zyxwvuts
zyxwvu
zyxwv
zyxwv
European Journal of Political Research 26: 59-79, 1994.
Publishers. Printed in the Netherlands.
0 1994 Kluwer Academic
The rise and decline of corporatism: The case of Sweden
LEIF LEWIN
zyxwvut
zyxw
zy
Uppsala University, Sweden
Abstract. Corporatism is a method to pacify intense minorities by giving them another o p p r tunity to influence politics when they have no chance in parliament.
possibility helps to
keep the system together; minorities get an incentive to stick to the system and social integration
is promoted. During the 19809 we have, however, witnessed a gradual decline of this neocorporatist model of interest representation. Europe is approaching the Ametican pluralist model
instead. Sweden, once the prototype of the Social Democratic Corporatist State, is the best
example of this change.
The problem of intensity
From a theoretical point of view corporatism can be understood as a way
out of the ‘intensity problem’ of democracy.
In a democracy the will of the majority prevails regardless of the intensity
of the citizens’ desires. Their preferences are to be counted, not weighed.
As a result, a lukewarm majority can at times hold sway over a strongly
committed minority. Groups that are especially affected by an issue then
have to accept taking a back seat to a greater multitude despite the fact that
those who constitute the latter do not care very much which outcome the
question takes. Such a decision-making procedure can naturally put a strain
on social integration. Conflicts of various kinds arise: strikes, revolutions,
and civil war spring from the frustrations of the dissatisfied. Ultimately a
minority that considers itself wronged in relation to its central concern can
decide to leave the community. The problem of intensity can pose a threat
to the survival of a society.
Is it possible within the framework of democratic theory to modify majority
rule so that it becomes easier - to cite a renowned article - ‘to preserve
good relations among the members of the group’ when those in the majority
have ‘mild‘ preferences and therefore ‘don’t much care’ which outcome
the question takes while the minority has ‘strong’ preferences and ‘cares
enormously’ which decision is taken (Kendd & Carey 1968: 5)?
Political scientists are at present inclined to answer this question in the
negative. The discourse confirms Robert Dahl’s well-known conclusion that
‘the analysis strongly suggests, although it does not prove, that no solution to
the intensity problem through constitutional or procedural rules is attainable’
(Dahl 1956: 119).
zyxwvu
zyxwv
60
zyxwvutsr
zyxwvu
zyxwv
z
zyxwv
zyxw
zyxwvu
Interest organizations. Instead a pragmatic solution has emerged: organizational participation is a way to handle the ‘intensity problem’. Since the
parliamentary channel cannot be formed to satisfy everyone, a second channel has been opened - a corporatist channel. The freedom to organize is as
fundamental to democracy as the principle of general and equal franchise.
Whenever dictatorships arise, one of the first steps is to crush free organizations. Furthermore, when special interests are concerned, they are given
extra influence in the political decision-making process and in the implementation of decisions. Consequently, by offering the intensely committed an
extra chance corporatism helps preserve peace in society.
The two-channel system makes the political decision-making process more
unpredictable than it would have been if one only had to count votes. This
uncertainty is a requisite for keeping the system together: there must be
some possibility that one’s proposal will be accepted despite one’s being in
a minority, a possibility that prevents one from immediately giving up. Even
though, to take examples from the Swedish case we are going to investigate,
neither the die-hards amongst opponents to increased taxes for house-owners
nor the strong advocates of a ban on pornography see any chance of convincing a majority in Parliament of the rightness of their view, it may nevertheless
be possible. for them to succeed by exerting influence through the Swedish
Association of House-Owners or the National Campaign against Pornography, respectively. Influence through organizations gives rational individuals
reason to stick to the system and to hope for the best, even if they for the
moment don’t control a parliamentary majority.
Social integration
‘Social integration’ is in other words what corporatism is supposed to further:
it is the dependent variable in our analysis.
The corporative society. As usual, the history of political thought helps us
get our bearings. The main thrust of the French Revolution and its complex
of democratic ideas was directed against the very notion that there is any
value in trying to establish social integration through organizations. The
remnants of a corporatist form of government inherited from the Middle
Ages were regarded by the radicals as nothing more than an instrument for
preserving the privileges of the .upper class. ‘Peace is a state of suppression’
(Machiavelli [1513]1983). The constitutional principle of the Revolution was
to sweep away the whole middle level of organizations and to attach liberties
directly to each individual perso’n. Estates, classes, and guilds would no
longer be legal subjects. Only the two extremities of society - the state and
the citizen - won political recognition. The classic formulation of this anticorporatist philosophy had been penned by Rousseau himself ‘In order for
z
zyxw
zyxwv
61
the general will to be well expressed, it is therefore important that there be
no partial society in the State’ (Rousseau [1762]1978:61).
The corporations were consequently abolished through a Revolutionary
decree. Political equality was to be realized by directing the citizens exclusively to the parliamentary channel, in which all would have the same influence
by having an equal vote.
It is against this background that the ideology of corporatism emerges.
Tired of the violence of the Revolution people began to turn again to
those who promised social harmony. The anti-revolutionary conservatism
constituted an attempt to regain lost social integration through corporative
arrangements: ‘To be attached to the subdivision, to love the little platoon
we belong to in society, is the first principle - the germ as it were - of public
affections. It is the first link in the series by which we proceed towards a
love to our country and to mankind’ (Burke [1790]1969: 135).
This defense of corporative society is subsequently repeated throughout
the conservative nineteenth century. Between the citizen and the state there
should be a finely meshed net of organizations whose purpose was to ensure
that this idyll continued to be undisturbed by social unrest (Bowen 1947).
Would the mass organizations of the twentieth century play the same role
for social integration as the old corporations? This became the great question
debated by the most powerful of the new coalitions - the trade unions. The
debate led, as is well known, to a split in the workers’ movement. The
revolutionary group under Lenin wanted to ‘crush capitalist society’. The
reformists’ policy became to start negotiating - on condition that there was
‘constant improvement in the condition of the workingmen’. The state took
on the role of carrying out and legitimizing this class compromise. In an
often cited work, this development is summarized as follows: ‘[Tlhe state
institutionalizes, coordinates, and enforces compromises reached by a class
coalition that encompasses both workers and capitalists’ (Przeworski & Wallerstein 1982: 215)
The French Revolution had eliminated corporatism and formal equality
had been proclaimed. However, in the liberal market thereby created it was
not equality but glaring class contradictions that emerged. The trade-union
movement therefore tried to achieve through negotiation the equality that
the doctrine favoured but seemed impotent to bring about. The organizations
were in turn embraced by the welfare state, and it was through them that
the citizens developed their loyalty to the state. The state became dependent
upon the organizations for the implementation of its policies. The leaders of
the’ organizations became more and more independent of their members
through support and subsidies from the state (Schmitter 1981; Offe &
Wiesenthal 1985;Birnbaum 1988;Williamson 1989).
From the history of political thought we thus learn that it was ‘social
integration’ the supporters of corporatism saw as its benefit, not ‘growth’
which economists often take as an obvious criterion of evaluation (Olson
1982) - even if the two sometimes but not always go together. The econom-
zyxwv
zyxwvu
z
zyxwvu
zyxwvu
ists’ point should however not be forgotten, for in many aspects it is possible
to reason about the effect on social integration as they do about the effect
on growth. So let us reconsider Olson’s argument for a moment. It says, as
we know, that organizations within states have an incentive to be free riders.
They direct their attention to redistribution instead of economic growth, and
there is no limit to how far they are prepared to go when it comes to passing
on costs to others, with catastrophic consequences for the economy as a
whole.
Olson notes an important qualification, however. If an organization becomes truly ‘encompassing’, he writes, it changes its attitude towards society
and plays the very role vis-A-vis economic growth that the theorists of corporatism have attributed to it with respect to the maintenance of social integration. This addendum explains why Olson finds the Swedish situation with
strong organizations and rapid growth ‘what our model would predict’: since
Sweden’s - and Norway’s - interest organizations are so large, their ambition
to further the general interests of society is what he expects.
Olson’s claim is ‘far from new. He merely repeats, albeit with unusual
brilliance and empirical thoroughness, the experience emphasized by the
theoreticians of corporatism, namely that organizations cultivate the social
solidarity of the.citizens. Olson distinguishes himself from his predecessors
only in the definite line he draws between the effects of small and large
organizations.
Following Olson, we meet with two hypotheses. According to the first,
the effect of the emergence of interest organizations, as long as they are
small, is that they work against social integration by attending solely to their
separate interests. According to the other hypothesis, once the organizations
have become encompassing, their effect will be the direct opposite: they
foster social peace, they constitute such a large part of society that they do
not judge it possible to roll over their costs to others.
zyx
zyxwvu
Peace in the labour market in Sweden. Since this article summarizes research
on the development of corporatism in Sweden, let us now turn to empirical
data to test the relation between organizational participation and social
integration. Sweden has been spared both revolutions and civil war but it
has suffered widespread conflicts in the labour market. These are described,
using an inverted scale, by the unbroken line in Figure 1.
The labour-market statistics include three measures: the number of workers affected by strikes and locksuts, the number of conflicts, and the number
of working-days lost through conflicts. Here the first measure is used but a
similar picture of the development in Sweden emerges regardless of which
measure is chosen. The frequent Strikes of the 1920s and 1930s were also
long-lasting; the few strikes of the 1950s, and 1960s were generally short.
Three periods in the development of corporatism in Sweden can be discerned: its rise between 1900 and 1935 with widespread conflicts, its maturity
zyxwvuts
zyxwvutsrq
zyxwvut
zyxwvutsrq
63
Dcgrse of
involved
1900
1910
1920
1-
1940
lOs0
lo0
1970
1980
lop0
103% 0
w
10
70
30
63
40
zyxwvuts
zyxwvutsr
5 3 5 0
Ij
33
il
1J
60
:7
a0
90
\
.
W
n
v
Maturity
RiSC
zyxw
Decline.
-
Fig. 1. Labour market peace and degree of organization in Sweden 1900-90.
Number of
workers involved in strikes and lockouts (thousands); iii~iiiDegrecof organization,LO (percent).
Sources: Swedish Official Statistics; Swedish Organizational Statistics.
in the period 1935-70 with steady improvement in relations between labourmarket parties, and its decline in 1970-90, when conflicts again increase.
At the beginning of the first period the unbroken curve plunges steeply.
Industrial peace was precarious and strikes increased as industrialization
proceeded. The disturbances culminated in the general strike of 1909 when
about 300 thousand workers were involved in conflicts. Conditions remained
poor throughout the 192Os, and early 1930s with between 70 and 80 thousand
workers involved in conflicts each year.
A noticeable improvement in industrial relations occurred in the mid193Os, which marks the beginning of the mature period. The number of
workers affected by labour-market codicts each year declined to 30 thousand
and continued to fall to 14 thousand. The low figures for the following years
- with about five thousand strikers per year is partly an artefact since they
-
64
zyxwvuts
zyxwv
zy
zyxwv
zyxw
reflect the agreements on stabilization policy reached during the war years.
The strike of 1945 was something of a reaction to this ‘voluntary’ moderation
in wage demands. The decades that immediately follow, 1950-1970, are quite
remarkable. In some years only a few hundred workers went on strike; the
average figure for the two five-year periods 1956-60 and 1961-65 is 15
hundred workers in conflict. It is during these years that the legendary
‘Swedish model’ for peaceful settlement of labour-market conflicts gains
international attention (Elvander 1988; Schiller 1988).
This trend was broken around 1970. During the period 1970-90 the number
of workers drawn into strikes and lock-outs increased again to 20, 30, or 40
thousand a year. The nadir was reached in 1980 when the Swedish Trade
Union Confederation (LO)put 100 thousand members on strike and the
Swedish Employers’ Confederation (SAF) countered by locking out 745
thousand workers, the most far-reaching labour-market conflict in Swedish
history.
In order to understand the importance of organizations for the level of
conflict in the labo’ur market, the data on strikes are set off against data on
organizations (the broken line in Figure 1). Only figures for LO are available
for the whole period and can be used here; white-collar workers did not
become organized until later (Kjellberg 1983).
The decline in peace in the labour market during the first period coincides
with a fourfold increase in LO’S membership during the first decade of the
c e n t u j to about 185 thousand trade union members in 1907 or about 15
percent of the work force. After defeat in the general strike of 1909, LO
lost half its members, and the decade after this was marked, as noted above,
by considerably fewer conflicts in the labour market. In consequence, the
two curves in Figure 1are mirror-reflections of each other. The data elegantly
support the first hypothesis: when organizations emerge they impede good
relations and when they are weakened, relations become more peaceful
again.
The period of maturity, during which peaceful relations between the parties
were reached, was characterized by a tremendous increase in the size of the
trade-union movement. The share of organized workers passed 35 percent
in the early 193Os, reached 70 percent by the end of World War Two, and
approached 90 percent in the heyday of the Swedish model. In absolute
figures this corresponds to about 600 thousand, one million, and one and a
half million members respectively. This supports the second hypothesis: when
organization was strong, relations were good in Sweden.
Finally, what does the relation look like during the period of decline,
1970-90? Peaceful conditions deteriorated again but the share of organized
workers remained high. This is not something that could be predicted with
the help of the theories of collehive action. For the time being we must
therefore leave the increase in conflicts in Sweden after 1970 unexplained.
How does Sweden compare with other industrialized countries in this
regard? The answer is that the Swedish development is rather unique. It
zy
z
zyxwv
z
z
65
might first be noted how long Sweden was spared more widespread labour
conflicts. The reason is of course that industrialization came to Sweden quite
late. The Sweden of a hundred years ago was a backward agricultural country
with a comparatively idyllic labour market. Secondly, once industrialism
finally amved in Sweden and conflicts began to break out in the labour
market, the level of conflict generally tended to persist - apart from the
decade after the general strike - while in other countries it declined. Even
as late as the mid-l930s,Sweden had a higher number of strikes than any
other western country except Norway. Thirdly, the number of strikes and
lock-outs declined drastically after this and was in the post-war period lower
than in any comparable country (Ross & Hartrnan 1960; Shorter & Tilly
1974;Kendall 1975;Crouch & Pizzorno 1978;Korpi 1983: chap. 8).
Some time around 1970 peaceful conditions in the labour market of industrialized countries began generally to deteriorate (Hibbs 1978). The greatest
change occurred in Sweden. Sweden could no longer exhibit more peaceable
relations than other countries and amved at a middle position with about
100 workers pdr ten thousand involved in labour-market contlicts. Similar
figures for the best countries (Switzerland, Austria, Japan, France, and the
USA) were in the tens, and for the countries with higher levels (Great
Britain, Australia, and Finland) were around one thousand, with Italy having
the highest level of all with 4.6 thousand (Swedish Institute for Social Research 1991).
With respect to the degree of unionization, Sweden remained at the top
of the list, however, with 86 percent of the work force belonging to a union.
This was followed by the other Scandinavian countries, Belgium, and Austria
while Japan (28 percent), the USA (18 percent), and France (15 percent)
came at the bottom. A statistical analysis of the relation between the degree
of unionization and the absence of labour-market conflicts with 16 countries
as units of observation demonstrates that the traditional corporatist experience no longer holds; nowadays a high degree of unionization does not yield
peace in the labour market (Table 2.3, in Lewin 1992:47).
This analysis of the level of peace in the labour market in Sweden points
out two problems. The first is how the group egoism characteristic of small
organizations is actually transformed into a concern for the broader interests
of society that marks encompassing organizations. The second problem is to
account for the sudden deterioration in the peacefulness of the labour market
in the period of decline 1970-90. Neither the first nor the second hypothesis
can explain this. Why was a situation that basically benefited all parties
suddenly abandoned?
zyxwv
Organizational participation
Before proceeding to the analysis of Swedish corporatism, let us consider a
conceptual problem concerning the independent variable.
66
zyxwvutsr
zyxwv
zyxwv
The definition of corporahm. The most important contribution has been
made by Philippe Schmitter, who got modem research on corporatism started
and made an illuminating distinction between corporatism and pluralism:
‘Corporatism can be defined as a system of interest representation in which
the constituent units are organized into a limited number of singular, compulsory,noncompetitive, hierarchically ordered and functionally differentiated
categories, recognized or licensed (if not created) by the state and granted
a deliberate representational monopoly within their respective categories in
exchange for observing certain controls on their selection of leaders and
articulation of demands and supports’ (1974).
It is not surprising that such a pedantic definition has in turn inspired a
formidable wave of exegetic comment; many of Schmitter’s followers have
run into considerable difficulties in trying to find their way out of this labyrinthine definition (Lewin 1992: 49-57). My conlusion is that it is vital to
distinguish between intentions, institutional arrangements, and outcomes and
let the definition of corporatism apply solely to the second of these three
phenomena. The intent with which some such arrangement is set up should
be left open to empirical investigation. Otherwise one might well ask whether
such collaboration between the state and organizations should not be called
corporatist because it was not begun deliberately but rather arose through a
random process uncontrolled by any particular actor as, it is often maintained, has occurred in the welfare state. Similarly, one might wonder
whether such collaboration is not to be called corporatist simply because it
does not result in the stipulated outcomes (‘in exchange for’) either regarding
the structure of the organizations or the implementation of welfare policy.
What the consequences of the collaboration might be is a matter to be left
open for hypotheses and not postulated by definition.
Inspired by Schmitter, I propose the following definition and hypotheses:
zyxw
zyxwvu
Definition:By ‘corporatism’ is meant the officially sanctioned participation
of organizations in decisions governing the affairs of the state or in their
administration, or similar actions camed out by organizations on behalf
of the state.
Hypothesis 1: The more corporatist the decision-making process, the more
monopolistic and hierarchically structured organizations will be and the
more centralized will be negotiations.
Hypothesis 2: The more corporatist the decision-making process, the more
successful welfare policy will be in the sense that outcome corresponds
with intent.
ZmtitutionaI arrangements. A large amount of data has been collected for
various indicators of corporatism as well as for the two hypoheses and will
zyxw
zyxwvu
z
zyxw
zyxw
zyxwvu
67
be introduced briefly here; a complete presentation is to be found in Lewin
1992.
One form of organizational participation is committee representation. Traditionally ministries have remained small in Sweden and executive functions
have been camed out by separate agencies that do not take orders from the
government. Ad hoc committees are set up to prepare major reforms. Earlier
those recruited to sit on such committees were almonst exclusively public
officials, later politicians (HesslCn 1927). From the 1930s, the representatives
of organizations entered and increased the share up to one third of all
members in the 1950s and early 1960s. Then followed a deep decline to
between 10 and 15 percent (Meijer 1969; Bengtsson 1975; Johansson 1992).
This rise and subsequent decline also corresponded to a shift in the nature of
the work camed out by the committees. While welfare policy was expanding,
committees were large and worked over a long span of time in secrecy with
the aim of reaching compromises satisfactory to all parties involved. In the
end they presented a thick report with costly proposals. Later, when the
government began to make cuts in the civil service, the committees became
smaller, they were more closely tied to the government, they worked more
quickly, and they stimulated debate with discussion papers. Once again
they were made up predominantly of public officials; the representatives of
organizations had to step back.
Another form of participation is the organizations’ consideration of reports.
When a committee has concluded its work, its report is sent out to various
bodies. Originally only state authorities had the right to be thus consulted.
Beginning in the 1930s it also became common to ask for the opinion of the
large organizations. This form of participation has declined in importance
today with the changing nature of the work carried out by such committees
(Swahn 1980).
A third form is representation on the boards of state agencies. Interest
organizations participate not only at the decision-making stage of public
affairs but also in their administration. During the period of maturity, organizations gained the right to be represented on the boards of governmental
agencies. Between the 1940s and 1970s, the proportion of agencies with
organizational representation increased from 28 percent to 74 percent (Molin,
Mhsson & Strdmberg 1975: 47). During the 1970s, one third of all the
members on agency boards were representatives of organizations (Hadenius
1978: 24-28; Westerstah1 & Persson 1975: 60-80). After this peak a dramatic
change occurred. Perhaps the clearest indication of decorporativization in
Sweden today is the fact that the interest organizations, headed by S A F ,
began to leave the boards of governmental agencies; the reasons for this
change will be discussed below.
In former times there was often direct contact between citizens and politicians in the local community and on this level the corporatist channel was
hardly developed. In the 193Os, however, a new form of organizational
participation in local government was introduced: Sweden switched to the so-
68
zyxwvut
zy
zyx
zy
zy
zyxw
called Ghent system of administration for unemployment insurance. In the
place of a state agency the trade unions now took charge of unemployment
insurance funds on municipality level and trade-union representatives were
given the right to decide which jobs an unemployed worker could refuse
without losing relief. This became an effective way for the unions to recruit
new members; in countries where the Ghent system is used, the degree to
which workers are unionized is comparatively high (Rothstein 1992). When
the crisis of the 1930s was followed by the economic regulations of the war
years, the same path was followed at the local level; war-time regulations
took on a highly corporative aspect. This is a tendency that has not been
reversed during the period of decline from 1970 to 1990. Organizational
participation is still large in local government (Soderlind & Petersson 1988:
95-96, 105; Strumberg & Westerstah1 1983: 298; Gustafsson 1988: 275-277;
Gidlund et al. 1982: 13, 134).
Finally, organizations participate through agreements. During the Great
Depression of the 1930s and the regulations of the Second World War, the
government begm‘to make agreements with various branches to assure
the success of its economic policy: agriculture, the building industry, the
automobile industry, the rental field, price regulation, temporary wagefreezes, etc: (Lewin 1967; Elvander 1969). With the return in recent decades
to a freer economy these regulations and agreements expired (Elvander
1988).
This is briefly the picture of the rise and fall of organizational participation
in Swedish politics in this century. These changes, we assumed in the two
hypotheses, have also had an impact on the structure of the organizations
and on the implementation of welfare policy.
Unions have gained and lost their monopoly. Between 1930 and 1970 there
was such an increase in the membership of three unions the Swedish Trade
Union Confederation (LO), the Central Organization of Salaried Employees
(TCO) , the Swedish Confederation of Professional Associations (SACO)
and their counterparts the Swedish Employers’ Confederation (SAF) and
the Federation of Swedish Farmers (LRF), that these organizations predominated in their respective areas. Their monopolistic position has subsequently been challenged with the emergence of a number of new organizations, a development that has led to extensive organizational disputes
concerning which union has the right to organize employees within some
particular area. The simple opposition between two main parties - LO and
S A F - has been transformed into an organizational structure of exceeding
complexity with the consequence that wage negotiations have become considerably more complicated (Elvander 1988: 253).
The organizations have taken on a hierarchic structure. During the period
of maturity the number of trade -unions decreased through amalgamation
(the number of unions that were members of LO declined from 43 to 21
while TCO’s membership similarly went from 43 to 21 and SACO’s from 91
to 25); the number of locals was reduced to a few ‘big locals’. The practice
zyxwv
zyxwvu
zyx
zyxw
zy
zy
zyxwv
69
of putting agreements before trade-union members for a direct vote ceased.
The number of people employed in the offices of the peak organizations in
Stockholm greatly increased.
Centralized negotiations have been established and abandoned. During the
economic boom of the 195Os, wage negotiations tended to degenerate into
free-for-alls; the later the trade unions reached an agreement, the more they
got. SAF therefore proposed centralized negotiations in which it would be
possible to hold back increases in costs. LO agreed to this after some hesitation. The first centralized negotiations were held in 1956 and, despite
general expectations, continued to be held for a quarter century. The actual
negotiations between LO and S A F were conducted by small delegations
consisting of,only three people from each side. The agreement reached
between LO and SAF then served as a norm for the rest of the labour
market. However, in 1983 the employers' organization withdrew from central
negotiations, an event that aroused international attention. When those who
do not make any clear distinction between institutional arrangements and
outcomes talk of the decline of the famous Swedish corporative system, they
cite as evidence the collapse of centralized wage negotiations (Lash & Urry
1987: 232-233, 236).
In order to evaluate the implementation of welfare policy according to
hypothesis 2, one should really go through every political decision and its
implementation, a task that is in principle possible but in practice of course
quite unfeasible. I have chosen instead to analyse the way in which the
principal goals of economic policy have been pursued; relevant data are
continually reported in official Swedish statistics, which, together with OECD
statistics, constitute the basis of the following presentation. The five goals
set out in the government's financial plan are usualIy considered to be prerequisites for welfare policy as a whole.
The most successful aspect of Swedish economic policy has been in achieving the goal o f j W employment. Sweden was a pioneer in adopting Keynesian
policies as early as the 1930s. These were supplemented after the war with
active labour-market policies, which amongst other things stimulated the
movement of labour. The high levels of unemployment of the 1930s, with
figures up to 23 percent, were eliminated and Sweden was without equal in
holding unemployment under two percent during the 195Os, 'Ws,and '70s.
Towards the end of the 198Os, however, this figure began to rise rapidly.
To some extent Sweden has paid for high levels of employment by losing
a steady price level. Compared with the other OECD countries, the rate of
inflation in Sweden during the first post-war decade was about average. In
the 1980s, by contrast, inflation in Sweden was each year higher than the
average in western Europe.
The balance of payments has fluctuated. After Sweden was obliged to
abandon the gold standard during the 19309, considerable uncertainty prevailed about what would be a reasonable exchange rate. When the pound
was devalued by 30 percent against the dollar in 1949, Sweden followed suit,
70
zyxwvutsrq
zyxwvut
zyxwvu
zy
as did many other western European countries. This started a long period
of economic expansion without any serious problems in the external balance.
The oil crises of the 1970s did create new problems but through repeated
devaluations the government attempted to improve Sweden’s competitive
position. The high rate of inflation and increases in production costs had
eaten up this advantage by the beginning of the 1990s, however, and new
devaluations were made.
As mentioned above, industrialism came to Sweden fairly late, but once
it got started, economic growth was more rapid and more constant than in
other countries. Japan alone has had a higher rate of economic growth than
Sweden over the past century: 32 percent per decade for Japan, 29 percent
for Sweden, 15-20 percent for other OECD countries (Kuznets 1971: 1114). This is probably the single most important factor that accounts for
Swedish welfare. The fact that this situation changed fundamentally in the
1970s and that the national economy has in the past few years shown no
growth at all or even a decline pose a serious threat to the future development
of welfare in Sweden.
In a country like Sweden with high taxes and many people employed in
the public sector, the government has the ability to make a considerable
impact on the structure of wages. Levelling of incomes is the official guideline.
For many years the government was successful in this aim. Swedes could see
how in the wage-pyramid depicted in official statistics they came closer and
closer to the apex as the wage differential was reduced year after year. This
trend was broken towards the end of the 1970s when adjustment to market
conditions and individual abilities again came to be accepted as norms for
determining incomes in Sweden.
To sum up, organizations were drawn into public affairs beginning in the
1930s;in the past decade or two their activities have been reduced in many
fields. Cooperation with the state has led to internal structural changes in the
participating organizations towards monopoly, hierarchy, and centralization.
The success governments have had in reaching their economic goals has
varied. Generally speaking, the period of maturity 1935-70 was extremely
successful: the spectre of mass unemployment was banished and Sweden
became a welfare state with a high and comparatively evenly spread standard
of living. The contribution made by the ‘encompassing’ organizations to this
development can perhaps be said to have amounted to help in reaching
general support for the goals of economic policy and their realization. Since
1970 the Swedish economy has performed less well. Conditions have been
characterized by stagnation and difficulties in achieving internal and external
balance.
zyxwvu
The guiding hand 1935-70. How-can we explain this rise and decline of
Swedish corporatism? What considerations, to begin with, persuaded the
central actors to cooperate in setting up corporatism in Sweden in the mid193os?
zyxw
71
This question focuses our interest on the talks held between the director
of SAF and the chairman of LO in the winter of 1936 concerning the turbulent
situation then prevailing in the Swedish labour market with widespread
strikes and lock-outs. To these men these conflicts constituted an extremely
serious social problem. In the public debate it was being asked how it could
be that a country with such a favorable economic development could be
plagued by labour-market conflicts to a greater extent than other similar
countries.
The government had made some attempt to promote peace but the organizations took over with direct contacts and an agreement between SAF and
LO was signed in 1938 at SaltsjObaden (outside Stockholm) that created a
set of rules for solving conflicts peacefully with a minimum of interference
from the government. Responsibility for maintaining peaceful relations was
assumed by the peak organizations; this feature is often pointed to as the
agreement’s ‘centralist tendency’. In the following years several further
agreements were signed by which relations between the two parties were
regulated in ode area after another (Westerstah1 1945: 189-209; Hadenius
1976: 47-49; SUderpalm 1980; Johansson 1989).
The remarkable thing about this agreement is that it was the direct opposite
of what was originally intended. The motive for it was to keep out the
government and to protect ‘the freedom of the labour market’ but the result
was to draw the organizations into the political process and to give them
responsibility for a vital area of public affairs. The intention may have been
‘pluralist’. The effect was to expand the role played by organizations in
public policy to such an extent that it deserves the name ‘corporatism’.
As we have seen, this role did nothing but increase over the next few
decades. Political scientists have established that by the end of the 1960s
Sweden was one of the most corporative societies in the world. After
Schmitter published his pioneering article, researchers began to collect comparative data and compile indices of corporatism. These indices seldom meet
the methodological criteria I have specified above. Intentions, institutional
arrangements, and outcomes with respect to organizational structure and
welfare policy are systematically mixed together. Wilensky takes the degree
of centralization in the trade-union movement as a measure of corporatism
(Wilensky 1976). Schmidt includes in his definition both the declared intentions of the parties and outcomes in terms of strikes and incomes policy
(Schmidt 1982). Lehmbruch adds the internal structure of the organizations
(Lehmbruch 1982). In his empirical studies, Schmitter himself discusses
something he calls the ‘governability’ of the various countries, a concept of
particular interest here because of its resemblance to what I have called
‘social integration’. Surprisingly enough, however, in developing his index
of corporatism he does not make use of any measure of the degree to which
organizations actively participate in the political process but instead relies
exclusively on data about the organizations’ internal structure (Schmitter
1981). Finally, when Lijphart and Crepaz published a survey article about
zy
72
zyxwvut
zyxwvu
research on corporatism in 1991, they did not set down any explicit criteria
at all but simply threw all previous lists into one summary index! Despite
this absence of rigour the results are remarkably congruent. It seems that
the participation of organized interests in politics is such a robust phenomenon that it survives even the roughest treatment. The various elements seem
to hang together so intimately that whichever one chooses to select for
examination, one manages to capture the whole complex of corporatism.
For the most part, the countries that come at the top, in the middle, and at
the bottom of the lists are the same. Around 1970, Sweden appeared to be
one of the most corporatist countries in the world, along with Austria and
Norway.
Union representatives had realized that the unseen hand of liberal ideology
was not sufficient to protect their interests. What was needed was also a
guiding hand of corporatism which could give the workers what they’could
not get through parliamentary means, since they would never achieve a
majority of their own. By the guiding hand of corporatism industrial workers
won especial infludnce in the construction of the welfare state and became
transformed from a dissatisfied group into society’s leading bearer of the
state’s interests.
zyxwvu
zyxwv
The grip slips 1970-90. How can, finally, the decline of corporatism be
explained? How did it subsequently come about that this state of affairs,
which was to everyone’s basic advantage, was abandoned? Despite all the
praise that has been sung about the benefits of ‘the Swedish model’, the
domestic actors were quite fed up with it by the end of the 1960s.They had
entered into the board rooms of various governmental bodies as interest
representatives but had instead, to use George Stigler’s phrase, become
captured by the state (Stigler 1971). SAF felt locked in as a hostage for
socialdemocratic policies it disliked on ideological grounds. LO felt trapped
into accepting a market economy, which set narrow limits on what the trade
unions could hope to achieve in the future.
It was LO that made the first move.
In the early 1970s LO launched an offensive that broadened the ambitions
beyond their traditionally accepted task. The decision of the Social Democrats to yield to LO’S position and introduce wage earners’ funds was the
most spectacular. On this issue LO had abandoned the reformist policy it
had held since the 19309 and gone back to an earlier marxist position that
ownership was decisive for social change. Part of the profit incurred by
businesses would be placed in large funds owned by the trade unions. These
funds were presented as an instrument by which the old capitalists would be
deprived of their power and the-trade unions made the most important
owners of industry.
LO also broke an old compromise by which the right of employers to be
solely responsible for organizing work and assigning tasks was recognized in
zyxwvu
zyxwv
zy
zyx
zyxwvut
zyxw
73
exchange for recognition on the part of the employers of the workers’ right
to organize; it forced through the law of co-determination, which guarantees
employees a say in the way businesses are run. State enterprise increased
markedly both through outright nationalization and through the establishment of new state-owned firms. The position of the trade unions vis-his
employers was thus strengthened in various ways.
LO seems to have been completely unprepared for SAF‘s countermove.
Opposition was certainly expected and LO was prepared to meet it at every
point both with legislation through the Social Democratic government and
through negotiations. However, S A F chose instead to bid farewell to the
whole corporative set-up; it broke the social contract from the 1930s’ one
might say. It broke off contacts with LO at the central level. It abandoned
the central negotiations for wage agreements. It said ‘no’to giving continued
priority to low-wage groups. S A F marched out of the board rooms of the
central government agencies (Lewin 1988; Henning 1980; Hadenius 1983;
De Geer 1989).
SAF‘s counte’rmove was well in keeping with social developments in other
countries. There are at present no quantitative estimates of the extent of
corporatism on which to base such a judgement. It seems that the numerous
methodological problems have deterred scholars from making further attempts at such measurement despite the fact that these problems are actually
rather trivial and could in principle be solved through more careful operationalizations. Evidence is gleaned instead from more penetrating case studies (Baglioni & Crouch 1990; Brunetta & Dell’Aringa 1990; Cox & O’Sullivan
1988; Mtny 1990; Schmitter 1989; Therborn 1987; Wilensky & Turner 1987;
Williamson 1989; Wilson 1990).
These scholars claim that the dissolution of corporatism since 1970 ultimately depends on the fact that it no longer succeeded in fulfilling its most
important task - to pacify intense minorities and promote social integration.
Instead it produced a new class society. The disadvantaged minorities were
no longer compensated through corporatist arrangementsand those favoured
by the system were not won over to social solidarity. ‘The important point
in the present context is that corporatism did not last. It may have been the
distant effect of the application of Keynes, but it led at the end of the day
to the protest of Schumpeter’s people, of the entrepreneur, or at any rate
of its ‘yuppie’ version, the casino capitalist’ (Dahrendorf 1989: 136).
As a political and moral force, corporatism is therefore considered to have
expended its power. The more direct causes of its dissolution are however
to be sought in material conditions. The prototype of the worker of the 1990s
is no longer the factory-worker, who could be replaced by thousands of his
comrades as long as they had the same physical powers, but rather the
highly trained specialist, whose contribution to production in ‘disorganized
capitalism’ is unique (Lash & Urry 1987).
The state, like capitalism, is the object of the same tendencies towards
dissolution. Macroeconomic programmes no longer work. Decentralization
-
zyx
74
zyxwvuts
z
and regionalization are the hall-marks of welfare policy whereas corporatism
flourished at the national level.
The trade unions themselves are in many western European countries
faced with the threat of becoming weaker. It is more difficult to recruit the
new specialist-worker to mass organizations. Flexibility is asked for but
unions have often been unable to muster more than sterile opposition.
The internationalization of business has accelerated the pace at which
corporatism has disintegrated in the West. It was long thought that small
European democracies managed better thanks to their corporatist structures.
Being far too dependent on their international trade to risk protectionist
experiments, these states opened their industry to tough international competition and thereby retained their efficiency; in their domestic policies they
later compensated various groups through corporatist agreements (Katzenstein 1985). It now turns out that this astute analysis was an epitaph rather
than a programme for the future. With the internationalization of capital,
enterprises make investments solely on the basis of profitability, without
regard to national 'considerations, and national corporatism has thereby lost
its compensatory role.
The tendency towards dissolution has been given strong impetus by ongoing European integration. Industry has been the main actor in this process,
whereas the trade unions, often in conjunction with Social Democratic
governments, have been the supporters of corporatism. This does not mean
that people cease to further their interests via organizations, but the mode
of participation becomes pluralist instead of corporative. Perhaps the future
for unions' influence in Sweden will be on the local level, where organizational participation has not disminished. A wave of small organizations representing special interests has swept over Brussels; American lobbying through
informal contacts intermixes with the complex bureaucracy of the EC. Besides, national corporatism was most successful in countries that are not
members of the EC: Austria, Sweden, and Norway.
Thus, in the study of corporatism, an intercontinental switch between
methodological and substantivefindings has occurred. When European scholars began the systematic study of interest organizations, they copied the
pluralist model and saw the world with American eyes. However, reality was
not comprised of a system of independent, competitive pressure groups but
rather of integrated, cooperative, hierarchically ordered organizations and
state authorities. Once European scholars had finally seen the object of
analysis in its proper light, however, they discovered that the system had
begun to change - towards greater pluralism! Brussels is now as much a home
for pressure-group politics BS Washington is. Sweden, once the prototype of
the Social Democratic Corporative State, is the best example of this fundamental change.
zyxwv
zyxw
z
zyx
75
zyxwvuts
60
zyxwvutsrqpon
zyxwv
zyxwvutsrq
f
zyx
zyxwvutsrqp
50
40
z
-
J7
30
-
-
I
19CO
-+
1
1973
1976
1979
I982
1405
:’I99
Elcction year
1991
zyxwv
Fig. 2. Distrust of parties and politicians 1 W 9 1 . ‘The parties are just interested in people’s
votes, not in their opinions.’ Sources: Gilljam and Holmberg, 1990; Department of Political
Science, University of Gothenburg. Data 1957-1991.
Trust and disappointment
Corporatism is fading away; as a result, social integration is deteriorating.
Intensely involved separate interests again challenge the trust in the Swedish
political system. Contempt for politicians, as it is usually called in the mass
media, has greatly increased since measurements were first made in election
studies in 1968. The share of respondents who agree with the statement that
the political parties are only interested in people’s votes, not in their opinions,
has increased from 37 percent in 1968 to 68 percent in 1991 (see Figure 2).
This collapse in confidence has been just about as great in all social groups
- among young and old, white-collar and blue-collar workers, in cities and
rural areas. All of the parties have suffered from the distrust, but the Social
Democratic Party, closely related to the corporatist state, has experienced
the most spectacular change; it was put out of office in 1991. This distrust
of politicians is not equally pronounced in all comparable countries. Sweden
is one of the countries in which the lack of faith has been greatest and most
regular (Gilljam & Holmberg 1990: 113-122, 307-308).
The Swedish people distrust interest organizations the most. Of eleven
76
zyxwvu
zyxwvutsrq
social institutions studied, none is less trusted than the trade unions; it is
worth noting that the government and parliament come only one or two
notches higher. The institutions citizens have most faith in are hospitals and
the police. Over time, the collapse has been especially great for the political
institutions (Elliot 1991).
According to the first hypothesis of this study, it is to be expected that
when small organizations spring up and articulate their particular interests,
social conflict will be the net effect. According to the second hypothesis, the
results of encompassing organizations will be the direct opposite: they are
so large that they have an interest in the totality and contribute to social
integration. Both of these hypotheses are confirmed by the material examined. Developments during the period of decline, when the organizations
remained large but the level of conflict rose once again, were left unexplained, however. My conclusion is that in the seventies the organizations
became overloaded. By that I mean that organizations were not only large
(like encompassing,organizations during the period of maturity) but also
considerably more ambitious far beyond their traditional tasks and made, as
we have seen, a number of promises to their members that they were later
unable to keep. Disappointment followed after the offensive in industrial
policy:
zyxwvut
zyxwvut
zyx
.
’
Organizations
Period
Social effec&
Small
Rise
Conflicts
Encompassing
Maturity
Good relations
Overloaded
Decline
Conflicts
Unionism has thus become an impossible business (Micheletti 1990). Inflation
and high taxes create demands for high wages. On the other hand lack of
growth forces the government to cut the budget. Cooperation with the state
then implies a restrictive wage policy - at the same time capital gains if the
policy is successful!
This crisis for Swedish corporatism was brought to a climax in February
1990 when the leadership of LO went so far down the anti-union path as to
be persuaded to support the government’s proposal to forbid strikes, a
decision that led to a storm of outraged protest. This event shows with
extreme clarity how an overloaded organization had become played out in
its traditional role of delivering loyalty and trust in the political system in
exchange for constant improvements for its members. During the period of
decline both improvements and trust have disappeared.
zyxwv
zyxwvutsrq
77
Acknowledgements
This paper summarizes a research project ‘Organized Interests and the Public
Good’ at the Department of Government, Uppsala University. It has been
translated by Donald Lavery. The project was sponsored by The Bank of
Sweden Tercentenary Fund. I am grateful to comments from my associates in
the project Christina Bergquist, Jorgen Hermansson, Bo Rothstein, Anders
Westholm and PerOla dberg.
zyxwvuts
zyxwvu
zyxwvu
zyxwvut
References
Baglioni. G. & Crouch, C., eds. (1990). European industrial relations: The challenge offlexibilify. London: Sage.
Bengtsson, U. (1975). PersonsammanGlttning i utredningsvbndet 1%-1973 [Composition of
committees 1968-19731. Department of Political Science, University of Gothenburg. Mimeo.
Birnbaum, P. (1988). States and collective action: The European uperience. Cambridge: Cambridge University Press.
Bowen, R.H. (1947). German theories of the corporative state with special reference to the period
1870-1919. New YorklLondon: McGraw-Hill.
Brunetta, R. & Dell-Aringa, C., eds. Labour relations and economic performance. London:
Macmillan.
Burke. E. ([1790] 1969). Rejkctions on rhe revolution in France. Ed. Conor Cruise O’Brien.
Penguin.
Cox, A. & O’Sullivan, N., eds. (1988). The corporate state: Corporatism and the state tradition
in Western Europe. Aldershot: Edward Elgar.
Crouch, C. & Pizzorno, A., eds. (1978). The resurgence of class conflict in Western Europe since
1%8, 2 vols. London: Macmillan.
Dahl, R.A. (1956). A preface to democratic theory. ChicagoLondon: The University of Chicago
Press.
Dahrendod, R. (1989). Tertium Non Datur: A comment on the Andrew Shonficld lectures,
Government and Opposition 24: 131-141.
Department of Political Science, University of Gothenburg. Data 1957-1991.
Elliot, M. (1991). Fd’rtroendekapital ifitskingring [Capital of trust in misappropriation]. Svensk
informations mediecenter.
Elvander. N. (1969). Intresseorganism’onerna i dugens Sverige [Interest organizations in Sweden
of today]. Lund: Glcenrp.
Elvander, N. (1988). Den svenska modcllen m e Swedish model]. Stockholm: Allmlnna farlaget.
De Geer, H. (1989). LvUnstervind och hdgervilg [In the wind of the left and the wave of the
right]. Stockholm: All m w a fbrlaget.
Gidlund, J., Engberg, J., Hallin. U.-B. & LindstrOm. A. (1982). Folkrllrelser och kommunrrlpolitik [Popular movements and local politics]. Ministry of Municipalities.
Gilljam, M. & Holmberg, S. (1990). Rdlt, bl&, grunt: En bok om 1988 drs riksdngsvd [Red,
blue, green: A book about the election of 19881. Stockholm: Bonnien.
Gustafsson, A. (1988). Kommwrol sjdlvsryrelse [Local self-government]. Stockholm: Liber.
Hadenius, A. (1976). Facklig organisarionsutvsckling [The development of trade unions]. Stockholm: RabCn and Sjagren.
Hadenius, A. (1978). Ambetsverkens styreker [The boards of authorities], Statrvetmkaplig
Tidrknift 81: 19-32.
zyxwvu
78
zyxwvut
zyxwvuts
zyxwvutsr
zyxwv
zyx
zyxwv
zyxwvu
Hadenius, A. (1983). Mcdbesthmanderefonncn [The reform of co-determination]. Stockholm:
Almqvist and Wiksell.
Henning, R. (1980). Partiema och Stdlverk 80 [The parties and 'Steelplant 80'1. Stockholm:
Liber.
Hesslen. G. (1927). Det svenska bmmitt4vUsendet'intill dr 1905 w e Swedish committee system
until 19051. Uppsala.
Hibbs, D.A.,jr. (1978). On the political economy of long run trends in strike activity. British
Journal of Political Science 8: 153-175.
Johansson, A.L.(1989). TillvUxt och Wpvamcrrbetc [Growth and class cooperation]. Stockholm:
Tiden.
Johansson, J. (1992). Detstatliga kommincvlvrcndet [The pubbc system of committees]. Edsbruk:
Akademitryck.
Katzenstein, P.J. (1985). Small states in world markets: Industrialpolicy in Europe. IthacalLondon: Cornell University Press.
Kendall. W. (1975). The labour movement in Europe. London: Allen Lane.
Kendall, W.& Carey, G.W. (1968). The 'intensity' problem and democratic theory, American
Political Science Review 6 2 5-24.
Kjellberg, A. (1983). Facwig organisering i rolv h d e r [Union organization in twelve countries].
Lund: Arkiv.
Korpi, W. (1983). The.democratic class struggle. London: Routledge & Kegan Paul.
Kuznets, S. (1971). Economic growth of notions. Cambridge: The Belknap Press of Harvard
University Press.
Lash, S. & Urry, J. (1987). The end of organized capitalism. Cambridge: Polity Press.
Lehmbruch, G . (1982). Introduction: Neo-corporatism in comparative perspective, in G.
Lehmbruch & P.C. Schmitter (eds.), Patterns of Corporatist Policy-Making. London: Sage.
Lewin, L. (1967). PlanhushAllningsdebatren [The debate on the planned economy]. Stockholm:
Almqvist and Wiksell.
Lewin, L. (1988). Ideology and strategy: A century of Swedish politics. Cambridge: Cambridge
University Press.
Lewin, L. (1992). Samh&llet och de organiserade intressena [Society and organized interests].
Stockholm: Norstedts.
Lijphart. A. & Crepaz. M.M.L. (1991). Corporatism and consensus democracy in eighteen
countries: Conceptual and empirical linkages, British Journal of Political Science 21: 235-
zyxwvuts
zyxwvutsr
256.
Machiavelli. N. ([1513] 1983). The Discourses. Ed. Bernard Crick. London: Penguin Classics.
Meijer, H. (1969). Bureaucracy and policy formulation in Sweden, Scandinavian Political Studies
4: 103-116.
Mtny. Y. (1990). Government and politics in Western Europe: Britain France, Italy, West
Germany. Oxford: Oxford University Press.
Mchclctti. M. (1990). Toward interest inarticulation: A major consequence of corporatism for
interest organizations. Scandinavian Political Studies 13: 255-276.
Molin. B., M h n , L. & StrOmberg, L. -(1975). Offentlig fdrvalbling [Public administration].
Stockholm: Bonniers.
Offe. C. & Wicscnthal. H. (1985). Two logics of collective action: Theoretical notes on social
class and organizational form, in J. Keane (ed.), Claw Offe: Disorganized Capitalism:
Contemporary TrMsfonnorionsof Work and Politics. Cambridge: Polity Press.
Olson, M. (1982). The rise and decline of M ~ ~ O ~ Economic
IS:
growth, stagflation and social
rigidities. New Havedbndon: Yale University Press.
Rzeworski, A. & Wallerstein. M.(1982,). The structure of class contlict in democratic capitalist
societies, American Political Science Review 76: 215-238.
Ross.A.M.& Hartman, P.T. (1960). Changingpatterns of industrial conflict. New York: Wiley.
Rothstein, B. (1992). Labor-market institutions and working-class strength, in S. Steinmo, K.
Thelen & F.Longstreth (eds.). Structuring Politics: Hiswrical Institutionalirm in Comparative
Analysis, Cambridge: Cambridge University Press.
z
zyxwvut
zyxwvu
zyxwvut
79
zyxwvut
zyxw
zyxw
zyxwvu
Rousseau, J.-J. ([1762] 1978). On the social contract. Ed. Roger D. Masters. New York: St.
Martin’s Press.
Schiller, B. (1988). Samarbete eller konflikt? [Cooperation or conflict?]. Stockholm: Arbetsmiljbfonden.
Schmidt, M.G.(1982). Does corporatism matter? Economic crisis, politics and rates of unemployment in capitalist democracies in the 1970s. in G. L-ehmbruch & P.C. Schmitter (eds.),
Patterns of Corporatist Policy-Making. London: Sage.
Schmitter, P.C. (1974). Still the century of corporatism?. Review of Politics 36: 85-131.
Schmitter, P.C. (1981). Interest intermediation and regime governability in contemporary Western Europe and North America, in S.D. Berger (ed.), Organizing Interests in Western
Europe: Plurdkm. Corporaritm, and the Transformotion of Politics. Cambridge: Cambridge
University Press.
Schmitter, P.C. (1989). Corporatism is dead! Long live corporatism, Government and Opposition
24: 54-73.
Schorter, E. & Tilly, C. (1974). Strikes in France 1830-1968. Cambridge: Cambridge University
Press.
Skierlind, D. & Peterson, 0. (1988). Svensk fdrvalmingspolirik [Swedish administrative politics]. Uppsala: Diskurs.
Werpalm. S.A.(1980). Arbetrgivama och Saltsjdbadrpolitiken [The employers and the politics
of Saltsjbbaden]. Stockholm: SAF.
Stigler, G.J. (1971). The theory of economic regulation, Bell Journal of Economics and Management Science 2: 3-21.
Strbmberg. L. & Westerstilhl, J., eds. (1983) De nya kommunerna [The new municipalities].
Stockholm: .Liber.
Swahn, U. (1980). Interest-representation in Swedish law-making 1922-1978: ‘Remiss’ circulation of Swedish government official reports. Paper presented at the Joint Sessions of
Workshops of the European Consortium for Political Research, Florence.
Swedish Institute for Social Research. 1991. Data.
Swedish Official Statistics.
Swedish Organizational Statistics.
’Iherborn. G. (1987). Does corporatism really matter? The economic crisis and issues of political
theory, Journal of Public Policy 7: 259-284.
Westersthl, J. (1945). Svensk fackf6reningsr6reIse [The Swedish trade unions]. Stockholm:
Tiden.
WesterstAhl, J. & Persson, M. (1975). Demokrati och intresserepresentation [Democracy and
interest representation]. Stockholm: Liber.
Wilensky, H.L. (1976). The ‘New Corporatism’, centralization, and the welfare state. London/
Beverly Hills: Sage.
Wilensky, H.L.& Turner, L. (1987). Democratic corporatism and policy linkages: The interdependence of indrcstrial, labor-market, incomes, and social policies in eight countries. Berkeley:
Institute of International Studies.
Williamson, P.J. (1989). Corporatism in perspective: A n introductory guide to corporaht Ilreory.
London: Sage.
Wilson, G.K. (1990). Zntrresr groups. Oxford: Basil Blackwell.
Address for correspondence: Leif Lewin, Department of Government, Skytteanum, Uppsala
University, P.O. Box 514, $751 2O,Uppsala, Sweden
Phone: +46 18 183 412; Fax +46 18 183 409