Conference PaperPDF Available

The Swedish Sonderweg in Question: Democratization and Inequality in Comparative Perspective, c. 1750–1920


Abstract and Figures

PLEASE NOTE! THIS PAPER WAS PUBLISHED OPEN ACCESS IN PAST & PRESENT, AUGUST 2019. READ HERE: THE CONFERENCE VERSION FROM 2017 IS NOT RELEVANT ANYMORE. In the twentieth century, Sweden became known as a country with an unusually egalitarian distribution of income and wealth, an encompassing welfare state, and an exceptionally strong Social Democracy. It is commonplace among historians and social scientists to consider these equal outcomes of the twentieth century as the logical end result of a much longer historical trajectory of egalitarianism, from early modern free peasant farmers or from a peculiar Swedish political culture, egalitarian and consensus-oriented. This essay questions this Swedish Sonderweg interpretation. It is shown that in 1900, Sweden had together with Prussia the most unequal voting laws in Western Europe, and more severe economic inequality than the United States. This puts the purported continuity from early modern equality to Social Democratic equality in question. It is instead argued that the roots of twentieth-century Swedish egalitarianism are much more modern than often argued, and lie in exceptionally well organized popular movements after 1870, with a strong egalitarian counter-hegemonic culture and unusually broad popular participation in politics.
Content may be subject to copyright.
The Swedish Sonderweg in Question:
Democratization and Inequality in Comparative Perspective, c. 1750–1920”
Erik Bengtsson
Department of Economic History, Lund University &
Economic History Unit, University of Gothenburg
Version 1.1. For presentation at University of Southern Denmark, 23 May 2017
Preliminary! All comments and criticisms welcome.
13 977 words, two tables, one figure
In the twentieth century, Sweden became known as a country with an unusually egalitarian
distribution of income and wealth, an encompassing welfare state, and an exceptionally strong
Social Democracy. It is commonplace among historians and social scientists to consider these
equal outcomes of the twentieth century as the logical end result of a much longer historical
trajectory of egalitarianism, from early modern free peasant farmers or from a peculiar Swedish
political culture, egalitarian and consensus-oriented. This essay questions this Swedish
Sonderweg interpretation. It is shown that in 1900, Sweden had together with Prussia the most
unequal voting laws in Western Europe, and more severe economic inequality than the United
States. This puts the purported continuity from early modern equality to Social Democratic
equality in question. It is instead argued that the roots of twentieth-century Swedish
egalitarianism are much more modern than often argued, and lie in exceptionally well organized
popular movements after 1870, with a strong egalitarian counter-hegemonic culture and
unusually broad popular participation in politics.
Keywords: inequality, Sweden, modernization, Sonderweg, Barrington Moore
Work on this paper has been financially supported by a Swedish Research Council grant for the project ”Growth
and inequality before the industrial revolution, Scania 1650 to 1850”, and the STINT grant “Poles apart: a long-
term perspective on inequality, industrialization and labour market institutions in Brazil and Sweden”. I am grateful
to Per Andersson, Anders Hylmö, Johannes Lindvall, Thomas Paster, Carolina Uppenberg and Erik Vestin for
comments. All errors and misunderstandings are my own.
I. A Swedish Sonderweg?
In history as well as in the social sciences, it is tempting to interpret modern-era outcomes, often
on the national level, in the spheres of politics and economics as the logical outcome of long-
term trends. One can think of Macfarlane’s (1978) thesis of a continuity of individualism in
English society since the thirteenth century, the idea of American Exceptionalism, or the
Sonderweg (special route) thesis in German history which claims that Germany fell prey to
Fascism because of its lack of an eighteenth or nineteenth century bourgeois revolution, which
left the country off the “normal” path of liberal modernization and instead put it on a Sonderweg
dominated by a conservative nobles and militaries (Wehler 1973; see Blackbourn and Eley 1984
for a critical discussion).
Sweden, in stark contrast to Germany, in the twentieth century became known for
its “middle way” between capitalism and socialism, as a country with high degrees of economic
equality, an encompassing welfare state, and exceptionally strong Social Democracy. It has
become commonplace to see Sweden’s twentieth-century egalitarianism as the outcome of a
long historical trajectory, a Swedish Sonderweg. In different formulations, the Swedish
Sonderweg thesis stresses the heritage of a free and politically active farmer class, economic
egalitarianism, and a responsive state. Within Sweden, this idea has reached “the status of a
national myth” (Hilson 2009, p. 138), as it has been embraced not only by professional
historians but also by political parties (Linderborg 2000) and a wider interested community.
The latest example is a successful and influential book of popular history, which present
equality as the “fate” of the Swedish people (Berggren and Trägårdh 2006, p. 44). The Swedish
Sonderweg analysis in its various guises also has gained currency in the international research
literature (Tilton 1974, Castles 1973, Delong 2007), and is standard in textbooks (e.g. Möller
2007 on Swedish political history).
However, this essay will show that the factual basis of the Swedish Sonderweg
interpretation actually is quite weak. For this reason, the interpretation must be reconsidered.
To investigate the Sonderweg interpretation, the essay discusses Swedish political and social
development from the mid-eighteenth century to the 1920s, the period when Sweden
modernizes and transforms from an agrarian society undoubtedly ruled by the monarchy and
nobility, to an industrial economy with political democracy. It is shown that the Sonderweg
thesis misconstrues what kind of road Sweden took to modernity.
In fact, Sweden c. 1850
1920 was severely unequal. In 1900, it was one of the countries in Western Europe with the
most restricted suffrage, and wealth was more unequally distributed than in the United States.
Thus if there was a long continuity of egalitarianism, it was certainly not unbroken as Gilded
Age Sweden was as unequal as any other capitalist country in that time. The more likely
explanation of Swedish twentieth-century equality, rather than any deep roots, is the in an
international perspective extraordinary degree of popular organization in the labour movement
and other popular movements in the period after 1870. The thick web of popular organization
then created a progressive counter-hegemony which facilitated rapid democratization and a
transition from a society marked by severe economic and political inequality in the 1900s to an
egalitarian society with a generous welfare state after the 1930s.
The contribution of the essay is two-fold. Firstly, it provides a critical discussion
of the image of Sweden widespread in history and social sciences as well as economics as
a country with an unbroken, long heritage of equality. Many of the references used by
international researchers are quite dated and here newer findings are synthesized. Secondly, the
essay contributes with an in-depth case study to the today very lively debate on the determinants
of long-run inequality (e.g. Piketty 2014, Milanovic 2016).
II. The Swedish Sonderweg view
There are three variants of Swedish Sonderweg analyses. The one following Moore (1966) is
materialist, focused on class relations, and stresses the role of the free peasantry. Another
version equally focuses on the peasantry but in a more culturalist way. The third focuses on
political culture.
I am here influenced by Eley’s (Blackbourn and Eley 1984, p. 154) argument on the understanding of the German
Sonderweg, on the need to redirect some of the focus from the longue duree to more short-term causes of the
Fascist disaster.
The Barrington Moore model of “routes to the modern world”
In historical sociologist Barrington Moore’s (1966) model of “routes to the modern world”,
whether a country ends up on a bourgeois-democratic, authoritarian-fascist, or communist
route, depends on the constellation of class forces in the country during modernization.
According to Moore, British and French modernization was led by bourgeois liberalism, while
in Prussia-Germany the dominance of the old landowning nobility (the Junkers) led to an
authoritarian route, and in Russia and China the absence of capitalist modernization in
agriculture led to a peasant revolt cum Communist revolution.
For Moore, the bourgeoisie is
strongly associated with democracy and liberalism.
In the 1970s, two Anglophone researchers tested Moore’s model on the Swedish
case. Castles (1973) found that when it came to commercialisation of agriculture and
replacement of the nobility, Sweden did not fit the model. The bourgeoisie was comparatively
weak in Sweden, and in fact subservient allies, after 1809, to the nobility, in precisely the kind
of coalition (“iron and rye”) which in Moore’s model would lead to authoritarianism. Castles
pointed to the existence of a strong peasant farmer class as the factor which saved Sweden from
the authoritarian route: “in a sense, the farmers held the line until industrialism produced a
liberal middle class capable of asserting its own rights” (Castles 1973, p. 327; cf. 330). Timothy
Tilton (1974) found better fit with Moore’s model. He claimed that Sweden unlike France,
Germany or Norman England never was fully feudal. With reference to an English-language
history of Sweden from 1931, he claims that the nobility only held 10 per cent of the land,
which is both quite unhistorical of course this share varied over time and quite misleading.
In fact, the nobility, which was around 0.5 per cent of the population, held about 20-25 per cent
of arable land from the late mediaeval period to 1600, 65 per cent in 1658, and one third in 1700
(Myrdal 1999, p. 334; Gadd 2002, p. 17).
Just like Castles, found that the peasantry was
extraordinarily important in Sweden, and politically played the liberal role played by the
bourgeoisie in Moore’s model: “The size and strength of the independent Swedish peasantry
Østerud (1978) presents an application and development of Moore’s model on the Scandinavian countries. For
criticism of Moore’s model in the German case see Eley (in Blackbourn and Eley 1984, especially pp. 5292).
Similarily, the otherwise excellent Rueschemeyer et al (1994, p. 93) rely on modernization theorists’ descriptions
from the 1950s which leads to an underestimation of the nobles.
can hardly be overemphasized, for in Sweden the peasants often played the role that the
bourgeoisie played elsewhere as an agency for preserving the balance between the monarchy
and the nobility.” (Tilton 1974, p. 565.) The analysis where a strong independent peasantry
plays the role of a guarantee of democracy in the lack of a strong urban bourgeoisie, is
represented in a different theoretical context also by for example Perry Anderson. Anderson
(1974, pp. 179180) pointed out that Sweden in the early modern period lacked strong cities
and therefore also a bourgeoisie. In his analysis, what prevented Sweden from taking the East-
Elbian road of development through full on feudal exploitation, was the strength of the
independent peasantry.
A culturalist peasant-oriented approach
There is also a much less materialist interpretation of Swedish history, which still sees a long
historical trajectory of egalitarianism. In Sorensen and Stråth’s (1998) interpretation, the
Scandinavian countries took a peasant-dominated route to modernity, which was marked by an
egalitarian society and an egalitarian culture. Sorensen and Stråth (1998, p. 1) claim that there
was a specific “Nordic Enlightenment” which “had the peasant as its foremost symbol”, “seen
as the mythical incarnation of education (bildning/dannelse), freedom, and equality.” Sweden
had peasant farmer representation in parliament since the fifteenth century, and this is symbolic
of a “Nordic Sonderweg” where the Social Democracy of the twentieth century is seen as a
continuation of a Lutheran peasant farmer culture where social liberalism and social democracy
melt together.
Trägårdh (1998) agrees: unlike in Continental Europe where democratization
was extending noble rights to everyone, in Scandinavia, the construction of modern democracy
can be seen as a process of generalizing the political culture of the local peasant assembly.”
(Trägårdh 1998, p. 258.) Here, the lack of feudalism and the political representation for farmers
In comparative political economy, the independent peasantry likewise, but later in time, puts Scandinavia on its
special route in for example the analysis of Katzenstein (1985, p. 35). Also pp. 140143 and 157159. Katzenstein,
who cites Tilton (1974) on the allegedly weak nobility in Sweden, also claims that “The absence of a strong feudal
tradition is equally striking in Scandinavia.” (p. 159).
Stråth (1988) in a study of democratization in Scandinavia explicitly advocates a Swedish Sonderweg analysis.
Åmark’s critical comments attached to Stråth’s chapter in many ways accord with the argument of this essay.
meant that farmers played a larger role, and twentieth century Social Democrats were the
inheritors of farmer populism (Trägårdh 1998, p. 259).
The political culture variant of the Sonderweg thesis
The historian Eva Österberg (1989, Österberg et al 2002) has launched an interpretation where
the “Swedish model” of consensus-seeking, compromise-friendly, rationalist policy making,
most often associated with the period after the Social Democratic ascension to government after
1932, actually starts in the sixteenth century. Österberg and her followers (Aronsson 1992,
Cederholm 2007) claim that the Swedish state from the early modern period on had
exceptionally strong ties to the local level of society, with state representatives present more or
less everywhere. Österberg stresses the relative dearth of peasant uprisings after the 1530s and
farmers’ political representation in parliament to state that there was a degree of consensus
between the rulers and the ruled. The great extent of peasant farmer involvement in local politics
and as lay representatives in local courts is stressed by Österberg as these involvements meant
regular contacts between the state and the ruled. A keyword in her argument is “interaction”,
and she claims that the frequent interactions between rulers and the ruled created a consensus-
oriented political culture in Sweden.
It is not easy to determine the degree to which Österberg’s description of Swedish
political culture is correct. Her interpretation has caught criticism for downplaying power
inequalities between the rulers and the ruled (Harnesk 2002, Linde 2009, Hilson 2009). Critics
have also claimed that the lack of peasant uprisings after the 1530s depended more on the
considerable repressive capacities of the standing armies developed in this period, than on the
consensus between rulers and the ruled. I believe that the critics have their points, but also want
to stress vis-à-vis Österberg (1) the existence of proletarian and semi-proletarian groups who
were excluded from the political arenas stressed by Österberg, and (2) the massive economic
and political inequalities that culminated in the late nineteenth century, and which mean that
there was no easy continuity from the peasant farmer egalitarianism claimed by Österberg and
followers for the early modern period, to the Social Democratic egalitarianism of the twentieth
century. I will get back to these points.
III. The Swedish ancien regime, c. 17501850
The Swedish Sonderweg view in all its guises stresses the role of the free farmers. And it is true
that when Prussia abolished serfdom in 1810 or Russia in 1870, this had since long happened
in Sweden, where serfdom gradually disappeared between 1000 and 1300 CE (Myrdal 1999,
pp. 9397). However, this does not mean that the Swedish economy and polity were egalitarian
and inclusive. The social structure and ownership of land in the eighteenth and nineteenth
centuries was surely more equal compared to Britain, where peasants were dispossessed early
and the agrarian economy famously dominated by noble and gentry estate owners (e.g. Lindert
1986), and more equal to the Russian serf economy. But at the same time, Swedish society was
more Prussian”, and generally more alike Continental Europe, than is assumed in the
Sonderweg analysis. I will here present two main objections to the peasant-centred Sonderweg
thesis. One, that while the peasant farmers were indeed acknowledged as political actors, they
were very much a subordinate actor in a game dominated, as on the Continent, by the monarchy,
nobles, and capital-owners. Two, the peasant-centred story misses the fact that the peasants
weren’t the lowest strata of society. The inclusion of this lower strata of labourers, and a correct
appreciation of the role of the nobility, in the analysis changes the tenor of the story.
Farmers and other people
In 1750 half of the Swedish population were free peasant farmers. However, this figure includes
the tenants of the nobility, so we must not assume that all of them owned land. It was only the
freeholders and the tenants on crown land who were represented in parliament; the nobility’s
tenants had no voting rights. Above the peasants in the social structure, there was a small slice
of nobles (0.5 per cent of the population), burghers, merchants, and persons of standing,
together about 6 per cent of the population. Below, crofters, rural artisans, soldiers, and workers
were about 43 per cent of the population. (See Bengtsson et al 2017a, Appendix). This was a
more equal social structure than that of in various ways Britain or Russia, but not equal.
From 1750 to 1850 landless rural groups grew so fast that in 1850 only one fourth of rural
families owned or controlled land (Wohlin 1909). That three-fourths of the rural population
were proletarian or semi-proletarian is not considered in the Sonderweg view, where all
emphasis is on the strength of the independent peasantry.
The English author Mary Wollstonecraft was on her trip to Sweden in 1796 shocked of
how servants were treated. She wrote:
“In fact, the situation of the servants in every respect, particularly that of the
women, shows how far the Swedes are from having a just conception of rational
equality. They are not termed slaves; yet a man may strike a man with impunity
because he pays him wages, though these wages are so low that necessity must
teach them to pilfer, whilst servility renders them false and boorish.”
(Wollstonecraft 1796, Letter III)
Traditionally the image has been that to be a servant in agrarian Sweden was a life-cycle
occupation; i.e., that children of farmers were servants in their youth, until they became old
enough to inherit the farm. Then, around the age of 30, they transitioned from being a servant
to being a self-owning farmer. However, as the population grew after 1750 and a smaller share
of the new generations could inherit or afford to buy a farm, the servant occupation became less
of a life cycle phenomenon and ever more people became life-long workers (Lundh 1999;
Uppenberg 2017).
Labour in Sweden was until 1885 regulated under a regime which stipulated that
if one could not live off one’s property, and did not have a job, one could be sentenced to forced
labour (cf. Johnsson 2016). Sweden was the last country in the Nordics to abolish this
regulation: Denmark and Norway did it in the 1850s, and Finland after 1863 (Johnsson 2016,
p. 14). That this “essentially feudal coercive measure” (i grunden feodalt tvångsmedel,
Johnsson 2016, p. 16) was in place until 1885 tells us something important about the nature of
Swedish society in this period. It was a society run by the aristocratic and bourgeois elite in the
name of authority and profit-maximization. In 1885 the old labour regime was replaced by
regulation of “drifters” and “vagrants”, which still allowed for the punishment of for example
strikers who were temporarily without a job.
Farmers and politics, c. 17501850
As discussed in section II, Castles, Tilton, Stråth, Anderson, Österberg and others have all
presented the peasant farmers and their political representation as the guarantor of a democratic
and egalitarian route to modernity in Sweden. For the farmers to play this function, they must
have (1) stood for an inclusive politics, and (2) exerted influence. Based on the Swedish-
language political history literature, I will now discuss these two points, beginning with the
The Swedish-language political history literature portrays the farmers very much
as a subordinate partner. There was no peasant farmer in the government until 1906, when the
progressive liberal Peterson in Påboda became the minister for education. Sweden did not have
a parliamentary system, where the government reflects the parliament’s majority, until 1920.
Until then, the government was the king’s council and his advisors, and picked by him. (On the
long struggle for parliamentary rule see for example Lindblad 2015. More generally see for
example Stråth 2012, pp. 155159.) Governments were dominated by estate-owners, militaries
and bureaucrats, with added capitalists and merchants after 1850. Among the four estates in the
four-estate diet which was in place until 1865, the farmers’ estate was the only one which never
got the right to elect its own speaker; this was done by the King. Furthermore, it was the one
estate which was not represented on the Privy Committee, the most important parliamentary
committee. Among the four estates, the nobility and the clergy played a conservative role. The
farmers were more oppositional. Since decisions were made by a majority of the estates, the
burghers had an important role in the middle (Tjerneld 1983).
So, the peasant farmer politicians were subordinate. We can still ask if they
ideologically played the role of a democratic-liberal avant-garde? (As in Castles 1973,
Anderson 1974, Tilton 1974, Alestalo et al 1987.) Many Swedish political historians,
unconscious of the great role played by the farmers in the international comparative literature,
have portrayed the peasant farmer politicians as narrow-minded, conservative, reactive, loyal
to the King, and focused on picturesque issues such as the right to home production of licquor
(e.g. Nilsson 1973, Bäck 1984, Nilsson 1994). This image is so common that one of the more
recent studies of farmer politicians asks in the title: “Primitive farmers?” (Claréus 2002. See
also the historiographic comments and references in Scherp 2013, pp. 1925.) Of course, there
are also a few historians who have viewed peasant farmers and their political representatives as
proactive agents in Swedish 19th century politics. Alexandersson (1975, pp. 7076) in his
dissertation about the peasant estate in parliament 17601772 revises the common image of the
peasant parliamentarians as unengaged and lazy. Likewise, Hultqvist (1954, 1955) shows both
that given that the lion part of public expenditures were funded with land taxes, and noble land
was exempt from tax, in reality the peasant farmers were the most heavily taxed group, and that
the bureaucracy was privileged and dominated by nobles. For these reasons, the intense
criticism from peasant politicians against state bureaucracy and public spending c. 18501880
must not necessarily be seen as the result of short-sighted penny pinching, but can rather be
seen as founded in real inequalities and injustice (cf. Christensen 1997). There are also some
instances in historical research which does not concern peasant politics in particular, but where
peasant politicians emerge as proactive and egalitarian forces. So Åkerman (1967, pp. 1011)
in his dissertation on the experiment with progressive income taxation in 1810 points to pressure
from below, especially from the peasants, as a cause of the reform. And in Christensen’s (1997)
dissertation on the role of the peasant estate in the 1840s debate on suffrage reform, the peasant
politicians indeed appear as conservative in issues of finance and banking, but not as void of
ideology and not as one-sided opponents of any public investments, as previous researchers
have claimed.
In conclusion, the farmer politicians at least to some degree were a democratic
voice. Christensen (1997) in the best study of this issue shows that in the nineteenth century
before the 1865 reform, the ancien regime was defended by a conservative group of bureaucrats
and estate owners, while the liberal-democratic opposition was led by anti-bureaucratic liberal
newspapermen, in an interesting alliance with the progressive wing of the farmers in parliament.
But Christensen also shows how the peasant farmers to a large degree focused on defending the
interest of their own class vis-à-vis their exploitation by estate owners and the state, and to a
much lower degree on the interests of the propertyless groups, in for example to access to
schooling (Christensen 1997, pp. 278281, 293294).
Not the downfall of the nobility
One of the most influential analyses of the alleged downfall of the Swedish nobility is Sten
Carlsson’s (1973). Carlsson pointed to especially two reforms as indicative. One, the opening
up of higher posts in the state for commoners by a reform in 1789. Two, the opening up that
commoners could buy tax-exempt land which until 1809 was reserved for nobles. On both
points, Carlsson’s interpretation coloured as it was by the Swedish belief in an egalitarian
Swedish exceptionalism has been shown to be flawed. Norrby (2005, p. 309) in his revisionist
dissertation on the Swedish nobility in the nineteenth century shows that the nobility continued
to dominate the higher echelons of the bureaucracy and military also after the 1789 reform, by
virtue of their superior education and social and cultural capital (also Sundell 2014 on
nepotism). On land-owning, Winberg (1985, pp. 164182, 200) showed that contrary to
Carlsson’s assumptions, the nobility in parliament in no way was united against the 1809
reform. On the contrary, a large share of them realized that the creation of a freer, more capitalist
market for tax-exempt land in no way forced them to sell, and would indeed increase the price
of the tax-exempt land. This increase in the value of the land would help the nobles to
restructure their investments when they so wanted, and research by Norrby (2005, pp. 7475)
and Ulväng (2013) shows that their ownership of other types of land indeed increased more
after 1809 than their ownership of tax-exempt land decreased. More recent research by Clark
(2014, ch. 2) shows that still today, Swedish nobles are severely over-represented in high-
prestige professions law and medicine, and are significantly wealthier than non-nobles. The
idea that the nobility lost out during an allegedly “bourgeois” nineteenth century once fit the
whig history of Marxists and liberals, but does not square with the facts of Swedish history (cf.
Bengtsson et al 2017b).
IV. The persistence of the Swedish old regime, 18501920
Between 1850 and 1920, Sweden industrialized, urbanized and moved, on a wobbly road, in
the 1910s to universal suffrage and parliamentary rule. This section shows that throughout this
period, the old regime (cf. Mayer 1981) persisted all the way into the First World War. Counter
to what could be assumed from the Sonderweg view, the development during this period was
not consensus-based, egalitarian, or smooth.
Economic inequality
At some point, when the empirical findings of every new study contradicts the old consensus,
the consensus must be scrapped. We are probably now at the point where the idea of continuous
egalitarianism in Scandinavia is due to be scrapped. Measuring economic inequality before
1900, the tax and census data that we associate with the modern state, is not easy, but in the last
fifteen years there has been a steady stream of top incomes studies, which can measure the top
of the distribution back to 1870 or so. There has been one study of Denmark (since 1870), one
of Norway (since 1875) and one of Sweden (since 1870). The researchers have in every case
found that contrary to what they expected and contrary to the common image of Scandinavia,
the country under investigation was not particularly equal around 1870 or before the First World
War. So write Roine and Waldenström (2008, p. 385): “Sweden was not more equal than other
Western countries at the beginning of the twentieth century; Aaberge et al. (2016) and
Atkinson and Søgaard (2016) reach the corresponding conclusion for Norway and Denmark,
respectively. Income inequality fell only after 1930, with the growth of redistributive taxation
a falling capital share of national income as well as union wage policies.
Recently, with co-authors I have shown that wealth inequality the most relevant
measure of economic inequality in pre-industrial society grew gradually in Sweden from 1750
to 1900. Inequality grew in many different ways. One powerful mechanism was the growth of
the rural proletariat, as discussed above: when in 1850 three fourths instead of one half of
families in the countryside were proletarian, this implied a bottom-led growth of inequality. On
the other side, inequality also grew from the top (or almost-top) as the non-noble but wealthy
class of merchants and capitalists grew in numbers as well as increased their wealth. The end
product is wealth distribution in 1850 and 1900 which is not especially equal in a comparative
perspective (Figure 1). In 1850, the ten percent wealthiest people held 81 per cent of the private
At the late twentieth century, Sweden was famous as a high-tax country. This, equally, is not a deep historical
legacy. Taxes as a share of GDP were only 7 per cent in 1900 and 10.3 per cent in 1930; after 1930 taxation grew
rapidly, to 20.3 per cent in 1950, and 37.0 per cent of GDP in 1970 (Henrekson and Stenkula 2015, Figure 1.2).
wealth in Sweden, as compared to 82 per cent in France, 84 per cent in the UK, and 70 per cent
in the US (in 1870). In other words, Swedish wealth distribution was about as unequal as French
and British, and significantly more unequal than American. In 1910, the picture is essentially
the same, with the top decile’s share at 88 per cent in Sweden, 89 per cent in France, 92 per
cent in Britain, and 81 per cent in the US.
Figure 1. Top decile’s share of total wealth: Sweden in an international perspective 1750–
Note. Sources: Lindert (1986, Table 4) for Britain 1740-1870. Shammas (1993, p. 424) for the US 1774 and 1870.
Bengtsson et al (2017a) for Sweden 17501900. Piketty (2014) for all other data.
Neither in terms of income nor wealth was Sweden especially equal in 1900. The Gilded Age
was quite gilded for the Swedish élites as for their American counterparts.
1740 1770 1800 1830 1860 1890 1920 1950 1980 2010
France Britain USA Sweden
Swedish Plutocracy in a Prussian Mirror
Sweden in the 18501920 period was not only a highly unequal economy, but also politically
highly unequal. A key indicator of how inclusive society is, is the share of the population who
were enfranchised. Table 1 shows this for parliamentary elections in the 1870s and around 1900,
for Sweden and seven other Western European countries.
Table 1. Average total franchise (of over 20 years of age)
26 %
29 %
5 %
18 %
38 %
9 %
55 %
10 %
15 %
9 %
26 %
19 %
42 %
36 %
38 %
4 %
35 %
3 %
24 %
38 %
Source: Aidt et al (2006), Table 2. See also Karlsson Schaffer (2010).
As we see in the table, Sweden after 1880 had a strikingly low share of enfranchised adult
citizens. Among the eleven Western European countries compared here, Sweden with 15 per
cent of adults has the lowest share enfranchised: much lower than relatively democratic Norway
(55 per cent), but also much lower than France, Austria, and even Germany (around 40 per
While the Swedish political system was highly retrograde, the central Statistics bureau was more up-to-date. In
1898 Statistics Sweden published a comparison of election systems in Sweden and fifteen other European
countries. The degree of enfranchisement, in elections to the second chamber of parliament, among men over 21
years in Sweden was 24 per cent in the comparison year (1896), which meant that Sweden had the most restricted
suffrage among the 16 countries, with the exception of Hungary with 20 per cent. Countries such as Great Britain
(65 per cent), Norway (77), the Netherlands (45), Belgium (79), France (84) and Greece (84) stood out as much
more inclusive. Statistiska Centralbyrån (1898), Table 20, p 140.
The election system on the local level, which also elected the electors to the first
chamber of the parliament, was even more unequal. A minimum level of income or wealth was
necessary for the right to vote, and among those with the right to vote, the number of votes was
distributed according to a rank of income and wealth. In urban municipalities, an individual
(which could also be a company) could control up to a hundred votes, and in rural municipalities
there was no limit at all. Famously, the prime minister of the 1880s count Arvid Posse, because
of the wealth that he had inherited in the shape of a family estate, controlled a majority of the
votes in the municipality Bergkvara (Mellquist 1974, p. 218). As Mellquist (1974, p. 9) put it
in his dissertation on the Swedish municipality voting system 18621900, all countries in
Europe restricted the franchise of the poor, but no system was as extreme as the Swedish one.
Mary Hilson (2009, p. 143) notes that Swedish politics in 1900 “appears as an example of
reaction with its unreformed Riksdag and a level of enfranchisement that was among the lowest
in Europe”.
Class character of the state
From the Swedish Sonderweg-story, we would expect Sweden to have a “prescient” welfare
state already in the late 1800s or early 1900s. This is not necessarily the case. As Lindert (1994)
shows, there was nothing special about the amounts, as a share of GDP, of the Swedish state’s
welfare expenditures (poor relief, health, etc.) between 1880 and 1920. Indeed, in 1913 the
share of state expenditures going to the military was especially high in Sweden, which is
especially notable for the fact that this was a country which had not been at war since 1814. 42
per cent of state spending went to the military, which may be compared to 31 per cent in
Norway, 16 in Denmark, 27 in the Netherlands and 25 in Belgium (Lindblad 2015, p. 69). The
description of Sweden as a “militarized alms house(det befästa fattighuset) from a Socialist
pamphlet of 1913 is not that far off the mark.
The speed at which the Swedish nobility lost political power in Sweden has been
significantly overstated. In Liberal as in Marxist historiography, the downfall of the nobility
post-1789 has been a convenient assumption, but historical research in the 1980s and 1990s
have knocked this assumption down all over Europe. As shown in section II, there is a
counterpart in Sweden, in that the nobles here were much wealthier than has been assumed.
Over the period 1844 to 1905, 56 per cent of government ministers were nobles (while about
0.5 per cent of the population were nobles), and the first non-noble prime minister only entered
the post in 1883 (Norrby 2011, pp. 246247). In this as in many other dimensions, Sweden was
more “Prussian” than the Swedish Sonderweg theory assumes. In 20 of the 45 years between
1875 and 1920, Sweden had a noble prime minister; the exceptions were from clerical, military,
bureaucratic and factory-owning backgrounds. Sundell (2014) shows that until the early
twentieth century the Swedish bureaucracy was marked by a significant concentration of
surnames: top bureaucrats were recruited from a small group of families (especially nobles).
A comparison with the other Scandinavian countries can be enlightening.
1876 when the office of prime minister was created to 1905, all Swedish prime ministers were
landowners, bureaucrats, militaries or industrialists. Karl Staaff, a lawyer, was in 1905 the first
pm from a “free profession”. Of the 17 prime ministers from 1876 to 1921, 7 were nobles. In
Norway on the other hand, none of the 14 prime ministers from 1873 to 1921 were noble, and
nine of them were lawyers, often also professors. Of the five non-lawyers, two were school
teachers, one an engineer, one a farmer, and only one was a career military officer (a general).
In other words, Norwegian recruitment to the highest public office was much more
“democratic” than that in Sweden. In general, Norway is compared to Sweden the more pure
example of the idea of the Nordic Sonderweg; as Sejersted (1992) has pointed out, Swedish
development from 1850 to 1920 was marked by large-scale capitalism and high-profile
capitalists such as the Wallenberg family; Norway did not have any Wallenbergs.
We may
also compare with Finland where, just as in Sweden, the forest and its produce was at the heart
of industrialization. However, in Finland unlike in Sweden individual farmers owned most of
the forest which meant that the large income streams from exports of timber and later paper and
A more or less populist critique of the privileged, wasteful and lazy bureaucrats was standard fare for the farmer
politicians during the last decades of the nineteenth century (see for example Hultqvist 1954, pp. 61, 64).
Information about the occupations of prime ministers in Sweden and Denmark, and their fathers, has been
fetched from Wikipedia: and . Corresponding information for Norway is from the
Norwegian centre for Research Data’s Archive of Politicians since 1814: .
The comparative political economy literature points out that Sweden in the twentieth century was a country not
only with a strong trade union movement, but also with strongly organized employers (e.g. Pontusson and Swenson
1996). This is in line with the relative strength of “big capital” in the early twentieth century.
pulp in Finland to a high degree ended up in the hands of the farmers (Alapuro 1988), while in
Sweden the incomes where much more concentrated to the companies. For this reason, while
Swedish economic inequality boomed from 1850 to World War One, in Finland it went in the
opposite direction. In terms of governments, Denmark was somewhere in-between the more
exclusive Swedish pattern and the more inclusive Norwegian one. Among its 21 prime ministers
(konseljpresidenter and statsministrar) from 1855 to 1920, four were noble. These four and
three others were estate owners, while eight were lawyers, one was a school teacher, three were
newspaper men. Only one was a military man and he was mostly a mathematician and teacher
at the military college. If we look at the social backgrounds of the prime ministers, it was also
more exclusive in Sweden.
There, until Staaff in 1905 all prime ministers had fathers who
were high-rank militaries, factory owners, estate owners, and in one case a bishop. There was
nothing like the farmer fathers of Norwegian prime ministers Sverdrup (188489), Løvland
(190708) and Konow (191012) or Danish Christensen (190508), and also nothing like the
carpenter father of CC Hall in Denmark (185759, 186063), the mason master father of JH
Deuntzer (190105), or the shoemaker father of CT Zahle (190910, 191320). The most
democratic recruitment to Swedish prime ministers before the mid-1920s was the vicar and
university teacher father of Staaff and the schoolmaster father of Nils Edén, a liberal and himself
a professor of history (191720).
Farmer politics, 18661920
In section IV I have discussed the peasant farmers’ politics in the four estates Diet up till 1865.
What happened after 1866, in the two-chamber parliament? Rather typical of the current
literature is Nilsson (1994, esp. pp. 22, 67) who in his dissertation on the first chamber of
parliament 18671886 claims that the economic elites in the first chamber were the driving
forces in the economic modernization of Sweden, with railroads and other important
investments, while the peasant-dominated second chamber was a conservative force. (See also
Kilander 1991 pp. 123ff; Nilsson 2001).
Åmark (2005, pp. 2930, 3435) in his comparative study of welfare policy in Norway and Sweden also stresses
that in the late nineteenth century Norway was a more egalitarian economy, with a less polarized politics.
Steadily from 1865 into the twentieth century, a healthy share of the 200-odd MPs
in the second chamber were farmers. There was no real party system then; MPs were rather
elected for their individual attributes than for their opinions. (The tariffs election of 1888 is
often credited as the break-through of the party system.) However, after 1865 most of the second
chamber MPs organized in some loose party organization. The foundation was the government-
friendly Ministerial Party organized in 1866 to support the liberal De Geer government. Against
this, the most important competitor was the Farmers’ Party (Lantmannapartiet), which also
organized already in 1866, with some tradition back to the four estates diet. Most the the second
chamber farmers belonged to this party, which was informally led by two noble estate owners,
Arvid Posse and Emil Key, and one farmer, Carl Ifvarsson. That estate owners and farmers
could unite has been debated in the literature, with a classical question being if Posse, who was
a well-known conservative and opponent of the 1865 representation reform, organized the party
to fool the farmers into a harmless identity politics (i.e. Bokholm 1998, pp. 144147).
The second chamber farmers seem to have prioritized as their number one issue
the lowering of taxes and the lightening of their costs for supplying soldiers to the army; both
taxation and the cost of soldiers lay disproportionately on the farmer class (e.g. Thermaenius
1928, Hultqvist 1954). The most democratic proposals on extensions of suffrage came from
urban liberals in alliance with farmers (Mellquist 1974). The moneyed bourgeoisie in the first
and second chambers certainly were contra Moore no proponents of democracy; so AO
Wallenberg in 1881 dismissed suggestions of suffrage extension as the egoistic ambitions of
some (disenfranchised) groups to increase their own influence at the expense of other
(enfranchised) groups (Mellquist 1974, p. 173). The sharpest opposition to the voting rights of
incorporated companies came from farmers from the north of Sweden, where mining and timber
companies often could dominate the local political scene (Mellquist 1974, p. 192). The more
democratically aligned farmers and urban liberals organized in the so-called New Liberal Party
18681871, and then on the left wing of the Farmers’ Party or the Centre Party.
On balance, contra the general tenor of the Swedish political history literature
(Nilsson 1973, Nilsson 1994), the farmer politicians after 1866 should not necessarily be seen
as parochial penny-pinchers. While they displayed strong conservative tendencies on a few
issues, such as finance (they preferred to have the central bank extend its remit rather than allow
private profit-making banks), they were, together with a small group of urban radical liberals,
the most democratically aligned group in parliament on issues such as suffrage reform, taxation
and schooling (Hultqvist 1954, Mellquist 1974, Christensen 1998). But contrary to the
Sonderweg perspective of the comparativists, the farmers did not rule the roost in Sweden in
this period. They were very much a subordinate group to the nobility-bourgeois fusion elite
which together with the royal house ruled the country.
Was there a continuity between farmers and the labour movement?
In analyses of the Swedish Sonderweg it is common to assume that the political representation
of the farmers meant that the state was more lenient on the nascent labour movement than more
exclusively aristocratic-bourgeois states were. However, this is questionable. As discussed
above, the bureaucracy in Sweden was dominated by the elite, a melange of bourgeoisie and
the nobility, and so were the governments at least until after 1905. Furthermore, the labour
regime was repressive. There is no evidence that the political presence of farmers increased
understanding towards the labour movement. This is unlike Finland, where such evidence does
exist (Stenius 1980; Alapuro 1988, pp. 101102, 105107). The Swedish state certainly used
less violence against the union movement than did its US counterpart (Archer 2008), but it was
not particularly friendly towards it either, as witnessed for example by the use of vagrancy laws
to sentence the 1879 Sundsvall strikers to forced labour, or the 1906 Mackmyra conflict where
the prime minister himself, Christian Lundberg, who was the CEO of the afflicted company
fired all workers who had joined a union and, in the middle of the winter, evicted them and their
families from their company housing (Casparsson 1966, pp. 8892). The state also took action
against the unions for example with the 1899 law (Åkarpslagen) which instated that it was
illegal to try to stop strike-breakers; in fact, the law even made it illegal for striking workers to
speak to the strike-breakers (Casparsson 1966, p. 149). Again, Finland, like Norway, appears
as the more “Nordic” country: more egalitarian, less conflict-ridden.
V. The counter-hegemony and the transition, 18701950
In Sweden, as in other states, the ancient regime lasted until the First World War (Mayer 1981).
The exceptional equality of Swedish economy and society c. 19201990 did not arrive as the
logical conclusion of a long historical continuity; pace Berggren and Trägårdh (2006), equality
is not the “fate” of the Swedes. The logical question is then: what caused the twentieth century
As discussed in section IV, economic inequality was only reduced after 1920.
Two mechanisms are stressed in the empirical literature: redistributive taxation, and the fall in
capital incomes (cf. Roine and Waldenström 2008). The fall in capital incomes after 1920 is
related to an increase in regulation after the First World War, as stressed by Piketty (2014), as
well as the strengthening of unions after 1920 (Bengtsson 2014). The emergence of
redistributive taxation and the build-up of the welfare state depended, as stressed by the power
resources school, upon the unique strength of the Swedish labour movement (Esping-Andersen
1985, Korpi 1982). We must then understand why the unions and the labour movement were
so strong in Sweden.
Madeline Hurd’s (2000) comparison of German and Swedish political-economic
development c. 18701930 provides a key part of the puzzle. Against the traditional
interpretation of Germany as authoritarian and Swedish as structurally democratic and
egalitarian, she points out that suffrage was as limited in Sweden as in Prussia before World
War One. She turns this fact in a very interesting direction, by pointing out that the late
democratization of Sweden actually fostered a liberal-socialist democratizing alliance which
put the country on a democratic and egalitarian rather than authoritarian route. Since suffrage
was so limited, petit bourgeoisie liberals and working-class socialists could unite strategically
for democracy, unlike Germany, where the greater inclusion of lower middle class men meant
that middle class liberals and haute bourgeoisie market liberals could unite around a program
of economic liberalism without democracy. Hurd’s study is a very important corrective to the
continuity assumption of the Sonderweg model: as she shows, in fact, the very lack of
democracy in Sweden could through its effect on social coalitions foster a thorough
As Hurd summarizes it: the plutocratic nature of the voting system itself shaped
the “small-folkalliance of petit-bourgeois and middle class liberals and Social Democratic
workers, which was to carry Sweden to democratization (Hurd 2000, p. 270). My interpretation
of Sweden is close to Hurd’s, but what I will do now is to present some facts about the key role
of the labour movement specifically, not only for democratization but also for the ensuing
equalization of incomes and wealth. Where my interpretation differs somewhat from Hurd’s is
in our estimation of the role of working class culture in Sweden. To Hurd (2000, p. 218), “Late
urbanization, combined with the rapidity and nonurban forms of much of Sweden's
industrialization, had resulted in a weak working-class associational culture.” However, in my
view she lends to much weight to beer-drinking as an indicator of working class culture.
German working class social life was more centred around beer than Swedish was, but this
difference does not mean that there was less of working class cohesiveness in Sweden just
that it was different. (On the centrality of the tavern in the German labour movement, see Lidtke
1985, pp. 20, 55; Roberts 1982.) Hurd points to the Socialist-Liberal alliance in Stockholm in
the 1890s and early 1900s as much stronger than in Germany and it is true that there was some
cultural convergence among working class leaders, to a petit-bourgeois pattern. But Hurd’s
(2000, p. 235) conclusion on this issue lifts an interesting question: “This approach toward
bourgeois culture lost the new worker some cultural freedom, but it gained him social power.”
Swedish working class associational culture might have been less distinctively proletarian and
less beer-soaked, but it was very strong in the numbers it mobilized and the working-class
consciousness that it bred.
In comparisons with Britain, it has been pointed out that Sweden saw less Lib-
Labism than Britain did (e.g. Fulcher 1991, pp. 6166). Britain industrialized before Socialism
had developed as an ideology, and thus got a strong Liberal current in its working class, with
many workers voting for the Liberals until the party system realignment in the 1920s. This has
Without using that terminology, Hurd’s analysis can be seen as what historical institutionalists and historical
sociologists would call a sequencing argument. Cf. Thelen (1999, p. 389), Mahoney (2000).
been seen by observers such as Perry Anderson as a cause of a weaker Socialist labour
movement in Britain than in the late industrializer Germany. However, in the light of Hurd’s
powerful investigation, we might want to modify this conclusion. It is true that the Swedish
labour movement after c. 1870 was more cohesively Socialist than its British counterpart. But
in Swedish civil society under the politically very exclusive oligarchy which reigned, Liberal
reformers and Social Democrats consorted in a manner more akin to Britain than to Germany.
The strength of the labour movement
Table 2 shows some key indicators of labour movement organization and working class
cohesion. Unlike the inequality figures in Figure 1 and the suffrage statistics in Table 1, this is
a field where Sweden at the end of the ancien regime actually stands out.
Cf. Berman (1998, pp. 381382), who for example cites the Social Democratic party leader Branting on the need
of a “big tent”.
Table 2. Working-class organization, c. 1910
party vote share,
last election
before 1914
Number of male
workers per
Union density
Party members
per 1000
United Kingdom
Note. Sources: voting and newspaper data from Luebbert (1991) pp. 167, 188. Union density from Crouch (1993),
pp. 114122. Peak pre-1914 party membership from Eley (2002), table 4.2, p. 66. Divided by country’s population
in 1900 from
The Swedish Social Democrats, formed as an explicit working-class party in 1889 and shaped
early on ideologically by the German Marxist tradition, were a very strong electoral force
already before World War One. With 36 per cent of the votes in 1911, even with restricted
suffrage (see Table 1), it was the most successful among the nine countries in the table. The
Swedish party along with its Nordic colleagues also was the strongest in terms of party
membership as a share of the population. The German SPD is famous as a mass party but with
its 19 party members per 1000 inhabitants in 1910, it was outdistanced by the Swedish SAP
with its 26 members per 1000 inhabitants. Likewise, union organization was extensive, even
though the Swedish figure in Table 2 is depressed by the loss in the general strike in 1909. From
the low point of 13 per cent in 1910, union density among workers increased to 31 per cent in
1920; before the loss in the general strike, it had been 26 per cent in 1908 (Kjellberg 1983). A
more exotic metric but very important to the argument that I want to make, is the sales of
working class newspapers per capita. Again, the Nordic countries are exceptional, in that the
labour movement-affiliated newspapers had incredible market shares. In Denmark, Norway and
Sweden for every 5 to 9 male workers, one labour newspaper was sold. This means that
Scandinavians to a much higher degree than other Europeans got their news from labour
movement intellectuals.
Furthermore, as Eley (2002, p. 44) points out, party newspapers were
often consumed collectively, available in cafés, clubs and bars. This is indicative of something
much larger, which is the strength of the working-class subculture in Scandinavia. Historical
research on this topic has to some degree focused on more purely cultural expressions such as
singing, drinking and so on (i.e. Lidtke 1985) or on the degree of rowdiness versus orderliness
among workers (e.g. Ambjörnsson 1988, Magnusson 1988). My point is something different.
The Swedish working class at the beginning of the twentieth century was poor and
disenfranchised and lived and worked in a deeply unequal economy, but they were
exceptionally organized. As many students of twentieth century Sweden have pointed out, a
Swedish citizen in this epoch could easily live his or her life entirely within the confines of the
labour movement subculture: live in the working-class housing cooperative HSB, shop at the
cooperative Konsum, enjoy oneself on the weekends at Socialist People’s Parks, go on vacation
within the labour movement travel company, get insurance from the cooperative Folksam, and
get buried by the labour movement-affiliated Fonus.
This socio-cultural complex has political
economy implications.
In the final quarter of the nineteenth century, the popular movements could
mobilize large sections of Swedish society. The temperance movement collected in 1909 1.9
million signatures for prohibition (Lundkvist 1974, pp. 4445), which is quite impressive in a
country with 5.5 million inhabitants. While the suffrage movement were small in terms of
members, at times it could mobilize greatly. So in 1893 for the “people’s parliament”, the
Stephens (1979, pp. 407408) compares the frequency of reading labour movement-sympathizing newspapers
in the British and the Swedish working class in the 1970s, and finds that Swedes indeed were much more likely to
read a Socialist-friendly newspaper. This led according to Stephens to stronger class consciousness in Sweden.
As Ölmedal (2016) points out, the labour movement and its associated cooperatives included at the apex of its
power major providers of housing, retail shopping, a savings bank, gas stations, lotteries, major newspapers, and
so on.
alternative, unofficial parliament organized by the movement, 149 856 people voted for the
different Liberal and Social Democratic candidates. This meant that the unofficial parliament
had 50 per cent more voters than the actual parliament in the elections to the second chamber
that year (Vallinder 1962, p. 90). Further, the suffrage movement’s petition for general suffrage
in 1897 attracted 363 638 signatures, of which only 65 879 had the right to vote to parliament
(Vallinder 1962, p. 202).
The labour movement counter-hegemony
The traditional measure of labour movement strength (not the least in the power resources
tradition) is electoral success and the degree of organization in trade unions. But as Jansson
(2012) points out, organization numbers are not everything. The cohesiveness of the
organization is almost equally important: to achieve something, the party or the union needs to
be able to agree on goals. Jansson, who studies the activities of the Workers’ Educational
Association, stresses the importance of workers’ education for the creation of a cohesive
reformist working class consciousness in Sweden after the Russian Revolution and until the
late 1930s.
My argument here is related, but more general. The width and depth of working-
class organization in Sweden in 1910 (or in 1950, for that matter) was unparalleled. The trade
unions as stressed by Korpi (1982) were important, the Social Democratic Party’s successful
alliance with the Farmers’ Party as stressed by Esping-Andersen (1985) was important, and the
workers’ education as stressed by Jansson (2012) was important. But the cumulative effect of
it all was important in a way that I believe has not been argued in the previous literature. The
total outcome of the exceptional organizational strength of the Swedish labour movement was
a working class with an exceptional degree of organization as well as ideological cohesion.
Lidtke (1985, p. 17) talks in his study of the German labour movement subculture of an
environment of clubs, activities, and relationships permeated with socialist political
Barrling Hermansson (2004) in her study of the organizational cultures in political parties shows that the SAP
is a quite hierarchical party, with a strong emphasis on cohesion and consensus (sometimes imposed from above).
assumptions and implications, and this applied even to labor movement associations that
officially denied all political purpose”. This kind of labour environment was even stronger in
Sweden than in Germany. As Lundkvist (1977) pointed out, around 1900, one third of Swedes
were members of the unions, the teetotalling movement and/or the free churches. Under the
very exclusive and political system in Sweden before 1909, the teetotallers and even the free
churches provided an organizational infrastructure which strengthened the nascent labour
movement. When the labour movement needed organization skills and meeting places, they
could cooperate with the two popular movements that were already there. The teetoalling
movement which spread in the mid-nineteenth century attracted members especially from
society’s lower strata and not the least among women. Their members were to a very large
extent excluded from official politics and general suffrage thus became a central demand for
the movement (Lundkvist 1974, pp. 3645). The movement worked as a “citizen school” for
lower class people who learnt political organizing and activism in the sober context (Lundkvist
1977, pp. 203212). Cross-organizing with the labour movement became common; in the
1910s, 84 per cent of Social Democratic MPs were teetotallers, and among the new Left
Socialist Party’s MPs in 1917, all were (Lundkvist 1977, pp. 175–177).
As Rueschemeyer et al. (1994, p. 50) argue, “it is the growth of a counter-
hegemony of subordinate classes and especially the working class developed and sustained
by the organization and growth of trade unions, working-class parties and similar groups that
is critical for the promotion of democracy.” They point to Scandinavia as an example of that
class organization and creation of a counter-hegemony can “change the balance of class power
in society” (Rueschemeyer et al., p. 274). I believe that this is very true for Sweden and the
egalitarianism of the twentieth century was rooted in this counter-hegemony rather than in a
long historical continuity.
Class consciousness
There are no survey studies of class consciousness in the first half of the twentieth century. The
Swedish National Election Survey, run by political scientists at the University of Gothenburg,
started in 1954. However, Särlvik (1970, p. 254) points out in his study of Swedish voters in
the 1950s that “The social bases of Swedish parties form a pattern that has hardly undergone
any profound change in regard to its main features since the 1920s.” The Swedish party political
landscape was organized in two blocs: on the Left the giant Social Democratic Party and the
small Communist Party, on the centre-right the Farmers’ Party, the Liberals and the
Conservatives. Särlvik points out that while voters of the three centre-right parties had quite
weak party affiliations and flowed between the three parties without too much afterthought, this
was not the case on the Left. The Social Democrats in the 1950s and 1960s had a remarkably
high share of voters who were “convinced adherents” of the party or “strongly identified” with
it (the survey question on this topic was changed in 1968). (pp. 256262) In 1968, when the
Social Democrats won a parliamentary majority, 51 per cent of their voters “strongly identified”
with the party, and another 25 per cent “weakly identified”; 18 per cent said that they only had
a preference for the party, without identification. Older voters were especially likely to strongly
identify with the party. In the Election Survey of 1968, among the respondents born before
1900, 51 per cent responded that they “strongly” identified with one political party. Only 10
per cent reported no party preference, and if one counts those with “weak” identification with
a party, then 75 per cent of voters identified. Strong party identification then gradually declined:
44 per cent among those born between 1900 and 1920, 33 per cent 192040, 23 per cent 1940
60, and 14 per cent among those born after 1960.
Class voting is important in Sweden and
was even more important in the first Election Survey, 1956, than it has been later. There is no
comparative information on class consciousness at the mid-twentieth century, but we may note
that in terms of party members’ share of the electorate, Sweden was from the 1950s (when the
statistics begin) to the 1980s was along with Austria the well-organized country. About 23-24
per cent of the electorate was organized in a political party (Scarrow 2002, Table 5.2).
Own calculations from SNES data.
The earliest survey of class consciousness in Sweden was made in the industrial
city Katrineholm in 1948. This showed that 80 per cent of manual workers and 50 per cent of
white collar workers considered themselves to be working class. Furthermore, a majority of the
inhabitants, when asked which class was in power in Swedish society, replied the working class.
When the study was replicated in post-industrial Katrineholm in the 1980s, the results were
different: fewer identified as working class, and no one believed that the working class was in
power. The 1990 public investigation into power in Swedish society concluded that: The
identification with a movement in the process of conquering power has been replaced with the
feeling of having left power to an elite.” (Maktutredningen 1990, p. 36). Günther Roth (1960,
p. 199) in his study of the German labour movement speculates that vulgar Marxism had a
special appeal among passive members of the movement. The workers’ self-biographies that
he cite show how rank-and-file members could believe in the power of Social Democracy and
Bebel as a coming power, which did not depend on their own activism. In the Swedish context,
we may remark that the party leader of the 1930s and 1940s Per Albin Hansson himself believed
in this. His biographer Isaksson (2000, pp. 48, 73, 106) portrays Hansson leaned over the social
groups tables produced by Statistics Sweden and which showed the rapid growth of social group
III, roughly the working class and the strongest basis of Social Democratic support, with a full
belief that this demography was destiny: population growth itself guaranteed a future Social
Democratic majority.
The immense organizational drive which marked Sweden from the 1880s to the
1920s changed the political ecology of the country for a long time. When the political
sociologist Rydgren (2002, p. 45) at the turn of the millennium wanted to explain why Sweden
then still had no successful extreme right party, he invoked the factor “enduring class loyalties”.
While the working class often can be recruited to extreme right parties as has to some degree
happened in Sweden since Rydgren wrote his article , in 2002 one could still point to the high
degree of class voting and union membership as an obstacle to far-right expansion within the
working class. Around 1990, as the public Power Investigation noted, the belief that the
working class was the majority of the country and ruled the country had eroded. But in
hindsight, what is remarkable is not that the working-class counter-hegemony decayed in the
1980s and 1990s, but that it lasted so long.
Class consciousness and ideology
Since Herbert Tingsten’s (1941) influential study, there has been an on-going if low-intensity
research debate on the ideological development of Swedish Social Democracy. Tingsten’s
argument was that SAP quickly deserted Marxism and its determinism, for a reformist social
liberalism which Tingsten claimed, in line with the Swedish Sonderweg analysis, much more
in accord with Swedish tradition. Lewin (1967) in a study of the planned economy objected
and instead claimed that the SAP in the first half of the twentieth century left Marxist
determinism for a more reformist version of Marxism. Ever since, there has been a debate on
the ideological composition of the SAP and whether it was a cause of the party’s success (e.g.
Berman 1998, Karlsson 2001). My take on this debate would be that while ideas are interesting
in themselves, the combination of Kautskyan Marxism, English guild socialism, liberal
reformism, and Bernsteinian revisionism in SAP was in fact not so special in a European
comparison, and not the cause of the party’s strength. The party’s unique strength was instead
the extraordinary degree of organization, based in the last third of the nineteenth century and
the peculiar timing of organization and democratization in this country. That the SAP had an
expansionist jobs policy in the 1930s may for that matter be related to the similar policy pursued
in working-class dominated municipalities already in the 1920s, where the policy was reached
at because of the material interest of the unemployed party and union members, rather than
from a reading of economic theory (Svensson 2004).
Working class consciousness was a lively topic for historical research during the
1970s. It was then often informed by a Leninist view where the explanandum, the thing to
explain, was: why weren’t the Swedes revolutionary? (E.g. Hentilä 1979.) From the teleological
viewpoint of Leninism, Swedish workers were assumed to have failed to live up to the Leninist
truth, and were misled by their reformist leaders. However, by now it seems apparent that this
was not the right question to ask. Today, what stands out when looking at working class
consciousness during the first half of the twentieth century is not the lack of revolutionary
sentiment, but the strength of class consciousness and the belief in Social Democratic progress.
The road to the alliance between farmers and workers
In political economy analyses such as Korpi (1983), Esping-Andersen (1985) and Luebbert
(1991), the alliance between the Social Democrats and the Farmers’ Party in the 1930s is seen
as key to the rise of Social Democratic hegemony. In relation to the focus on farmers and
workers in this essay, it is worth revisiting the famous “cow trade” of 1932 that started the 44-
year period of uninterrupted Social Democratic government.
In Denmark, the Social Democrats governed from 1929 with the small radical
liberal party Radikale Venstre. This type of alliance was in the mind of the Swedish SAP’s
leader Per Albin Hansson as well, as he reflected over the possibilities of an alliance with CG
Ekman, the Liberal Party leader (Isaksson 2000, pp. 128, 202). In 1932, when SAP won the
Swedish election, the Danish colleagues had shifted to an alliance with the farmer-dominated
Venstre party. The SAP sought a coalition partner but the leadership was worried about the
possible attractions to German fascism within the Farmers Party; here it should be remembered
that the Farmers’ Party was formed during the First World War after a coup against the
government staged by the royal house and military establishment with farmers as foot soldiers
(Lindblad 2015). The farmers’ party (Bondeförbundet) in other words had no roots in the for its
time relatively democratic farmers’ party of the nineteenth century (Lantmannapartiet). The
Social Democratic finance minister Sköld and his wife were given the task to invite top farmers’
party politicians for dinner at the Skölds’ house, and discreetly gather information on their
possible Hitlerist/Lappo sympathies. Sköld and the other top SAP representatives such as
Wigforss concluded in their reports to Per Albin that while the farmers’ party was very pro-
eugenics, they did not sympathize with Nazism. The SAP could then negotiate with the farmers
as well as the Conservatives and the Liberals about the formation of a coalition government.
The Conservatives and the Liberals declined, and the SAP ended up with a complete and within
the party rather controversial surrender to the farmers on agricultural policy, to get a functioning
coalition (Isaksson 2000, pp. 239245).
Previous research (Korpi 1983, Esping-Andersen 1985, Luebbert 1991) is
completely right in emphasizing the 1932 coalition between the farmers’ party and the Social
Democrats as a massively important factor for the unique influence of the Social Democrats in
Sweden. But it is imperative not to interpret this coalition as a logical or unavoidable outcome
of a long continuity of friendly farmer-worker relations. On the contrary, when in 1914 50 000
farmers marched the streets of Stockholm in support of the King’s power and against
parliamentary democracy, the labour movement countered with a 100 000 strong march the
next day, with the opposite message (Lindblad 2015, pp. 1315, 1822). The farmers’ party
was not the first choice coalition partner for the SAP in 1932; the liberals were.
The take-over of the state
The Social Democratic labour movement was well-organized. It also steered the state in an
efficient way. When the Social Democrats governed the country after 1932, they knew that the
bureaucracy was staffed by people who did not sympathize with their policy aims. They
understood, stresses Rothstein (1986) in his book on the “Social Democratic state”, that they
had to people the bureaucracy with new cadres, sympathetic to the new policies, to be able to
implement them effectively. Rothstein compared the success of Social Democrats in steering
actual policy in two fields: labour market policy and education policy. He distinguishes between
the Weberian bureaucracy and a cadre model of bureaucracy, which focuses on value-
rationality rather than instrumental rationality. The labour market board was full of Social
Democratic cadres and steered with Social Democratic goals in mind, while the schooling board
was less committed to equality. Rothstein claimed that this difference led to a more successful
from a Social Democratic point of view policy on the labour market policy area.
An important precondition of why the Social Democrats could re-focus the
bureaucracy was that the educational wing of the labour movement was so strong, as stressed
above. Another probable cause of this successful cadre bureaucracy project was that Sweden
even in its reactionary guise in the late nineteenth century had relatively high levels of literacy
and numeracy. This is one field where we can indeed hint a historical Swedish Sonderweg:
considering how backward the economy was in 1800 or 1750, literacy was remarkably
widespread (Sandberg 1979). The legacy of popular education probably facilitated the growth
of popular movements and their competence after 1870, and thus indirectly influenced the
egalitarian turn of Sweden in the twentieth century. Furthermore, the relative closeness between
intellectual liberals and the labour movement in the early twentieth century, under the very
exclusive political regime which was in place then, probably gave a lasting legacy of easier
recruitment of left-liberal intellectuals to steer the Social Democratic state. The strength of the
Social Democratic cadres meant that they could build an advanced welfare state with ambitious
redistributive aims. Thus, together with the exceptional strength of the trade unions, the
foundations for a new equality were put in place after the 1920s. (Cf. Östberg 1996, pp. 255
257, on the importance of the cadre for political power.) Along with literacy levels high for the
level of economic development of the country, another historical legacy in Sweden was that of
an efficient state. That the Swedish state was relatively competent and powerful internally
already in the sixteenth and seventeenth centuries was argued by Tilly, and has found support
in later historical research (Hallenberg 2001). In Heclo’s (1974) state-centred story, this is key
to the growth of the ambitious Swedish welfare state in the twentieth century. However, as
Premfors (2003) points out, the competent bureaucrats of the old regime did not want
democratization on the contrary, this was enforced on them. Nevertheless, the competence of
the state facilitated ambitious Social Democratic policies as well, once democracy was in place
in 1921.
VI. Concluding reflections
There was no simple continuity from an alleged peasant farmer egalitarianism to a Social
Democratic one. On the contrary, from circa 1920 to 1980, Sweden experienced a rapid
transition, which built on the popular mobilization of the previous decades. This contradicts the
Swedish Sonderweg thesis in all its versions the peasant-focused Moorian approach, the
cultural version, and the political culture version. The roots of modern Swedish egalitarianism
were the huge organizational drive of the “popular movements” in the decades after 1870. This
counter-hegemony achieved through union and political activity equalization of votes, incomes,
and wealth. In this way, the essay supports the institutional approach to understand the
determinants of economic inequality (cf. Milanovic 2016). Politics is, as Hacker and Pierson
(2012) say, organized combat and the Swedish working class was the well-organized in the
world; this meant that Sweden became the most Social Democratic country in the world.
Another reason to doubt the Swedish Sonderweg thesis, is the rapid increase in
economic inequality in Sweden since the 1980s. If equality is a long-lasting national trait, a
factor built into an eternal political culture, indeed the “fate” of the Swedish people (as Berggren
and Trägårdh 2006 claim), then equality should not be threatened just because the Social
Democrats and the labour movement more generally have been weakened since the 1980s. The
Gini coefficient of incomes according to OECD has increased from 21.2 in 1975 and 19.8 in
1981, the lowest point, to 28.1 in 2014, the latest available year. With this, Sweden is now the
eleventh most equal country, beaten by all the Nordic countries but also for example Slovenia,
Belgium and Austria.
Likewise, the share of total income accruing to the ten per cent with the
highest incomes has grown from 22 in the early 1980s to 28 in 2015 according to the World
Wealth and Income Database. Economic and political equality in twentieth century Sweden
was a not an eternal, unmoveable national trait, but a socio-political construction. As such, it
can also be abolished by means of politics.
Aaberge, Rolf, Anthony B. Atkinson and Jørgen Modalsli (2016) “On the Measurement of Long-Run Income
Inequality: Empirical Evidence from Norway, 1875–2013”, Statistics Norway Research Department
Discussion Papers No. 847 (Oslo: SSB).
Åberg, Martin (1998) Samförståndets tid: Konflikt, samarbete och nätverk i svensk lokalpolitik (Stockholm:
Historiska Media).
Aidt, Toke, Jayasri Dutta and Elena Loukoianova (2006) “Democracy comes to Europe: Franchise extension and
fiscal outcomes 1830–1938”, European Economic Review 50: 249283.
Åkerman, Sune (1967) Skattereformen 1810: Ett experiment med progressiv inkomstskatt (diss., Uppsala
Alapuro, Risto (1988) State and Revolution in Finland (Berkeley: University of California Press).
Alestalo, Matti and Stein Kuhnle (1987) “The Scandinavian Route: Economic, Social, and Political Developments
in Denmark, Finland, Norway and Sweden”, International Journal of Sociology 16: 3-38.
Alexandersson, Erland (1975) Bondeståndet i riksdagen 17601772 (Lund: Gleerups).
Åmark, Klas (2005) Hundra år av välfärdspolitik: Välfärdsstatens framväxt i Norge och Sverige (Umeå: Borea).
Ambjörnsson, Ronny (1988) Den skötsamme arbetaren: idéer och ideal i ett norrländskt sågverkssamhälle 1880
1930 (Stockholm: Carlssons).
Anderson, Perry (1974) Lineages of the Absolutist State (London: Verso).
Archer, Robin (2008) Why Is There No Labor Party in the United States? (Princeton, NJ: Princeton University
Aronsson, Peter (1992) Bönder gör politik: Det lokala självstyret som social arena i tre smålandssocknar, 1680
1850 (Bibliotheca Historica Lundensis 72. Lund University Press).
Atkinson, Anthony B. and Jakob Eghold Søgaard (2016) “The Long-Run History of Income Inequality in
Denmark”, The Scandinavian Journal of Economics 118(2): 264291.
Bäck, Kalle (1984) Bondeopposition och bondeinflytande under frihetstiden: Centralmakten och
östgötaböndernas reaktioner i näringspolitiska frågor (Stockholm: LTs förlag).
Barrling Hermansson, Katarina (2004) Partikulturer: Kollektiva självbilder och normer i Sveriges riksdag (diss.,
Uppsala University).
Bengtsson, Erik (2014) “Labour's share in twentieth-century Sweden: a reinterpretation”, Scandinavian Economic
History Review 62(3): 290314.
Bengtsson, Erik, Anna Missiaia, Mats Olsson and Patrick Svensson (2017a) “Wealth Inequality in Sweden, 1750–
1900”. Forthcoming in the Economic History Review.
Bengtsson, Erik, Anna Missiaia, Mats Olsson and Patrick Svensson (2017b) “The Wealth of the Richest: Inequality
and the Nobility in Sweden, 1750–1900”. Mimeo, Lund University.
Berggren, Henrik and Lars Trägårdh (2006) Är svensken människa? Gemenskap och oberoende i det moderna
Sverige (Stockholm: Norstedts).
Berman, Sheri (1998) “Path Dependency and Political Action: Re-examining Responses to the Depression”,
Comparative Politics 30(4): 379400.
Blackbourn David and Geoff Eley (1984) The Peculiarities of German History: bourgeois society and politics in
nineteenth-century Germany (Oxford: Oxford University Press).
Bokholm, Rune (1998) Kungen av Skåne: En bok om statsmannen Arvid Posse (Lund: Historiska Media).
Carlsson, Sten (1973) Ståndssamhälle och ståndspersoner 17001865: studier rörande det svenska
ståndssamhällets upplösning. 2nd ed. (Lund: Gleerups).
Casparsson, Ragnar (1966) LO: Bakgrund, utveckling, verksamhet (Stockholm: Prisma).
Castles, Francis G. (1973) “Barrington Moore’s Thesis and Swedish Political Development”, Government and
Opposition, pp. 313331.
Cederholm, Mathias (2007) De värjde sin rätt: senmedeltida bondemotstånd i Skåne och Småland (diss., Lund
Christensen, Jan (1997) Bönder och herrar: Bondeståndet i 1840-talets liberala representationsdebatt. Exemplen
Gustaf Hierta och JP Theorell (diss., Gothenburg University).
Claréus, Anders (2002) Primitiva bönder? Något om allmogens syn på statsmakt politik och nation under 1700-
talet”, in Åsa Karlsson and Bo Lindberg (eds.) Nationalism och nationell identitet i 1700-talets Sverige
(Uppsala: Opuscula Historica Upsaliensia), pp. 3547.
Clark, Gregory (2014) The Son Also Rises (Princeton, NJ: Princeton University Press).
Crouch, Colin (1993) Industrial Relations and European State Traditions (Oxford: Clarendon).
Delong, Brad (2007) Book Review: Growing Public by Peter Lindert, Journal of Economic History 67(1): 250
Doyle, William (1992) “Myths of order and ordering myths”, in M.L Bush (red.), Social Orders and Social Classes
in Europe since 1500: Studies in Social Stratification (London and New York: Longman), pp. 218229.
Eley, Geoff (2002) Forging Democracy: The History of the Left in Europe, 18502000 (Oxford: Oxford University
Esping-Andersen, Gøsta (1985) Politics against Markets: The Social Democratic Road to Power (Princeton, NJ:
Princeton University Press).
Gadd, Carl-Johan (2002) Det svenska jordbrukets historia, III. (Stockholm: Natur och Kultur/LT förlag).
Hallenberg, Mats (2001) Kungen, fogdarna och riket: Lokalförvaltning och statsbyggande under tidig Vasatid
(Stockholm/Stehag: Brutus Östlings Bokförlag Symposion).
Harnesk, Börje (2002) “Den svenska modellens tidigmoderna rötter?”, Historisk tidskrift 122(1): 7890.
Heclo Hugh (1974) Modern Social Politics in Britain and Sweden. From Relief to Income Maintenance (New
Haven, CT: Yale University Press).
Henrekson, Magnus and Mikael Stenkula (2015) Swedish Taxation since 1862 (Stockholm: IFN).
Hentilä, Seppo (1979) Den svenska arbetarklassen och reformismens genombrott inom SAP före 1914 (diss.,
University of Helsinki).
Hilson, Mary (2006) Political Change and the Rise of Labour in Comparative Perspective: Britain and Sweden
18901920 (Lund: Nordic Academic Press).
Hilson, Mary (2009) “A Consensual Democracy? The Historical Roots of the Swedish Model”, in Political
Outsiders in Swedish History, 18481932, ed. by Lars Edgren and Magnus Olofsson (Newcastle:
Cambridge Scholars Publishing), pp. 133156.
Hultqvist, Per (1954) Riksdagsopinionen och ämbetsmannaintressena: från representationsreformen till 1880-
talets början (Göteborg: Wettergren och Kerber).
Hurd, Madeleine (2000) Public Spheres, Public Mores, and Democracy: Hamburg and Stockholm, 18701914
(Ann Arbor: U of Michigan Press)
Isaksson, Anders (2000) Per Albin, IV: Landsfadern (Stockholm: Wahlström & Widstrand).
Jansson, Jenny (2012) Manufacturing Consensus: The Making of the Swedish Reformist Working Class (diss.,
Uppsala University).
Jansson, Torkel (1982) Samhällsförändring och sammanslutningsformer: Det frivilliga föreningsväsendets
uppkomst och spridning i Husby-Rekarne från omkring 1850 till 1930 (diss., Uppsala University).
Johnsson, Theresa (2016) Vårt fredliga samhälle: ”Lösdriveri” och försvarslöshet i Sverige under 1830-talet
(Uppsala: Acta Universitatis Upsaliensis.)
Karlsson, Sten O. (2001) Det intelligenta samhället: En omtolkning av socialdemokratins idéhistoria (Stockholm:
Karlsson Schaffer, Johan (2010) “The forgotten revolution: Challenging conventional wisdom on Sweden’s
transition to democracy”. Mimeo, available at
forgotten-revolution-2010-04-27.pdf (retrieved 23 September, 2016)
Katzenstein, Peter J. (1985) Small States in World Markets: Industrial Policy in Europe (Ithaca, NY: Cornell
University Press).
Kjellberg, Anders (1983) Facklig organisering i tolv länder (Lund: Arkiv).
Korpi, Walter (1983) The Democratic Class Struggle (London: Routledge & Kegan Paul).
Lewin, Leif (1967) Planhushållningsdebatten (Stockholm: Almqvist & Wiksell).
Lidtke, Vernon D. (1985) The alternative culture: socialist labor in imperial Germany (Oxford University Press).
Lindblad, Hans (2015) Karl Staaff: Försvaret och demokratin (Stockholm: Ekerlids förlag).
Linde, Martin (2009) I fädrens spår? Bönder och överhet i Dalarna under 1700-talet (Hedemora: Gidlunds).
Lindert, Peter (1986) “Unequal English Wealth since 1670”, Journal of Political Economy 94(6): 11271162.
Lindert, Peter (1994) “The Rise of Social Spending, 1880–1930”, Explorations in Economic History 31: 137.
Linderborg, Åsa (2000) Socialdemokraterna skriver historia: Historieskrivning som ideologisk maktresurs 1892
2000 (Stockholm: Atlas).
Luebbert, Gregory (1991) Liberalism, Fascism or Social Democracy: Social Classes and the Political Origins of
Regimes in Interwar Europe (New York and Oxford: Oxford University Press).
Lundh, Christer (1999) The Social Mobility of Servants in Rural Sweden, 17401894, Continuity and Change
14(1): 5789.
Lundkvist, Sven (1974) Politik, nykterhet och reformer: En studie i folkrörelsernas politiska verksamhet 1900-
1920 (diss., Uppsala University).
Lundkvist, Sven (1977) Folkrörelserna i det svenska samhället 1850-1920 (Stockholm: Almqvist & Wiksell
Magnusson, Lars (1988) Den bråkiga kulturen: förläggare och smideshantverkare i Eskilstuna 18001850
(Stockholm: Författarförlaget).
Macfarlane, Alan (1978) The Origins of English Individualism: The Family Property and Social Transition
(Oxford: Blackwell).
Mahoney, James (2000) “Path Dependence in Historical Sociology”, Theory and Society 29(4): 507548.
Mahoney, James (2003) “Knowledge Accumulation in Comparative Historical Research: The Case of Democracy
and Authoritarianism”, in James Mahoney and Dietrich Rueschemeyer (ed.), Comparative Historical
Analysis in the Social Sciences (Cambridge: Cambridge University Press), pp. 132176.
Maktutredningen (1990) Demokrati och makt i Sverige: Maktutredningens huvudrapport. SOU 1990:44
(Stockholm: Fritzes).
Mayer, Arno (1981) The Persistence of the Old Regime (New York: Pantheon).
Mellquist, Einar D. (1974) Rösträtt efter förtjänst? Rösträttsdebatten om den kommunala rösträtten i Sverige
18621900 (Stockholm: Stadshistoriska institutet).
Milanovic, Branko (2016) Global Inequality: A New Approach for the Age of Globalization (Cambridge, MA:
Harvard University Press).
Möller, Tommy (2007) Svensk politisk historia: Strid och samverkan under tvåhundra år (Lund: Studentlitteratur).
Moore, Barrington (1966) Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the
Modern World (Boston: Beacon Press).
Myrdal, Janken (1999) Det svenska jordbrukets historia, II: Jordbruket under feodalismen 10001700.
(Stockholm: Natur och Kultur/LT förlag).
Nilsson, Göran B (1973) ”Svensk liberalism vid mitten av 1800-talet", in Steven Koblik (ed.) Från fattigdom till
överflöd (Stockholm: Wahlström & Widstrand).
Nilsson, Torbjörn (1994) Elitens svängrum: Första kammaren, staten och moderniseringen 18671886
(Stockholm Studies in History 50. Almqvist och Wiksell International.)
Norrby, Göran (2005) Adel i förvandling: Adliga strategier och identiteter i 1800-talets borgerliga samhälle (diss.,
Uppsala University. Studia Historica Upsaliensia 217).
Norrby, Göran (2011) Ordnade eliter: Organiseringen av Nordens statsbärande skikt 16601920 (Stockholm:
Ölmedal, Ida (2016) ”Imperiet går tillbaka”, Fokus, 29 January 2016. Available at (accessed 1 March 2017).
Östberg, Kjell (1996) Kommunerna och den svenska modellen: Socialdemokratin och kommunalpolitiken fram till
Andra världskriget (Stockholm/Stehag: Brutus Östlings Bokförlag Symposion).
Österberg, Eva (1989) “Bönder och centralmakt i det tidigmoderna Sverige. Konflikt - kompromiss - politisk
kultur”, Scandia 55(1): 7395.
Österberg, Eva, Irene Andersson, Lars Hultman, Kenneth Johansson and Roger Johansson (2002) ”Fredliga Moder
Svea socio-politiskt våld och den svenska modellen”, in Eva Österberg (ed.) Socialt och politiskt våld:
perspektiv på svensk historia (Lund: Historiska Media), pp. 1146.
Østerud, Øyvind (1978) Agrarian Structure and Peasant Politics in Scandinavia (Oslo/Bergen/Tromsø:
Piketty, Thomas (2014) Capital in the Twenty-First Century (Cambridge, MA: Harvard University Press).
Pontusson, Jonas and Peter Swenson (1996) Labor Markets, Production Strategies, and Wage Bargaining
Institutions: The Swedish Employer Offensive in Comparative Perspective”, Comparative Political Studies
29: 223250.
Premfors, Rune (2003) Democratization in Scandinavia: The Case of Sweden”, Score Rapportserie 2003:8
(Stockholm: Stockholm Center for Organizational Research).
Roberts, James S. (1982) “Drink and the labour movement: the Schnaps boycott of 1909”, in Richard Evans (ed.)
The German Working Class 1888-1933: The Politics of Everyday Life (London: Croom Helm), pp. 80107.
Roine, Jesper and Daniel Waldenström (2008) “The Evolution of Top Incomes in an Egalitarian Society: Sweden,
1903–2004”, Journal of Public Economics 92: 366387.
Roine, Jesper and Daniel Waldenström (2009) “Wealth Concentration over the Path of Development: Sweden,
1873–2006”, Scandinavian Journal of Economics 111(1): 151187.
Roth, Guenter (1963) The Social Democrats in Imperial Germany: A Study in Working-Class Isolation and
National Integration (Totowa, NJ: The Bedminster Press).
Rothstein, Bo (1986) Den socialdemokratiska staten: reformer och förvaltning inom svensk arbetsmarknads- och
skolpolitik (Lund: Arkiv).
Rueschemeyer, Dietrich, Evelyne Huber Stephens and John D Stephens (1994) Capitalist Development and
Democracy. Reprint. (Oxford: Polity Press).
Rydgren, Jens (2002) “Radical Right Populism in Sweden: Still a Failure, But for How Long?”, Scandinavian
Political Studies 25(1): 2756.
Sandberg, Lars G. (1979) “The Case of the Impoverished Sophisticate: Human Capital and Swedish Economic
Growth before World War I”, Journal of Economic History 39(1): 225241.
Särlvik, Bo (1970) “Voting Behavior in Shifting ‘Election Winds’”, Scandinavian Political Studies 5: 241283.
Scarrow, Susan (2002) “Parties Without Members? Party Organization in a Changing Electoral Environment”, in
Russell J. Dalton and Martin P. Wattenberg (eds.) Parties Without Partisans: Political Change in Advanced
Industrial Democracies (Oxford: Oxford University Press), pp. 80102.
Scherp, Joakim (2013) De ofrälse och makten: En institutionell studie av riksdagen och de ofrälse ståndens politik
i maktdelningsfrågor 1660-1682 (diss., Stockholm University).
Sejersted, Francis (1993) “Den norske Sonderweg”, in Demokratisk kapitalisme (Oslo: Universitetsforlaget), pp.
Shammas, Carole (1993) “A new look at long-term trends in wealth inequality in the United States”, American
Historical Review 98(2): 412431.
Sørensen, Øystein and Bo Stråth (1997) “Introduction”, in Sørensen and Stråth (eds.) The Cultural Construction
of Norden (Oslo: Scandinavian University Press), pp. 124.
Stenius, Henrik (1980) “The breakthrough of the principle of Mass organization in Finland”, Scandinavian Journal
of History 5: 197217.
Stephens, John D. (1979) “Class Formation and Class Consciousness: A Theoretical and Empirical Analysis with
Reference to Britain and Sweden”, British Journal of Sociology 30(4): 389414.
Stephens, John D. (1989) “Democratic Transition and Breakdown in Western Europe, 1870-1939: A Test of the
Moore Thesis”, American Journal of Sociology 94(5): 10191077.
Stråth, Bo (1988) “Continuity and Discontinuity in Passing Front I and II. Swedish 19th Century Civil Society:
Culture, Social Formations and Political Change”, in Bo Stråth (ed.) Democratisation in Scandinavia in
Comparison (Gothenburg: Department of History, University of Gothenburg), pp. 2141.
Stråth, Bo (2012) Sveriges historia 18301920 (Stockholm: Norstedts).
Sundell, Anders (2014) “Nepotism and Democracy”, Quality of Government Working Paper Series 2014:16
(Gothenburg: University of Gothenburg).
Svensson, Rebecca (2004) När järnarbetare hanterar spaden och målaren knackar makadam. Om
arbetslöshetspolitik i en arbetarstyrd kommun, Västerås, under 1920-talets krisår (Diss, Acta Universitatis
Upsaliensis, Uppsala studies in economic history 70, Uppsala).
SNES, Swedish National Election Studies Program. Data available from
The project which produces the data is
Statistiska Centralbyrån (1898) Statistisk Tidskrift, issue 113-115 (Stockholm: SCB).
Thelen Kathleen (1999) “Historical Institutionalism in Comparative Politics”, Annual Review of Political Science
2: 396404.
Thermaenius, Edvard (1928) Lantmannapartiet: Dess uppkomst, organisation och tidigare utveckling (diss.,
Uppsala University).
Tilly, Charles (1990)
Tilton, Tim (1974) “The Social Origins of Liberal Democracy: The Case of Sweden”, American Political Science
Review 68(2): 561571.
Tingsten, Herbert (1941) Den svenska socialdemokratiens idéutveckling, two volumes (Stockholm: Tiden).
Tjerneld, Andreas (1983) Från borgarståndets storhetstid: Statsbudgeten som partiskiljande fråga i den sena
ståndsriksdagen (diss., Stockholm University).
Ulväng, Göran (2013) ”Betydelsen av att äga en herrgård: Herrgårdar, ståndsgårdar och gods i Uppsala län under
1700- och 1800-talen”, Historisk Tidskrift för Finland 98(2): 287316.
Uppenberg, Carolina (2017) “The Servant Institution during the Swedish Agrarian Revolution the Political
Economy of Subservience. Forthcoming in an anthology edited by Jane Whittle.
Vallinder, Torbjörn (1962) I kamp för demokratin: Rösträttsrörelsen i Sverige 1886-1900 (Stockholm: Natur och
Wehler, Hans-Ulrich (1973) Das deutsche Kaiserreich 18711918 (Göttingen: Vandenhoeck).
Winberg, Christer (1985) Grenverket. Studier rörande jord, släktskapssystem och ståndsprivilegier. (Stockholm:
Institutet för rättshistorisk forskning).
Wohlin, Nils (1909) Den jordbruksidkande befolkningen i Sverige 17511900. Emigrationsutredningen, bilaga IX
(Stockholm: P. A. Norstedts & söner).
Wollstonecraft, Mary (1796) Letters Written During a Short Residence in Sweden, Norway, and Denmark.
Available at: .
World Wealth and Income Database. (Accessed 1 March 2017).
... There are variations across the region (Keskinen, Skaptadóttir & Toivanen, 2019), but let me exemplify with the case of Sweden. Bengtsson (2019) argues that it is commonplace amongst historians and social scientists to consider the Swedish version of the 'Nordic model' of welfare and equality as being the logical end result of a long historical trajectory of egalitarianism extending forward from early-modern free peasant farmers, or a peculiar Swedish political culture that was always egalitarian and consensus-oriented. Political parties and a wider interested community also embrace this national myth. ...
... However, challenging this, Bengtsson shows that the factual basis for this myth is weak and that Sweden in 1900 had some of the most unequal voting laws in Western Europe, and more severe economic inequality than the United States. Rather, the well-organized popular movements that emerged after 1870, with a strong egalitarian counter-hegemonic culture and unusually broad popular participation in politics, lie at the roots of twentieth-century egalitarianism in Sweden (Bengtsson, 2019). Gender researchers have pointed to the related national myths about gender equality in Sweden. ...
... Williamson (2015) asked whether Latin America missed the "twentieth-century levelling" and responded in the affirmative. In fact, similar standpoints have recently emerged in the discussion about the roots of Swedish equality (Bengtsson, 2019). Some scholars claim that Scandinavian egalitarianism was shaped in the pre-industrial period (Rothstein & Uslaner, 2005), while more recent research provides evidence that the low-income concentration of today is the result of developments in the twentieth century Gärtner & Prado, 2016;. ...
... Williamson (2015) has argued that Latin American inequality is the result of missing the levelling that took place in developed countries during the second and third quarters of the twentieth century. Gärtner and Prado (2016) and Bengtsson (2019) have contended that Swedish equality is the result of developments after World War I. The first part of our empirical investigation into standards of living and inequality in Sweden and Brazil was concerned with rates of change and comparative levels of GDP and real wages. ...
Is the present-day pattern of high inequality in South America and low inequality in Scandinavia deeply rooted in the past or the outcome of developments during the twentieth century? In this chapter, we examine this issue by providing new evidence on real and relative wages in Brazil and Sweden between 1830 and 1920. We find that real wages were comparable in the two countries until the late 1880s, after which the living standards of Swedish workers surged, while they stagnated in Brazil. As real wages followed the trajectory of GDP per capita in both countries, we find little evidence of economic dualism in explaining the development of wages. However, the gap in GDP per capita was much larger than for real wages prior to the 1880s, so we also examine the spread of earnings across space, skills, and sex in the early twentieth century. Regional inequality was greater in Brazil than in Sweden, while differences across skills and sex were similar. We conclude that the major divergence in workers’ welfare took place in the fourth quarter of the nineteenth century, whereas most other aspects of inequality began to diverge during the twentieth century. The colonial legacies of Brazil and the possibly ancient roots of Swedish equality thus seem to offer little insight into today’s patterns of inequality.
Organizing rural workers has always proved to be a challenge for the labour movement. This was especially the case in Scandinavia where well into the industrial era, labour and property relations in the agricultural countryside remained essentially feudal in character. Nonetheless, and especially in the rich agricultural districts of the southernmost province of Skåne, the Swedish labour movement had succeeded spectacularly by the interwar years. Perhaps unintuitively, a key to its success was that it focused as much money and energy on constructing new spaces of culture and leisure – so-called People's Houses and People's Parks – as it did to direct workplace organizing. Drawing on Kevin Cox's concepts of “spaces of dependence” and “spaces of engagement,” this paper explains how and why Sweden's labour unions succeeded in remaking Skåne's political geography and transformed the region into one of the strongest social-democratic districts in early-twentieth century Sweden.
Full-text available
According to Dayan and Katz, media events require live transmission, firmly dating their birth to the age of television. However, the application of such a media-centric criterion runs the risk of projecting an a-historical perspective to the phenomenon as such. By contrast, using the example of the public funeral of the Swedish King Oscar II in 1907, the purpose of this article is to scrutinize and categorize the different efforts on part of the journalists to create media events before broadcasting. Four types of journalistic practices are uncovered: (1) highlighting the attention of media in a broad sense in public space (such as church bells, flags, shop windows, etc.) as well as journalistic presence, to endow eventfulness to the occasion; (2) acting as witness ambassador to evoke the senses of the media audience; (3) mediating the witnessing public participating on location to emphasise the engagement of the whole society; (4) using different kinds of narrativization, such as cross-cutting, to bring a sense of immediacy to the reporting. In terms of theory, the ambition is to draw attention to the journalist’s role as a historical narrator, as well as to bring the historical perspective back to the discussion on media events.
Denmark, Sweden and the Netherlands have different historical patterns of industrialization, but developed similar patterns of industrial coordination and cooperation. Theories accounting for industrial relations systems (economic structure, power resources, and party/electoral systems) have difficulty accounting for the similarities among these cases. Therefore, we explore the historical depictions of labor appearing in literature to evaluate whether cross-national distinctions in cultural conceptions of labor have some correspondence to distinctions between coordinated and liberal industrial relations systems. We hypothesize that historical literary depictions of labor are associated with the evolution of industrial systems, and apply computational text analyses to large corpora of literary texts. We find that countries (Denmark, Sweden and the Netherlands) with coordinated, corporatist industrial relations in the 20th century share similar cultural constructions about labor relations dating back to at least 1770. Literary depictions found in modern coordinated/corporatist countries are significantly different from those found in Britain, a country with liberal/pluralist industrial relations systems. The research has significance for our understanding of the role of culture in the evolution of modern political economies.
In this paper, we mobilize newly available historical series from the World Inequality Database to construct world income distribution estimates from 1820 to 2020. We find that the level of global income inequality has always been very large, reflecting the persistence of a highly hierarchical world economic system. Global inequality increased between 1820 and 1910, in the context of the rise of Western dominance and colonial empires, and then stabilized at a very high level between 1910 and 2020. Between 1820 and 1910, both between-countries and within-countries inequality Q5 were increasing. In contrast, these two components of global inequality have moved separately between 1910 and 2020: Within-countries inequality dropped in 1910-1980 (while between-countries inequality kept increasing) but rose in 1980-2020 (while between-countries inequality started to decline). As a consequence of these contradictory and compensating evolutions, early 21st century neo-colonial capitalism involves similar levels of inequality as early 20th century colonial capitalism, though it is based on a different set of rules and institutions. We also discuss how alternative rules such as fiscal revenue sharing could lead to a significant drop in global inequality.
Full-text available
The article examines how people within the Church of Sweden’s leadership tried to solve ‘the problem of vagrancy’ in Sweden in the early twentieth century. In focus are the priest John Melander and the deacon Josef Flinth, who advocated and realized various activities for categories of poor and mobile men in the population. These interventions, defined as help-to-self-help, differentiated between the ‘worthy’ and the ‘unworthy’ needy. In publications and lectures, Melander and Flinth presented arguments to transfer ‘unworthy’ categories to the ‘worthy’, thereby expanding the community of value. This expansion was conditioned, however, by boundaries drawn regarding ideas on belonging and ethnicity. Working in the borderlands of the community as part of a Christian calling, Melander and Flinth contributed to the expansion of social work in the early twentieth century.
Full-text available
The rise of authoritarian populism has forced many democracies to consider how best to defend democracy against its inner enemies. In the literature on democratic self-defense, one often distinguishes between three models: a legal (militant), political (procedural) and social (integrational). If much scholarly attention is on the merits and limits of the first two models, the social model has fallen behind. This is surprising given its success in the interwar years in many Scandinavian countries, and the empirical correlation between high levels of social equality and high levels of political tolerance. This article examines the merits and limits of the social model. More specifically, it makes two contributions. First, it introduces ‘the social security’ approach proposed by early Swedish social democratic thinkers as an alternative to ‘the social homogeneity’ approach proposed by Hermann Heller. The aim is to show that they provide different solutions to the loser's dilemma: the fact that losers in a democratic election must be ready to support the winners, whose decisions are at odds with their own convictions. Second, the article examines a common objection against the social security approach, namely, that it politicizes democracy, and thereby undermines the distinction between procedure and substance in the defense of democracy.
Sweden’s institutionalized employment protection legislation, ‘LAS’, is interesting theoretically because parts of it are semi-coercive. The semi-coerciveness makes it possible for firms and unions under collective agreements to negotiate departures from the law. Thus, the law is more flexible than the legal text suggests. The present study explores intended and unintended consequences of LAS as experienced by managers of smaller manufacturing companies. The results suggest that managers support the idea of employment protection in principle but face a difficult balancing act in dealing with LAS. From their point of view, the legislation’s institutional legitimacy is low, producing local cultures of hypocrisy and pretense. The article gives insights into how institutions aimed at specific, intended behavior sometimes end up producing unintended consequences fostering the opposite.
Surprisingly little attention has been devoted to thinking about knowledge accumulation in the social sciences. Although many social scientists agree that a basic purpose of their research is to produce cumulative knowledge about the world, they do not often reflect on the nature of such accumulation and how it takes place. Currently, we lack a clear answer to questions such as: What constitutes knowledge accumulation in the social sciences? How can it be measured and compared across different scholarly research programs? What promotes or inhibits knowledge accumulation in the social sciences? One might be inclined to see these questions as raising epistemological issues that cannot be resolved across disparate scholarly communities. However, I shall argue in this essay that most social scientists in fact agree on how to answer these questions, at least implicitly and in broad terms. In attempting to outline this shared position, I seek to initiate a larger discussion about how to study knowledge accumulation in social science analysis. My immediate goal is to assess the extent of knowledge accumulation that has taken place in the field of comparative historical analysis, a research area that is sometimes criticized as having failed to achieve cumulative knowledge. I limit my discussion to studies of the origins of democratic and authoritarian national regimes.
How much of our fate is tied to the status of our parents and grandparents? How much does this influence our children? More than we wish to believe. While it has been argued that rigid class structures have eroded in favor of greater social equality, The Son Also Rises proves that movement on the social ladder has changed little over eight centuries. Using a novel technique-tracking family names over generations to measure social mobility across countries and periods-renowned economic historian Gregory Clark reveals that mobility rates are lower than conventionally estimated, do not vary across societies, and are resistant to social policies. The good news is that these patterns are driven by strong inheritance of abilities and lineage does not beget unwarranted advantage. The bad news is that much of our fate is predictable from lineage. Clark argues that since a greater part of our place in the world is predetermined, we must avoid creating winner-take-all societies. Clark examines and compares surnames in such diverse cases as modern Sweden, fourteenth-century England, and Qing Dynasty China. He demonstrates how fate is determined by ancestry and that almost all societies-as different as the modern United States, Communist China, and modern Japan-have similarly low social mobility rates. These figures are impervious to institutions, and it takes hundreds of years for descendants to shake off the advantages and disadvantages of their ancestors. For these reasons, Clark contends that societies should act to limit the disparities in rewards between those of high and low social rank. Challenging popular assumptions about mobility and revealing the deeply entrenched force of inherited advantage, The Son Also Rises is sure to prompt intense debate for years to come.