University Press.
Overton, M. and B. M. S. Campbell (1996). Production and productivity in English
agriculture 1086–1871. Histoire et Mesure, XI-3/4, 255–297.
Palm, L. Andersson (1997). Gud bevare utsädet!: produktionen på en västsvensk
ensädesgård. Djäknebol i Hallands skogsbygd 1760–1865. Stockholm: KSLA.
Persson, K. G. (1999). Grain Markets in Europe 1500–1900. Integration and Deregulation.
Cambridge: Cambridge University Press.
Popkin, S. L. (1980). The Rational Peasant: The Political Economy of Peasant Society.
Theory and Society, 9:3, 411–471.
Schön, L. (1995). Historiska nationalräkenskaper för Sverige: Jordbruk med binäringar
1800–1980. Lund: Ekonomisk-historiska föreningen.
Schön, L. (1997). Internal and external factors in Swedish industrialization. Scandinavian
Economic History Review, 45, 209–223.
31
Schön, L. (2000). En modern svensk ekonomisk historia. Tillväxt och omvandling under två
sekel. Stockholm: SNS förlag.
Scott, J. C. (1976). The moral economy of the peasant: rebellion and subsistence in southeast
Asia. New Haven: Yale University Press.
Scott, T. (1998). Introduction. Scott, T. (ed.), The Peasantries of Europe: from the Fourteenth
to the Eighteenth Centuries. London/New York: Longman.
Slicher van Bath, B. (1963). The agrarian history of western Europe A. D. 500–1850,
London: Edward Arnold.
Sommarin, E. (1939). Det skånska jordbrukets ekonomiska utveckling 1801–1914. Del 2.
Lund: Skrifter utgivna av de skånska hushållningssällskapen.
Svensson, P. (2001). Agrara entreprenörer. Böndernas roll i omvandlingen av jordbruket i
Skåne 1800–1870, Lund: Almqvist & Wiksell International.
Svensson, P. (2006) Peasants and Entrepreneurship in the Nineteenth-Century Agricultural
Transformation of Sweden. Social Science History 30:3, 387–429.
Turner, M. E., Becket, J. V. and B. Afton (2001). Farm Production in England 1700–1914.
Oxford: Oxford University Press.
Weibull, J. (1952). Tionden i Skåne under senare delen av 1600-talet. Lund: Gleerup.
Wohlin, N. (1909). Den jordbruksidkande befolkningen i Sverige 1751–1900: statistiskdemografisk studie på grundval af de svenska yrkesräkningarna. Stockholm: Norstedt.
Wolf, E. R (1966). Peasants. Englewood Cliffs. NJ: Prentice Hall.
de Vries, J. (1975). Peasant Demand Patterns and Economic Development: Friesland, 1550–
1750. In Parker, W. N. & E. L. Jones (eds.) European Peasants and Their Markets.
Essays in Agrarian Economic History. Princeton: Princeton University Press.
de Vries, J. (2001). The transition to capitalism in a land without feudalism. In
Hoppenbrouwers, P. and J. L. van Zanden (eds.) Peasants into farmers? The
transformation of rural economy and society in the Low Countries (middle ages – 19th
century) in light of the Brenner debate. CORN Publication Series 4. Turnhoet: Brepols
Publishers.
Zanden, J. L. van (1999). Wages and the standard of living in Europe, 1500–1800. European
Review of Economic History, 2, 175–197.
32
Appendix 1
Table 7. GLS regression, impacts on grain production 1711–1860
(Random-effects, Gaussian)
Mantal
Freeholders & crown tenants
Man. by: Tenant to peasant
Man. by: Demesne
Man. by: Servants solo
Man. by: Person of rank
Early enclosures 1
Early enclosures 2
Radical enclosures
Initially solitary
Grain price 7 years av.
Grain price change vari.coff.
Soil conditions
Trade deregulation 1775
Trade deregulation 1810
Trade deregulation 1825
Constant
Group variable: farms
Number of observations
Number of groups
R-sq:
within
between
overall
Wald chi2(16)
Prob > chi2
Coef.
6.926682
.1799259
.1009317
-.3148599
.1220462
.009073
.2198753
.1262532
1.011946
-.3833656
.242400
-.4728325
.2863213
.0831925
-.0869927
.3712715
-1.840845
Std. Err.
.2261312
.0754443
.0608523
.0963993
.0895649
.0650052
.0236279
.0413472
.0283123
.1748623
.0063156
.0370008
.0255923
.0183661
.0238497
.0192073
.1766212
P>|z|
0.000
0.017
0.097
0.001
0.173
0.889
0.000
0.002
0.000
0.028
0.000
0.000
0.000
0.000
0.000
0.000
0.000
49 141
1 397
0.2028
0.5191
0.4333
13718.93
0.0000
If we compare this table’s estimates on impacts on production with table 5: s impacts on
commercialisation, we find that the overall explanatory value of this regression is higher,
mainly because of impacts on variations within the farms. This is due internalization of
changes in household sizes into the dependent variable in the commercialisation approach.
This also explains the diverging outcomes of the 1825 trade regulation dummy. However, the
general effects of the different variables are much the same in the two regressions.
33