(Work and Welfare in Europe) Martin Schröder (Auth.) - Integrating Varieties of Capitalism and Welfare State Research - A Unified Typology of Capitalisms (2013, Palgrave Macmillan UK)

Work and Welfare in Europe
Series Editors: Denis Bouget, University of Nantes, France; Jochen Clasen,

University of Edinburgh, UK; Ana Guillén Rodriguez, University of Oviedo,
Spain; Jane Lewis, London School of Economics and Political Science, UK and
Bruno Palier, Sciences-po Paris, France
Titles include:
Sigrid Betzelt and Silke Bothfeld
ACTIVATION AND LABOUR MARKET REFORMS IN EUROPE
Challenges to Social Citizenship
Sonja Drobnic and Ana M. Guillén Rodriguez

WORK–LIFE BALANCE IN EUROPE
The Role of Job Quality
Colette Fagan, Maria González Menèndez and Silvia Gómez Ansón

WOMEN ON CORPORATE BOARDS AND IN TOP MANAGEMENT
European Trends and Policy
Neil Fraser, Rodolfo Gutierrez and Ramon Pena-Cassas

WORKING POVERTY IN EUROPE
Patricia Frericks and Robert Maier

THE LIFE COURSE AND THE ECONOMIC SUSTAINABILITY OF
THE WELFARE STATE
Paolo Graziano, Sophie Jacquot and Bruno Palier

THE EU AND THE DOMESTIC POLITICS OF WELFARE STATE REFORMS
Europa, Europae
Karl Hinrichs and Matteo Jessoula

LABOUR MARKET FLEXIBILITY AND PENSION REFORMS
Flexible Today, Secure Tomorrow?
Trudie Knijn
WORK, FAMILY POLICIES AND TRANSITIONS TO ADULTHOOD
IN EUROPE
Colin Lindsay and Donald Houston

DISABILITY BENEFITS, WELFARE REFORM AND EMPLOYMENT POLICY
Ive Marx and Kenneth Nelson

MINIMUM INCOME PROTECTION IN FLUX
Birgit Pfau-Effinger and Tine Rostgaard

CARE BETWEEN WORK AND WELFARE IN EUROPEAN SOCIETIES
Martin Schröder
INTEGRATING VARIETIES OF CAPITALISM AND WELFARE STATE RESEARCH
A Unified Typology of Capitalisms
Rik van Berkel, Willibrord de Graaf and Tomáš Sirovátka
THE GOVERNANCE OF ACTIVE WELFARE STATES IN EUROPE
Work and Welfare in Europe

Series Standing Order ISBN 978–0–230–28026–7
(outside North America only)
You can receive future titles in this series as they are published by placing a
standing order. Please contact your bookseller or, in case of difﬁculty, write to us
at the address below with your name and address, the title of the series and the
ISBN quoted above.
Customer Services Department, Macmillan Distribution Ltd, Houndmills,
Basingstoke, Hampshire RG21 6XS, England
Integrating Varieties of
Capitalism and Welfare
State Research
Martin Schröder
University of Marburg, Germany
© Martin Schröder 2013
Softcover reprint of the hardcover 1st edition 2013 978-1-137-31029-3
All rights reserved. No reproduction, copy or transmission of this
publication may be made without written permission.
No portion of this publication may be reproduced, copied or transmitted
save with written permission or in accordance with the provisions of the
Copyright, Designs and Patents Act 1988, or under the terms of any licence
permitting limited copying issued by the Copyright Licensing Agency,
Saffron House, 6–10 Kirby Street, London EC1N 8TS.
Any person who does any unauthorized act in relation to this publication
may be liable to criminal prosecution and civil claims for damages.
The author has asserted his right to be identified as the author of this work
in accordance with the Copyright, Designs and Patents Act 1988.
First published 2013 by
PALGRAVE MACMILLAN
Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited,
registered in England, company number 785998, of Houndmills, Basingstoke,
Hampshire RG21 6XS.
Palgrave Macmillan in the US is a division of St Martin’s Press LLC,
175 Fifth Avenue, New York, NY 10010.
Palgrave Macmillan is the global academic imprint of the above companies
and has companies and representatives throughout the world.
Palgrave® and Macmillan® are registered trademarks in the United States,
the United Kingdom, Europe and other countries.
ISBN 978-1-349-45660-4 ISBN 978-1-137-31030-9 (eBook)
DOI 10.1057/9781137310309
This book is printed on paper suitable for recycling and made from fully
managed and sustained forest sources. Logging, pulping and manufacturing
processes are expected to conform to the environmental regulations of the
country of origin.
A catalogue record for this book is available from the British Library.
A catalog record for this book is available from the Library of Congress.
Contents
List of Figures and Tables viii

Acknowledgements x
Introduction 1
1 Welfare State Research and Varieties of Capitalism 5
1.1 Welfare state research 5
1.1.1 Weaknesses of Esping-Andersen’s welfare
regime typology 13
1.2 Varieties of capitalism 15
1.2.1 Complementarity 23
1.2.2 The coordination index 25
1.2.3 Weaknesses of Hall and Soskice’s varieties
of capitalism typology 25
1.3 Why it makes sense to combine nations into regimes 28
2 Empirical Indicators and Existing Typologies 31
2.1 Verifying diversity empirically 31
2.2 Combining the country clusters of
existing typologies 37
2.2.1 Three capitalisms: Schmidt’s typology 38
2.2.2 Four capitalisms: Boyer’s typology 39
2.2.3 Five capitalisms: Amable’s typology 41
2.2.4 Again five capitalisms: Ebbinghaus’ typology 44
2.2.5 Ever more diversity: Crouch and
Streeck’s typology 47
2.3 Dealing with outlier countries 50
2.4 A methodological interlude – how to build typologies 55
3 A Unified Typology of Capitalisms 58
4 How Complementarities Stabilize Three Capitalisms 63
4.1 Welfare regimes and industrial relations 64
4.2 Welfare regimes and corporate governance 68
4.3 Welfare regimes and skills creation 70
4.4 Welfare regimes and employer associations 74
4.5 Varieties of capitalism and decommodification 77
4.6 Varieties of capitalism and retirement policy 81
v
vi Contents
4.7 Varieties of capitalism and power resources 84

4.8 Résumé: Are complementarities the answer? 87
5 Diversity’s Source: Three Policy Styles,
Three Capitalisms 91
5.1 Liberal capitalism 97
5.1.1 The United Kingdom 98
5.1.2 The United States 102
5.1.3 Ireland 106
5.1.4 Canada 108
5.1.5 New Zealand 111
5.1.6 Australia 114
5.1.7 Summing up liberal capitalism 115
5.2 Conservatively coordinated capitalism 118
5.2.1 Germany 120
5.2.2 Austria 122
5.2.3 Switzerland 124
5.2.4 Belgium 126
5.2.5 France 128
5.2.6 The Netherlands 130
5.2.7 Italy 133
5.2.8 Spain 135
5.2.9 Portugal 138
5.2.10 Japan 140
5.2.11 Summing up conservatively coordinated
capitalism 142
5.3 Social democratically coordinated capitalism 143
5.3.1 Sweden 145
5.3.2 Denmark 148
5.3.3 Finland 151
5.3.4 Norway 153
5.3.5 Summing up social democratically
coordinated capitalism 155
5.4 Summary and limitations of the approach 157
6 What Can a Unified Typology Explain? 162
6.1 How a unified typology can explain more
than existing typologies 163
6.2 How a unified typology can explain trajectories
of liberalization 168
Summary and Conclusion 172
Contents vii
Notes 176
References 181
Appendix 210
A Data used 210
B Factor loadings 217
Index 221
List of Figures and Tables
Figures
1.1 Taxes and social security contributions for different

average production workers in 2011 8
1.2 Benefit entitlements in the first year of unemployment
in 2007 9
1.3 Gini index of household incomes (total population)
after taxes and transfers in the late 2000s 10
1.4 Employment rates of the working population
(aged 15–64) in 2011 10
1.5 Service sector employment as share of total
employment (civilians only) in 2009 11
1.6 Health and social sector employment as share of
total employment (civilians only) in 2009 (2008
for some countries, 2006 for France) 12
1.7 Rights of works councils in 2010 20
1.8 Level of wage bargaining in 1990 and 2010 21
1.9 Market capitalization of stock-noted companies as
percentage of GDP in 2007 and 2009 22
1.10 OECD indicator of employment protection in 2008 23
1.11 Level and coverage of collective bargaining and degree
of co-determination in 2010 24
2.1 Production systems and welfare states in one grid 34
2.2 Scree plot of principal component analysis 35
2.3 Dendrogram of hierarchical cluster analysis 36
3.1 Typologies and their congruence 59
4.1 Central union power density and welfare liberalism 66
4.2 High-income strata and market capitalization 69
4.3 Employment protection and coordination 73
4.4 Replacement wage and coordination 80
viii
List of Figures and Tables ix
4.5 Pension replacement rate of prior wage and coordination 82

5.1 Agreement: government should provide job and
unemployment compensation 95
6.1 Coordination and employment volume 164
6.2 Female employment as percentage of total employment 165
6.3 Female elite decision-makers as percentage of total 166
6.4 Women in leading management positions as
percentage of total 167
Tables
1.1 Three welfare regimes in comparison 6

1.2 Characteristics of coordinated market economies
and liberal market economies 19
2.1 Correlations between production and welfare regimes 32
2.2 Capitalisms structured into one framework 33
2.3 Regulation theory, welfare state research, varieties
of capitalism – different theories, complementing
explanations 42
2.4 Incoherencies in welfare and production regimes 51
4.1 Correlations between historical industrial relations
and welfare regimes 65
4.2 Correlations between current industrial relations and
welfare regimes 67
4.3 Correlations between employment protection
and coordination 74
4.4 Correlations between employer organization and
welfare state characteristics 76
4.5 Correlations between decommodification and
coordination 79
4.6 Correlations between pension system decommodification
and coordination 83
4.7 Correlations between organized labour and welfare state
characteristics 86
5.1 Three types of capitalism 158
Acknowledgements
I would like to thank Aaron Boesenecker, Jason Beckfield, Peter Hall,

Patrick LeGalès, Tobias Lenz, Wolfgang Streeck, Helmut Voelzkow,
Vivien Schmidt and Kathleen Thelen for their invaluable help with this
book. All remaining errors are the sole responsibility of the author.
x
Introduction
Even though most scholars acknowledge that welfare states interact pro-
foundly with the organization of the economy, there is no typology that
combines the analysis of welfare states and production systems. Instead,
‘neither students of the welfare state nor students of political economy
viewed the welfare state as an integral part of national economies’
(Pierson 2000: 793; for a similar argument, cf. Deeg and Jackson 2006:
31; Soskice 2007: 92). Since about 10 years, efforts exist to connect
welfare regimes1 and modes of economic coordination to make up for
this (Ebbinghaus and Manow 2001a; also cf. Huber and Stephens 2001a;
Pierson 2000). But existing attempts stop short of offering a typol-
ogy that indeed unifies the study of production and welfare regimes
(cf. the attempts in Cusack, Iversen and Soskice 2007; Iversen and
Soskice 2006). Thus, whereas most scholars agree that welfare and pro-
duction regimes ‘tend to coincide’ (Esping-Andersen 1990: 159), there
has not yet been an attempt to integrate welfare and production regimes
into one unified typology, which captures the diversity of production
and distribution in one framework. Therefore, this book develops such a
typology, combining the study of production systems with the study of
welfare states. To do this, it combines the two (arguably) most common
typologies of advanced capitalist countries, ‘varieties of capitalism’2 (see
Hall and Soskice 2001a) and the ‘three worlds of welfare’ (see Esping-
Andersen 1990, 1999).
It might seem surprising that these two typologies can be integrated
in one. Whereas varieties of capitalism analyses how firms coordinate
with their environment, welfare state research asks how citizens are
confronted with incentives and duties in the distribution of welfare.
Thus, on the micro level, the approaches analyse different depend-
ent variables: varieties of capitalism analyses firms; Esping-Andersen’s
1
2 Integrating Welfare and Production Typologies
welfare typology analyses what benefits and duties individuals have in

their welfare regime. Both use their differing micro-level analysis to
explain different macro-level indicators (performance, welfare budgets,
industrial specialization). Yet, these seemingly unrelated approaches
arrive at broadly similar country groupings.3 Namely, countries that
rely on the market for the governance of their production system have
a liberal welfare state; countries that coordinate their economy usually
have either a ‘conservative’ or ‘social democratic’ welfare state.4
These overlaps allow combining varieties of capitalism and welfare
state research, which is the aim of this book. It thus shows that welfare
states and production systems are not unrelated to each other, but pro-
vide a common picture of capitalist diversity. In showing this, I aim to
use the largest possible number of OECD countries. However, of the
34 OECD countries for which suitable data is available, the regime
type of 14 countries has not (yet) been determined (Chile, the Czech
Republic, Estonia, Greece, Hungary, Iceland, Israel, Korea, Luxembourg,
Mexico, Poland, Slovenia, the Slovak Republic and Turkey). Combining
existing typologies, I can only use countries that have already been
categorized. This leaves the following 20 countries: Australia, Austria,
Belgium, Canada, Denmark, Finland, France, Germany, Ireland, Italy,
Japan, the Netherlands, New Zealand, Norway, Portugal, Spain, Sweden,
Switzerland, the United Kingdom and the United States.5
Why is a unified typology an improvement over existing attempts to
classify welfare and production regimes, on which it builds? This book
is the first attempt to show how conservative and social democratic
welfare states can only survive within a coordinated market economy
and vice versa, while liberal welfare states can only survive when cou-
pled with liberal market economies – and vice versa. Also, this book is
the first comprehensive attempt that shows how a conservative ‘policy
style’ structures how continental European countries and Japan organ-
ize their production and welfare system. So it is not coincidental that
coordinated market economies evolved next to conservative welfare
states, giving rise to what I call a ‘conservatively coordinated’ variety
of capitalism, which tries to stabilize society through the welfare state
and the production system. In turn, nationwide solidarity meant that
Scandinavian social democratic welfare states evolved alongside nation-
ally coordinated production systems, giving rise to what I call ‘social
democratically coordinated capitalism’. The anti-statism of Britain
and its former colonies led to a policy style of liberalism that not only
pervades the welfare state, but also the production system of English-
speaking countries. It is this notion of a policy style, which aligns the
Introduction 3
welfare state and the production system to a common principle, that

provides an improvement on the existing literature.
Apart from Esping-Andersen’s (1990) own work and the original 2001
typology by Hall and Soskice, existing attempts of typologizing lay the
foundation for a unified typology of capitalisms (cf. Ebbinghaus and
Manow 2001a; Amable 2003; Boyer 2005; Iversen and Soskice 2006).
But these existing attempts stop short of producing a typology in which
all coordinated countries are different from liberal countries, while at
the same time coordinated countries can be split into two subtypes: one
centred on upholding social stability and one centred on nationwide
solidarity. This gives rise to a group of English-speaking liberal coun-
tries, which can be distinguished from a group of Scandinavian social
democratically coordinated and a group of conservatively coordinated
countries (continental Europe plus Japan).
While the first chapters of this book show why a unified typology of
welfare and production regimes is possible, the last chapters show the
added explanatory value of such a unified typology. Therefore, Chapter 1
introduces the typologies that this book combines, their relationship to
each other and the problems that beset them. Thus, section 1.1 introduces
welfare state research and section 1.2 introduces varieties of capitalism.
I show why both typologies, by and large, still give an accurate picture
of capitalist diversity, in spite of problems in how they classify countries.
Section 1.3 deals with popular criticism against the two approaches and
how this has to be dealt with when combining them. It exposes the
problem. Chapter 2 gives quantitative empirical evidence why it makes
sense to combine the two typologies. It does so by using a database (see
Appendix A), which quantifies welfare and production arrangements
using a large number of variables. Section 2.1 performs a principal factor
and a hierarchical cluster analysis on this data. Section 2.2 shows that
the country groupings of other established typologies can gainfully be
combined with varieties of capitalism and welfare state research. Section
2.3 shows that a number of countries fit uneasily in any typology, and
even more so in attempts to combine them. This leads to some reflec-
tions on how a unified typology of capitalisms can be constructed and
what its limits must be (section 2.4). Chapter 3 builds on this, present-
ing the outlines of a unified typology by distinguishing between lib-
eral, social democratically coordinated and conservatively coordinated
countries.
It is one thing to state that welfare and production systems coincide;
it is a different task to explain why. The combination of varieties of capi-
talism and welfare state research opens our eyes as to why welfare and
4 Integrating Welfare and Production Typologies
production systems coincide. Trying to explain this, Chapter 4 shows

why the proposed typology makes sense by presenting a critical outline
of complementarities between certain production and welfare regimes.
I argue that coordinated production systems are complementary to social
democratic and conservative welfare arrangements, while liberal produc-
tion systems flourish in liberal welfare arrangements. Chapter 5 looks
into each country in depth to explain why a typology makes sense in
spite of the problems connected to it. It shows that welfare and produc-
tion systems historically moved ‘in tandem’. When countries reform their
production system in a liberal direction, they tend to liberalize their wel-
fare state as well, and vice versa. This shows how welfare and production
system reforms proceed in a liberal variant of capitalism (section 5.1), in
a conservatively coordinated type of capitalism, which is mainly centred
on upholding social stability (section 5.2), and in a social democratically
coordinated capitalist variety, in which solidarity plays a central role
(section 5.3). This chapter thus argues that the development of welfare
and production institutions followed similar logics within countries and
clusters of countries. It argues that countries have a congruent policy style,
which they apply to the regulation of their welfare and of their produc-
tion system, aligning both to a common principle within countries and
families of countries. More to the point, if societies embrace the market
for the governance of their welfare system, they also embrace it for the
governance of their production system – and vice versa. Introducing this
conception, according to which an underlying regulatory style binds cer-
tain welfare and production systems together, I introduce a sociological
conception of how common values explain political economies. Given
the problematic standing of typologies between real cases and ideal
types,6 section 5.4 summarizes to what degree we can see similarity in the
different countries. Chapter 6 shows how a welfare typology can explain
more when used together with varieties of capitalism and vice versa.
Notably, I show how a unified typology can explain varieties of female
labour participation and of liberalization better than existing typologies
used in isolation.
The book concludes by proposing avenues for future research that
open up due to a combination of varieties of capitalism and welfare
state research, showing why it makes sense to combine the two most
popular approaches to explain capitalist diversity into one. Thus, this
book provides a new tool for future research: a unified typology to ana-
lyse similarities and differences in capitalist countries through the com-
bined lens of which so far are the most popular, but as of yet isolated
approaches to analysing capitalist diversity.
1
Welfare State Research and
Varieties of Capitalism
This chapter introduces Esping-Andersen’s welfare and Hall and

Soskice’s production typologies. Readers who are well accustomed to
them can omit this chapter.
1.1 Welfare state research
Drawing on Esping-Andersen (1999: 33), one can define welfare state

research as that field of study that is devoted to analysing who man-
ages social risks (mainly divided between: the state, the family or the
market) and how these risks are managed. In contrast to varieties of
capitalism, welfare state research focuses more on how economic value
is (re)distributed than on how it is generated.
As distributional outcomes of market transactions can be unacceptable
for the embedding society, it can create social rights and duties to change
these outcomes. It is these rules, their emergence and their effects, which
welfare state research examines (Esping-Andersen 1990: 11).1 Differently
from varieties of capitalism, Esping-Andersen’s typology explains capital-
ist diversity out of class conflicts, with the key question being how the
working class formed a coalition with different social groups to develop
different welfare states (Esping-Andersen, 1990: 18). In Scandinavia, the
working class formed a coalition with small, capital-intensive and politi-
cally well-organized farmers and then took the middle class on board by
providing high-quality social services and public jobs. In liberal coun-
tries, the middle class could largely care for itself in the market, so the
welfare state became residual, caring only for the poor. In continental
Europe, labour-intensive large-scale farmers were in a coalition with
conservatives that isolated the labour movement. A state-administered
system of welfare benefits tied them to the state by protecting them
5
6 Integrating Varieties of Capitalism and Welfare State Research
Table 1.1 Three welfare regimes in comparison
Liberal Social Conservative

democratic
Role of – Family Marginal Marginal Central

– Market Central Marginal Marginal
– State Marginal Central Subsidiary
Dominant mode of solidarity Individual Universal Kinship
Dominant locus of solidarity Market State Family
Degree of decommodification Minimal Maximal High (for
breadwinner)
Underlying dominant value Liberty Equality Stability
Paramount examples USA Sweden Germany, Italy
Source: Esping-Andersen (1999: 85).
against social risks. These different class struggles led to three welfare
regimes: a ‘liberal’, a ‘conservative’ and a ‘social democratic’ one.2 The
table above sums up their main characteristics.
In the liberal regime, markets are the central means of allocation.
Welfare programs are largely limited to means-tested poor relief (cf.
Bradshaw et al. 1997), distinguishing between one’s ability and one’s
willingness to work. Thus, only the undeserved poor (those that are not to
blame for their situation) constitute the deserving poor (deserving social
assistance). As the bulk of the pensions, unemployment and health
system is therefore based on private provision, decommodification3 is
low – the state does not protect against the market. The United States
is seen as the paramount case for this regime type. Canada, Australia,
Ireland, New Zealand and the United Kingdom are more ambiguous but
fall more in the liberal category than in any other.4
The social democratic regime can be seen as the conceptually opposed
model, as it is ‘committed to comprehensive risk-coverage, generous
benefit-levels, and egalitarianism’ (Esping-Andersen 1999: 78). The
model nonetheless embarked from the same starting point as the liberal
one. It preferred a ‘Beveridge’ system of flat-rate citizen entitlements to
a ‘Bismarckian’ system, which calculates benefits based on prior earn-
ings. However, whereas the liberal system progressively moved towards
residualism, limiting benefits as much as possible, the social demo-
cratic countries enlarged their welfare programs towards universalism,
taking in the entire population. Accordingly, these countries decom-
modified workers; they reduced their market dependency to maximize
Welfare State Research and Varieties of Capitalism 7
social equality (Esping-Andersen 1999: 78f.). Their welfare state is

encompassing, mainly because social benefits are based on citizenship
rather than absolute need or prior contributions. Thus, public unem-
ployment, pension, and health schemes cater to everyone, rendering
private provision unnecessary, even for the most affluent. In addition,
a large public sector generates a high employment rate and a generous
public infrastructure of childcare, health and other services, usually
provided free of immediate charge. On the downside, the costs of these
services entail a high tax burden. Yet, social spending of these states is
not the highest in the OECD (cf. OECD 2006: 181). Instead, it is not so
much by the overall level of social spending that they distinguish them-
selves, but by the large dispersion of social benefits towards all social
strata. Sweden and Denmark are the purest embodiments of this type of
regime, Finland and Denmark resemble it closely.
In the conservative regime, the welfare state is neither residual nor
encompassing. It is based on the concept of insurance, as opposed to
basing benefits on means testing or citizenship. By contributing to
social-security schemes, citizens acquire entitlements in case of sickness,
unemployment or retirement. Neither provision of high-quality public
services nor means-tested poor relief unsettles the social stratification
of these countries (both concerning upward and downward mobility).
Instead, by relating benefits to prior contributions, ‘welfare insurances’
stabilize the existing social stratification. Germany is the paramount
case of this regime type, whereas Austria, Belgium, France, Italy, Japan,5
the Netherlands,6 Portugal and Spain resemble it closely.
Esping-Andersen arrives at this threefold typology by relying on
quantitative data. He sees countries with the highest scores on popula-
tion coverage and homogeneity of social programs as social democratic.
Those with the highest scores on private contribution to health and
retirement schemes and the highest degree of means-tested welfare
programs are liberal. Conservative countries score highest on the seg-
mentation of welfare programs along occupational status lines. Esping-
Andersen (cf. 1990: 69ff. as opposed to 1999) later somewhat refined and
rearranged countries. To illustrate differences between the countries and
the regime they belong to, the following chart shows wage deductions
of three average production workers. The leading bar represents levies for
an average worker with a wife, two children and an average wage.
The liberal countries, apart from Great Britain, are all on that half of
the countries that levy the lowest financial charges on wages. Under
those countries that levy the highest charges, one finds social demo-
cratic and conservative ones, with none of the two regime types having
One-earner married couple at 100% of average earnings, 2 children

Two-earner married couple, one at 100% of average earnings and
the other at 67%, 2 children
Single person at 100% of average earnings, no child
45
40
35
30
25
20
15
10
0
er s
Ire nd
w Sp d
a n
rtu d
na l
Ja da
rm n
st y
d ra a
ng e
m
e y
Ne No den
er ay
Fi nds
lg d
Au ium
nm ia
k
Ca ga
Sw Ital
ar
itz ate
Au an
lan
Ze ai
Po lan
Ge pa
ite F rali
Ki nc
Be n
De str
do
th rw
nla
la
la
Sw d St
ite
Ne
Un
Un
Figure 1.1 Taxes and social security contributions for different average
production workers in 2011
Source: OECD (data from 2012). URL: http://stats.oecd.org.
clearly more or less heavy deductions than the other. The same average
production worker who has to give up a varying share of his wage would
probably be happy to hear that he also gets something back. The fol-
lowing chart draws out average replacement rates for the first year of
unemployment of a worker aged 40 with an uninterrupted employment
record before unemployment, averaged for four different stylised family
situations (single and one-earner couple, with and without children)
as well as for two earning levels (67% and 100% of average full-time
wages).
Again it is the liberal countries, together with Italy and Japan, which
replace wages the least in case of unemployment. In short, it is these
90
80
70
60
50
40
30
20
10
0
Ki tes
Ze ly
Au nd
Ire n
Ca nd
Fi a
Au d
ia
Be y
Sw m
Fr n
De ce
k
th ain
No ds
Po ay
itz gal
d
an
ar
ali
pa
d
n
lan
Ita
str
do
an
lan
rw
ala
la
na
nla
ed
Ne Sp
Sw tu
a
lgi
nm
str
rm
Ja
er
ite St
ng
r
er
Ge
Un ited
w
d
Ne
Un
Figure 1.2 Benefit entitlements in the first year of unemployment in 2007

Source: OECD (2009: 76). URL: http://dx.doi.org/10.1787/706364844714.
countries which tend to collect the least in the form of taxes and social
security contributions, but which also give the least in terms of unem-
ployment benefits. It is conceivable that these welfare state differences,
some favouring redistribution more than others, go along with different
income structures. The chart on the next page illustrates that this is the
case. The bars show the Gini index of household incomes after taxes
and transfers; thus they show how unequally income is distributed
between households after these have paid for whatever the welfare state
costs and received whatever it hands out.
The social democratic welfare states stand out as having the most egal-
itarian income distribution. The liberal welfare states are on the other
side of the figure, they are most unequal, together with Portugal, Italy
and Japan. Ireland is an exception to the liberal regime; it is more equal
than many conservative countries. Most of the conservative countries are
between the two extremes of egalitarian social democratic welfare states
and inegalitarian liberal ones. As different welfare regimes exert system-
atically different charges on employment, while also creating employ-
ment in the public sector, they also generate different employment rates.
Figure 1.4, which is also on the next page, illustrates this, showing what
percentage of all that are of working age are in employment.
Countries with the lowest employment rates are conservative, together
with Ireland, meaning they use their labour pool the least. For a liberal
10
0.4
0.38
0.36
0.34
0.32
0.3
0.28
0.26
0.24
0.22
0.2
No rk
Fi n
Be nd
Sw ay
um
Fr ia
ce
nd
Ge nds
itz ny
d
ain
da
Ze n
Au nd
ng ly
P om
St l
es
a
e
lan
pa
ali
Ita
str
ite ug
a
rw
an
Sw rma
at
ed
nla
Ne ela
na
ala
Sp
nm
lgi
d
str
la
Ne Ja
Au
er
Un ort
Ca
er
r
De
Ki
d
th
d
ite
Un
Figure 1.3 Gini index of household incomes (total population) after taxes and
transfers in the late 2000s
85
80
75
70
65
60
55
50
ly
Ire n
Be d
um
ce
St l
Fi s
d
Ki an
Ca m
Au a
ia
Ze y
Au nd
De alia
k
en
No s
y
d
a
e
an
ar
a
ai
lan
lan
Ita
ite ug
str
do
an
at
lan
Sw rw
nla
na
ala
Ne wed
Sp
lgi
nm
str
Ne erm
Ja
Un ort
er
ng
Fr
er
itz
S
P
th
G
w
d
ite
Un
Figure 1.4 Employment rates of the working population (aged 15–64) in 2011
Source: OECD (data from 2012). DOI: 10.1787/emp-table-2012-1-en.
welfare state, the US also has a very low employment rate. Contrary
to this, the three social democratic countries of Norway, Sweden and
Denmark have the second, fourth and fifth highest employment rates.
Switzerland has the highest employment rate and the Netherlands have
the third highest, but this is mainly because the Netherlands has a lot
of part-time employment. Liberal countries have intermediate employ-
ment rates. However, not only is the rate of employment structured
through welfare regimes; employment also takes place in different sec-
tors in different welfare regimes. The conservative countries have few
service-sector jobs, as their welfare states tax these price-sensitive jobs
heavily, financing costly transfers, instead of public service sector jobs,
so that overall employment in the (public and private) service sector is
low. Figure 1.5 shows this, plotting what percentage of the entire work-
force is employed in the service sector.
While conservative countries (together with Finland, New Zealand
and Ireland) have the lowest proportion of their workforce in the
service sector, liberal and social democratic welfare states (together
with the Netherlands and France) have the largest service sectors.
While both social democratic and conservative welfare states are costly
and thereby make work more expensive, depressing work in the cost-
sensitive service sector, the charges of the social democratic welfare
85
80
75
70
65
60
55
50
l
ly
Ge pan
Au y
ia
ain
Ze nd
nd
itz nd
Be nd
Au um
No lia
Ca ay
Fr a
Sw ce
Un De den
Ki ark
th om
St s
es
ga
an
ite and
d
Ita
str
a
rw
an
at
la
ala
Sw rela
la
na
tu
Sp
lgi
nm
Ne ngd
str
rm
Ja
e
Ne Fin
er
r
Un erl
Po
d
w
d
ite
Figure 1.5 Service sector employment as share of total employment (civilians

only) in 2009
states do not damage service sector employment as much, because they

finance jobs in the public health and social service sectors, as Figure
1.6 shows.
The social democratic welfare states (together with the Netherlands)
finance extensive employment in health and social services. This makes
them different from liberal welfare states, which have no resources to
finance large-scale employment in social sectors. The extensive employ-
ment that they finance in social services also distinguish the social dem-
ocratic welfare states from conservative ones, which use their revenue to
finance social transfers, instead of public services. So while conservative
welfare regimes, especially the Mediterranean countries, have to live
with low employment rates, social democratic countries have higher
employment due to their large public sector; liberal countries have
higher employment rates due to private sector employment.
For the purpose of this book, these are the main differences of wel-
fare regimes. The classification of welfare states into three regime types
had a tremendous impact on the social sciences. As Wincott (2001:
409) argues, ‘more than any other scholar, Gøsta Esping-Andersen has
influenced the way in which we view the (comparative) landscape of
the welfare state’. But as the preceding figures show, some institutions
of some countries diverge from what would be typical for their regime
25
20
15
10
05
00
Sp l
ain
ly
Au e
ia
Ze n
Au nd
C lia
itz da
Ge and
y
d
m
Be s
Fi m
S nd
er n
De ds
ay
No rk
ga
an
e
c
pa
lan
e
Ita
str
ite do
a
a
an
at
lan
rw
ala
Sw ana
nla
Ne wed
rtu
lgi
nm
str
rm
Ne Ja
l
er
ite Ire
St
U n ng
Fr
Po
Ki
th
w
d
Un
Figure 1.6 Health and social sector employment as share of total employment
(civilians only) in 2009 (2008 for some countries, 2006 for France)
Source: OECD (data from 2011). URL: http://dx.doi.org/10.1787/888932524032.
type.7 This provoked criticism again Esping-Andersen’s typology. The

following section deals with this criticism, to discuss whether Esping-
Andersen’s typology is still a useful way of classifying countries.
1.1.1 Weaknesses of Esping-Andersen’s welfare regime typology

Castles and Mitchell (1993) and Castles (1996) voiced a popular criticism
against the three regime types. They argued Australia and New Zealand
should be seen as ‘wage-earner’ welfare states, constituting a fourth
regime. The labour courts of Australia and New Zealand decommodify
workers and provide welfare guarantees via the wage arbitration system,
from which a relatively egalitarian wage structure resulted. In this sense,
the labour courts of these countries were functionally equivalent to
conservative and social democratic welfare states. However, as Castles
(1996) himself and Castles, Gerritsen and Vowles (1996) mention, these
arrangements have been cut back significantly (also cf. Menz 2005b).
In addition, a higher unemployment rate made the primary income
distribution, through which these countries achieved a relative degree
of egalitarianism, less significant. Therefore, the welfare state structure
of these countries, which is liberal, gains importance. So by now, these
countries actually have a very inegalitarian income distribution (cf.
Figure 1.3). Echoing these arguments, Esping-Andersen (1999: 89f.)
refused to add an additional fourth ‘Antipodean’ welfare regime to his
typology.
Ferrera (1996), Leibfried (1992) and Lessenich (1994) demanded that
the Mediterranean welfare states should form a fourth category. These
Southern European countries have a distinctive position due to high
employment protection, which they combine with rudimentary welfare
benefits. Their low welfare benefits make family solidarity important.
Lastly, scholars also mentioned that Japan can be seen as a distinct wel-
fare regime, mainly due to its underdeveloped welfare state (cf. Jones
1993; Pontusson 2005b). Indeed, Esping-Andersen (1999: 91) claims
Japan can be seen as having a residual welfare state. However, Japanese
labour market arrangements and, to a lesser degree, family ties, act as
a functional equivalent to a strong welfare state. Even though Esping-
Andersen saw Japan to be a liberal welfare state in 1990, he argued
that it belonged to the conservative group in 1999 (Esping-Andersen
1999: 90f.). He claims that Japan’s religious values are close to the
‘catholic subsidiarity principle’ (Esping-Andersen 1999: 92; for a discus-
sion of this, also cf. Hamilton and Woolsey 1988: 20ff.). Secondly, he
claims that a corporatist welfare system exists in Japan. This is because
‘kinship’ and ‘family’, which are the dominant locus of solidarity in
Japan, also constitute the conservative regime (see Table 1.1). Southern
European welfare states and Japan can therefore be seen as a subvariety
of the conservative welfare regime. On these grounds, Esping-Andersen
(1999: 93ff.; also cf. Bernard and Saint-Arnaud 2003) refused to split up
the group of conservative countries.
Esping-Andersen’s typology explains why members of a society have
social rights and how these rights contribute to (or hinder) economic
development. His typology sums up the ‘deal’ a society offers its indi-
viduals concerning rights and duties in the form of social entitlements
on the one side and social obligations on the other. This is why, even
though there is important and well taken criticism against Esping-
Andersen’s typology, I share the view that ‘as an organizing principle
for comparative studies of welfare states the typologies have proven to
be a very robust and convincing tool’ (cf. Arts and Gelissen 2002: 155).
Consequently, studies stress that the stability of welfare regimes has not
been eroded (cf. de Beer, Vrooman and Wildeboer 2001; Bernard and
Saint-Arnaud 2003; Scharpf and Schmidt 2000a; b; Starke, Obinger and
Castles 2008). So even though Esping-Andersen’s typology has some-
what come of age, it is still the gold standard:
A case can be made for extending the number of welfare state regimes,
perhaps to four or even five, with a Mediterranean grouping the most
consistently suggested extension. [However,] extending the typology
to five or more models would sacrifice the explanatory parsimony of
his [Esping-Andersen’s] original thesis and that being so one might
as well return to individual country comparisons. A principal value
of Esping-Andersen’s three ideal-types of welfare regimes is that it
provides abstract models, so that deviations from the ideal types can
be noted and explained.
(Arts and Gelissen 2010: 577, 581)
There are also good theoretical reasons to use Esping-Andersen’s three-

fold typology. Notably, with his division into three welfare regimes,
Esping-Andersen captures how countries can either use the market or
diverge from it by a) conservatively upholding social stability or by
b) organizing nationwide solidarity. The three underlying principles
behind these three welfare regimes, liberalism, conservatism and social
solidarity, can actually be seen as the three most general principles of
organizing society in the Western world, for example epitomized by the
dominant separation of political parties into liberals, conservatives and
social democrats. Either societies use the market to organize themselves,
or they circumvent markets, be it by stressing social solidarity, be it

by stressing social stability. Either you want society to stay as it is, or,
if not, you can go in a more individualistic or alternatively a more
egalitarian direction. One may well argue that this distinction between
ascriptivism, individualism and egalitarianism, as others have called it,
exhausts the universe of social justice views that a society can follow
(cf. the typology in Douglas 1982). In this sense, a valid argument can
be made that Esping-Andersen’s threefold typology actually exhausts
the known combinations of social ordering principles that societies can
pursue. Thus, the southern European welfare regimes can be described
as even more conservative versions of the conservative regime type,
with their extreme emphasis on social stability. The Antipodean coun-
tries can be described as a liberal regime, with social democratic traits.
Therefore, even exceptions to Esping-Andersen’s threefold typology can
be described within its framework, which is a powerful theoretical rea-
son to retain it. I will therefore remain with Esping-Andersen’s original
three regime types. However, I will embrace the notion that countries
not only follow one regime but embody features of different regimes to
varying degrees.
1.2 Varieties of capitalism
What is specific, one might even say revolutionary, about Peter Hall and
David Soskice’s (2001a) typology is that it places the organization of
firms at the centre of analysis. This is also what contrasts their typology
to Esping-Andersen’s.
In contrast to previous conceptions of capitalist diversity – which

placed the constitution of welfare regimes or predominant forms of
labour organization in the analytical foreground [see Olson, 1965;
Goldthorpe, 1984; Esping-Andersen, 1990], utilizing measures such
as degree of labour commodification or union density in the genera-
tion of cross-sectional taxonomies of capitalist systems – the varieties
school privileged the organization and regulation of production in
its classificatory and explanatory schema.
(Peck and Theodore 2007: 737f.)
Hall and Soskice draw heavily on the neo-corporatism literature,

which – grossly simplified – postulates a U-shaped curve, linking the
degree of organization of an economy to its efficiency (cf. Crouch 1993:
12; Lehmbruch and Schmitter 1979, 1982; Olson 1982; Calmfors and
Driffill 1988). With this logic, the typology is often seen as being rooted
in a functionalistic explanation of institutions: these are upheld as long
as they support the competitive advantage of firms. In turn, what firms
are interested in is either a completely ‘coordinated’ or a completely
‘liberal’ institutional environment. Hall and Soskice (2001b) look at
five institutional areas to distinguish between economic liberalism and
coordination: 1) industrial relations – the setting of wages and work-
ing conditions, which can either be organized through liberal market
principles or negotiated and in this sense coordinated, 2) vocational
training – which can also be structured according to liberal market
principles or organized through the state or associations, 3) corporate
governance and financing – relations between firms and their investors,
which can be governed through short-term stock investments or strate-
gic long-term stakeholders, 4) inter-firm relations, which can again be
governed through market mechanisms or strategic cooperation between
firms and 5) relations of firms with their employees, which can again
be based on short-term calculations of economic advantage on each
side or by strategic, long-term cooperation. In distinguishing these five
fields, Hall and Soskice establish a dualistic classification of production
systems, which they name liberal and coordinated market economies.
An important starting point of varieties of capitalism was the observa-
tion that countries with what Hall and Soskice call coordinated institu-
tions show superior performance in industries that are mainly based on
improving and upgrading established technology, while liberal countries
show comparative advantage in emerging industries, in which innova-
tion is not linked to prior product development. Companies that con-
tinue improving existing products require a workforce which must be
able to perform complicated tasks autonomously with little managerial
control. As crucial resources such as knowledge, expertise and skills are
concentrated in the hands of employees, relations between companies
and their workforce often take the form of a game-theoretic – and there-
fore fragile – cooperation. Firms invest in the skills of their employees and
guarantee stable employment. In turn, workers refrain from using their
high skills to ‘hold up’ management. This cooperation between workers
and their firms is liable to break down unless other institutions support
it. Notably, cooperation is only possible if companies refrain from poach-
ing qualified workers from other companies, thereby free riding on the
qualification efforts of competitors by using the workers that these other
firms trained. To prevent this, centralized bargaining levels out wages,
while associations or the state monitor firm behaviour (cf. Lauder, Brown
and Ashton 2008: 23). As firms know that workers and competitors have
an incentive to cooperate, associations, networks or the state can diffuse

technology, help build stable relations between companies and support
cooperative research. This sharing and joint development enables com-
panies to produce high-quality products that allow for a higher profit
margin than mass production. Macroeconomically, the system is advan-
tageous as centralized trade unions and employer associations can agree
on wage and price restraints to prevent cost-push inflation.
Another cornerstone of the coordinated production system is the
financial sector, which can support either long-term innovation or
short-term flexibility. Coordinated market economies have bank-based
financing, so companies can get patient capital (Hall and Soskice
2001b: 22). Even though banks expect good returns in the long run,
their patient capital enables firms to sustain employment levels dur-
ing temporary recessions and thus protect their highly qualified core
workforce. Bank-based capital obviously promotes the representation
of bank-representatives in the enterprise’s board of trustees. By being
present there, banks acquire insider information on companies and
can even restructure them if they see their investments endangered.
As financial interlocking between banks and firms reduces the publicly
available share of stock, companies have to be less concerned about
hostile takeovers and can instead devote their attention to product and
process innovation.
Thus, firms in coordinated market economies reap the advantages
from cooperating in situations that resemble a prisoner’s dilemma. To
avoid defection, an external agency must ensure that each side cooper-
ates, which then allows for coordination. This external agency does not
have to be an association, as long as it ensures that ‘firms rely more
heavily on non-market relationships to coordinate their endeavours
with other actors’ (Hall and Soskice 2001b: 8; also compare the descrip-
tion by Hall 2001: 56f.; Kenworthy 2006: 71). If companies trust each
other – for whatever reason – then they can also coordinate in the
absence of strong associations.
The crucial point is that coordination (supported by whoever it might
be), gives rise to strategic cooperation between firms. This is the case
for Austria, Germany, Belgium, the Netherlands, Switzerland, Sweden,
Denmark, Norway, Finland and Japan. Therefore, these countries
are called coordinated market economies. Other countries are more
ambiguous, notably France, Italy, Spain and Portugal. But as much as
the blocking out of short-term market considerations is a blessing for
certain industries, it is a disadvantage for others. Arrangements that
permit companies stable cooperation lead to a lack of flexibility. It is
easy to imagine how an internet start-up, a young biotechnology firm

or an independent media-production firm will be less happy with coor-
dinated arrangements than a long-term oriented car producer. This is
because after, say, a biotechnological product is developed, it must be
put on the market as quickly as possible. After a film is finished, atten-
tion has to be directed to a new project.
Not continuously improving, but rapidly developing new products,
companies in such industries need a dynamic institutional environment
that allows them to continuously reorient their production. Thus, the sta-
bility of institutional arrangements in coordinated market economies is a
disadvantage for companies in very dynamic industries. Instead, firms in
fast-moving industries need flexible labour markets to recruit and dismiss
employees, often based on projects. To act fast, they need a management
that can take decisions unilaterally. They also need flexible forms of
capital, to invest in new projects and they need it quickly, as they have
to take their chance before a competitor does. If someone else is faster,
then a frequent way to ‘diffuse’ technology is to buy competitors or else
be bought by them. This needs well-functioning markets for corporate
governance. Ireland, New Zealand, Australia, Canada, the United States
and the United Kingdom belong to this regime type of liberal market
economies. The heart of this system consists of risk taking; everything
that impedes market competition is thus detrimental to the functioning
of these production systems (for the last two paragraphs, cf. Hall 2002;
Hall and Soskice 2001b; Soskice 1999).8 Taken together, the differences
between coordinated and liberal market economies can be summed up as
two coherent systems, which the following table illustrates.
In distinguishing coordinated and liberal market economies, the vari-
eties of capitalism approach explains why companies have a competitive
advantage in certain industries, while they are disadvantaged in others.
Coordinated arrangements allow for strategic cooperation (by limit-
ing the use of the market). Liberal arrangements allow for allocating
resources efficiently (by promoting the use of the market). To sum up
the difference, whereas companies in liberal market economies are able
to quickly redeploy resources to more dynamic environments, compa-
nies in coordinated market economies can carry out long-term projects
in high-quality markets that are based on specific skills, but they are
slow to redeploy resources. The difference between the two types of
countries is evident in a number of ways, which Table 1.2 on the follow-
ing page sums up. One important difference are works councils, which
can promote long-term cooperation between management and labour
to continuously improve existing products. But at the same time, works
Table 1.2 Characteristics of coordinated market economies and liberal market

economies
Coordinated market Liberal market economies

economies
Industrial Cooperative Antagonistic

Relations Long job tenure Flexible labour market
Coordinated wage bargaining Atomistic wage bargaining
Coordination by the state, Weak organized labour and
associations or groups of capital, state supports atomistic
companies wage bargaining
Corporate Patient capital brings Short-term capital brings
financing long-term orientation short-term orientation and
flexibility
Concentrated ownership of Dispersed ownership
companies
Corporate Stakeholder model Shareholder model
Governance Cooperation with other Mergers and acquisitions,
companies hostile takeovers
Skills Investment in specific skills Investment in general skills
creation
Innovations Incremental Radical
councils are a disadvantage in liberal countries, where consultation

with them would make it difficult for management to quickly redeploy
resources from one project to another. Fittingly, as Figure 1.7 on the
following page shows, almost no liberal countries have works council
representation, while almost all coordinated countries do.
Figure 1.7 plots the variable ‘WC_rights’ from the 2011 version of the
‘Database on Institutional Characteristics of Trade Unions, Wage Setting,
State Intervention and Social Pacts’ (ICTWSS). The measure presented here
takes the value of 0 if works councils do not exist, only have information
rights, or no rights and no possibility for sanctions at all. The measure takes
a value of 1 if works councils have social rights; 2 if they have economic
and social rights, and 3 when works councils have economic and social
rights including codetermination. The picture that emerges is fairly clear.
In every liberal country apart from Ireland, works councils either do not
exist or do not have any rights apart from being informed; starkly opposed
to this, in all non-liberal countries apart from Portugal, works councils
have rights that go beyond this. Austria, Germany and the Netherlands
have the strongest co-determination. The four Scandinavian countries and
France and Belgium follow them. Switzerland and Japan, and the southern
European countries of Italy and Spain, have weak works councils.
0
Ze lia
nd
ng s
a
Po m
Ire l
d
n
ain
ly
Fi d
Sw nd
Be en
De ium
No rk
ay
er e
Ge ds
Au y
ia
ga
e
an
lan
lan
ad
pa
c
Sw Ita
str
do
a
ra
rw
Ne ran
lan
ala
nla
ed
ite Sta
rtu
Sp
nm
lg
rm
Un Can
Ja
Ne ust
er
F
itz
Ki
d
A
th
ite
w
d
Un
Figure 1.7 Rights of works councils in 2010

Source: ICTWSS Database 2011.
Hall and Soskice describe a similar clustering between liberal and

coordinated market economies for the level at which wages are bar-
gained. However, their data is from the 1990s and wage bargaining
systems have been liberalized since then (cf. Baccaro and Howell 2011).
This begs the question whether there is still a clear clustering in coordi-
nated and liberal countries when looking at more recent wage bargain-
ing levels. Following the definitions of the ICTWSS database, five levels
of bargaining can be distinguished. However, the fifth and highest level,
economy-wide enforceable agreements between central organizations,
does not exist any longer in the countries under scrutiny here. So Figure
1.8 distinguishes four bargaining levels. The fourth and empirically
highest level is that of mixed industry and economy-wide bargaining;
here, central organizations negotiate non-enforceable central agree-
ments (guidelines) and/or key unions and employer associations set
patterns for the entire economy. The third level is industry-wide bar-
gaining, but with no nationwide pattern setting, limited involvement
of central organizations, and limited freedom for company bargaining.
The second level is mixed or alternating industry- and firm-level bar-
gaining, with weak enforceability of industry agreements; the first level
stands for the absence of all these modes of collective bargaining, mean-
ing fragmented bargaining, mostly at company level (Visser 2011: 7).
In the following figure, countries are arranged according to their 2010
(black bar) bargaining levels. Next to this is the 1990 (grey) bargaining
level to capture changes since then.
0
ite S a
Ki ates
Ne Aus m
Ze alia
Ire d
Fr nd
Po nce
Sw ap l
itz an
Fi nd
Sw nd
De en
Sp k
n
Be ly
m
er y
er s
Au ny
ria
ga
a
ar
G nd
Un ited nad
an
ai
Ita
do
iu
w
a
st
la
la
a
ed
rtu
nm
r
lg
la
m
al
nl
Ne Nor
t
er
ng
Un Ca
th
w
d
Figure 1.8 Level of wage bargaining in 1990 (grey) and 2010 (black)
Again, there is a fairly clear clustering in liberal and non-liberal

countries. All six liberal countries have wage bargaining on the lowest
or second-lowest level. While there have been changes since 1990 (the
data that Hall and Soskice rely on), these hardly contradict their theory.
In fact, the picture is clearer now than it was in 1990, since the bargain-
ing level of Australia, New Zealand and Ireland declined. France is the
only non-liberal country with comparatively low bargaining coordina-
tion. As mentioned above, another crucial institution to differentiate
between liberal and coordinated countries is the financial system.
Liberal market economies have more equity-based capital. This means
that they should have a highly valued stock market relative to their
GDP. Figure 1.9 on the following page sketches out the market capitali-
zation for different countries to check whether this is the case.
To check for the impact of the financial crisis in 2008, Figure 1.9
includes data for 2007 (grey) and 2009 (black). As the figure shows, lib-
eral market economies (and the extreme outlier of Switzerland) indeed
have bigger capital markets. However, Ireland and New Zealand, while
liberal, nonetheless lack well-developed capital markets. Although
Sweden is coordinated, its market capitalization almost equals that of
the United States. However, apart from the extreme outsiders of Ireland
and New Zealand, a fairly clear picture emerges again.
Another point where liberal and coordinated market economies
should diverge is employment protection. The long-term oriented
coordinated countries should protect employment more than liberal
300
250
200
150
100
50
0
Au nd
Fi ly
Ze al
No um
nm y
ria
er d
Ne Por ny
Be and
Ne Ja k
er n
Fr ds
Sp e
Un Sw in
St n
ite Ca es
Ki ada
A dom
itz alia
nd
De a
ar
G lan
th pa
e
Ita
w tug
a
w
a
n
an
at
la
st
ite ed
la
i
Sw ustr
lg
m
la
al
n
r
Ire
er
ng
d
d
Un
Figure 1.9 Market capitalization of stock-noted companies as percentage of GDP
in 2007 (grey) and 2009 (black)
Source: World Bank (data from 2011), World Development Indicators Database.
countries, which should allow companies to be flexible in the hiring

and firing of labour. However, since Hall and Soskice’s publication,
employment protection has been liberalized. So do Hall and Soskice
still have a valid point in stressing institutional differences between
the two groups of countries? Figure 1.10 on the next page uses the
2008 version of a widely used generic OECD measure of employment
protection, which can range from 0 to 6, reflecting how hard it is for
companies to dismiss 1) short-term workers, 2) long-term workers and
to conduct 3) mass-layoffs.
The congruence with Hall and Soskice’s 2001 separation in liberal and
coordinated market economies is striking. The six countries with the
weakest employment protection are the six liberal countries. Conversely,
all non-liberal countries have stronger employment protection.
Using recent data, Hall and Soskice’s distinction between liberal and
coordinated market economies thus clearly still captures important
differences between countries. To be certain, not all indicators cluster
as clearly as employment protection does. Not all liberal countries are
liberal in all respects and not all coordinated countries are coordinated
3.5
3
2.5
2
1.5
1
0.5
0
ite Ca tes
Ne King da
Ze m
Au land
Ire lia
nm d
er n
Be taly
G gium
Sw Ja d
itz pan
Sw ark
Fi ds
Au d
ria
No ny
Po ay
Fr al
Sp e
n
De lan
th de
c
n
an
ai
g
w do
a
rw
an
na
st
la
a
rtu
str
I
la
m
Ne e
nl
er
St
l
er
d
ite
d
Un
Un
Figure 1.10 OECD indicator of employment protection in 2008

in all respects. What is important though is that the different parts of

economic regulation are not isolated but interact through complemen-
tarity, forming a coherent overall system.
1.2.1 Complementarity
Institutions support each other’s functioning, which is commonly
understood as complementarity. One prominent example is that bank-
based financing, centralized wage bargaining and long-term employ-
ment go together, as each of these contributes to the long-term stability
that the other institution needs for its functioning. For example, a com-
pany strategy of improving a product over the long term relies not only
on patient investors, but also on a long-term oriented labour force and
long-term cooperation with supplier-firms. If one of these is lacking, the
other elements function less well. Crucially, each of these aspects sup-
ports the other, forming a fully coordinated system.
Conversely, it makes more sense for a firm to use a flexible short-term
market strategy if it has access to flexible labour and flexible capital.
Thus, the central idea of the varieties of capitalism approach is that
coherence, understood as uniformly flexible or uniformly long-term
oriented institutions, leads to complementarity, understood as positive
effects of institutions towards each other.9 In short, either fully liberal or
fully coordinated countries are economically successful, so that ‘nations
with a particular type of coordination in one sphere of the economy
should tend to develop complementary practices in other spheres as

well’ (Hall and Soskice 2001b: 18).
Figures 1.1 to 1.10 have already shown that coordinated and lib-
eral countries are similar. However, if complementarity exists, we
should also see that within countries, different institutions are similar.
Figure 1.11 shows the coverage of collective bargaining, together with
the bargaining level and the power of works councils (both adjusted to
vary between 0 and 100) to indicate how different institutional subsys-
tems are similar.
Not only do coordinated countries have more powerful works coun-
cils (which is the leading indicator of the figure), their wage bargaining
also takes place at a higher level and more of their workers are covered
by collective wage agreements. There are of course some exceptions to
this. Few workers in coordinated Japan and Switzerland are covered
by collective wage agreements, while liberal Australia has a compara-
tively high rate. Yet overall, there is a fairly strong correlation (r²=.595)
Coverage of collective bargaining

Level of collective bargaining
Rights of works concils
100
y = 3,9711x + 24, 941
90
R2 = 0,595
80
70
60
50
40
30
20
10
0
es
da
l
n
ly
ce
No n
Be y
m
er y
s
ga
ar
n
nd
an
ali
pa
lan
ai
an
e
Ita
do
iu
at
an
rw
Ne ma
na
ed
tu
Sp
nm
str
lg
la
al
Ja
nl
St
er
ng
Fr
Ca
Sw
Ze
Fi
er
Po
Au
itz
De
Ki
d
th
G
ite
Sw
w
d
Ne
Un
ite
Un
Figure 1.11 Level and coverage of collective bargaining and degree of co-
determination in 2010
between the indicators. Note however, that Ireland is excluded, since

there was no data on its collective wage bargaining coverage, but it
probably would have distorted the clear-cut picture that emerges here.
1.2.2 The coordination index

Hall and Gingerich (2004, 2009) tried to quantify this institution-wide
degree of coordination or liberalism by measuring its different elements.
Using a principal factor analysis, they built an index that is made up
of sub-indices measuring: 1) shareholder power (legal protection and
influence of dispersed shareholders relative to managers or dominant
shareholders), 2) company-concentration (share of firms with large
shareholders to firms with dispersed shareholders), 3) size of the stock
market (stock market equities as share of GDP), 4) level of wage coor-
dination (dominant level where unions and employer associations
coordinate their wage claims), 5) degree of wage coordination (degree to
which wage coordination is strategic) and 6) labour turnover (share of
employees who held their job for less than one year – for all indicators
cf. Hall and Gingerich 2004: 11). The ensuing index of coordination is
correlated to all six individual factors, which can be seen as an empiri-
cal argument for the coherence/complementarity thesis. However, apart
from the problem that their data is for 1990–5 and thus out of date, it is
disputable whether the index covers all of the important variables of the
varieties approach (cf. Kenworthy 2006). Yet, the coordination index is
arguably the best measure available to quantify how much countries are
coordinated or liberal – if at least for a lack of alternatives. The following
sections therefore use it as a proxy to measure the degree of coordina-
tion in different countries. The indicator also allows classifying Italy,
and, to a lesser degree France, Spain and Portugal as coordinated market
economies, as these countries have higher scores on the coordination
index than liberal market economies (cf. Hall and Gingerich 2004: 14).
Before we use this index however, some weaknesses in Hall and Soskice’s
underlying typology need to be addressed.
1.2.3 Weaknesses of Hall and Soskice’s varieties of capitalism

typology
Varieties of capitalism gains explanatory power by combining the micro
with the macro level. On the micro level, it shows how companies are
influenced by the environment that they are embedded in. On the
macro level, it shows how a coherent institutional system of national
governance is built by giving actors guidelines in uncertain, game-
theoretical dilemmas. The approach thus illustrates how company
behaviour, if occurring en masse, forms a production system and it then

shows how this production system in turns affects company behaviour.
This in itself is a considerable achievement.
However, a weakness of this approach is that the welfare state hardly
plays any role in it (Soskice 2007: 92). Also, by classifying into only two
types of countries, the Hall and Soskice typology is maximally parsi-
monious but also minimally complex, so that other typologies capture
more empirical variety between countries (Deeg and Jackson 2006: 21).
Apart from this inevitable trade-off between parsimony and exactness,
the varieties approach is also attacked for claims that it only embodies
if interpreted in a certain way. Thus, Colin Crouch’s (2005a: 442) criti-
cism, to cite one example, sometimes misses the mark. He claims that
Hall and Soskice try ‘to allocate every developed capitalist economy
to one or other of two categories’. If carefully scrutinized, the varieties
approach makes more modest claims. It simply states that the two cat-
egories of market economies are to be seen as ‘ideal types at the poles of
a spectrum along which many nations can be arrayed’ (Hall and Soskice
2001b: 9). However, it is the essence of an ideal type that it never fully
reflects the complexities of empirical reality (cf. Weber 1949: 90, 43).
Therefore, Hall and Soskice (2001b: 21) indeed do not claim that they
try to allocate every country to one or another ideal type, as their critics
insist they do. Instead, they admit that a number of ambiguous cases
exist.
In addition, by stating that ‘according to the Hall-Soskice model, it
is not possible for firms within a liberal market economy to succeed at
incremental innovation’ and by arguing that the ‘model allows the firm
virtually no autonomy outside its national macroeconomic context’
(Crouch 2005a: 444), a deterministic power is ascribed to the approach
that it never claimed to have. Hall and Soskice (2001b: 17 – italics
added) simply claim that nations ‘tend to specialize in specific types
of production or products’. This however does not rule out variation
within nations. In fact, Hall and Soskice (cf. 2001b: 15) even explicitly
acknowledge such variation. Later works of their critics therefore actu-
ally concede that country-internal variation can be reconciled with the
nation-centred varieties approach (Crouch and Voelzkow 2009; Crouch,
Schröder and Voelzkow 2009).
If a biased understanding of the approach is applied however, then
the approach itself will appear crude. But this is not the fault of the
approach. To give an illustrative example, Hall and Soskice stress three
times in their introduction that their models of capitalism should be
understood as ideal types. They evoke the term ‘tend to’ 20 times and use
it for virtually every causal influence they describe. Thus, the approach
should not – but is often – criticized for a rigidity that it does not
embody. This is how Peter Hall (2009) replied to criticism by Wolfgang
Streeck (2009). Hall (2009: 491) claimed that:
most of the views Streeck attributes to that approach reflect a crude

caricature rather than a close reading of the relevant literature. That
is deliberate. Streeck spends the better part of a page (pp. 19–20),
explaining that he is not going to engage with the subtleties of the
varieties of capitalism literature on the grounds that the latter are
simply concessions designed to save the approach from its critics but,
in so doing, he constructs a windmill of his own at which to tilt.
Thus, while critics claim that the varieties approach, in short, makes
the world too simple, proponents of the approach claim that its critics
ascribe a rigidity to it that it was never meant to have. However, even
proponents of the varieties school admit that the approach is not good
at conceptualizing gradual change (cf. Hall and Thelen 2009). As later
chapters will show, a combination with welfare state research can com-
pensate for this problem. Therefore, I advocate combining the original,
namely dualistic varieties approach, with welfare state research. Two
more reasons exist to use the original, dualistic varieties approach. First,
most scholars criticize the approach vigorously, but in the end still
adopt its central distinction between coordinated and liberal market
economies (cf. the discussions in Hancké, Rhodes and Thatcher 2007a;
Hancké 2009; Crouch and Voelzkow 2009). In this sense, most scholars
take ‘an easy way out’ of the varieties of capitalism debate. On the one
hand, they (rightly) criticize that the world is more complex than the
varieties of capitalism approach portrays. On the other hand, the very
same scholars often fall back on the same coordinated versus liberal mar-
ket economy distinction that they criticize, implying that there must be
something to this ‘all-too simplistic’ distinction after all. Thus, if one
likes the typology or not, even critics admit that – similar to Esping-
Andersen’s typology – the ‘attention to institutions as building blocks
for coordination, and the coordinated market economy – liberal market
economy typology that resulted, have become the stock in trade of
political economists everywhere’ (Hancké, Rhodes and Thatcher 2007b:
36; empirically, cf. Schneider and Paunescu forthcoming). Second, as
even many critics of the varieties approach argue, there is a deeper,
theoretical reason, why a twofold approach is useful. As coarse as such
an approach is, a twofold differentiation makes it possible to distinguish
between market-led forms of capitalism and all other forms. That scholars
bemoan that ‘all other forms’ is too broad a category is a valid objection
and indeed this book will argue that it is helpful to distinguish between
two types of coordination: conservatively- and social-democratically
coordinated capitalisms. But such a more fine-grained distinction into
different types of coordination only makes sense in the shadow of a
typology which has established that all coordinated countries are dif-
ferent from all liberal countries in the first place, as the former types of
countries circumvent the market. That this circumvention happens in
different ways is another question, which the original varieties typology
does not contradict, but provides the foundation for.
In fact, the varieties approach lays the foundation for most other
recent typologies of capitalist diversity, as the following chapters will
show. Here, it is enough to bear in mind that I will use the most basic
distinction, notwithstanding acknowledgement that it can be broken
up into more fine-grained analytical units. Before proceeding however,
section 1.3 will give some information as to why in spite of all the valid
criticisms against Hall and Soskice’s varieties typology it makes sense
to use typologies at all and why it consequently also makes sense to
combine them. Readers who have little doubt about this can skip to
Chapter 2.
1.3 Why it makes sense to combine nations into regimes
Criticism that is pertinent already if addressed to varieties of capital-

ism or welfare state research alone – and the preceding section showed
that there is quite a substantive amount of it (for a good overview, cf.
Crouch 2005a, b; Amable 2003) – still gains in salience when addressed
to an effort to combine the two typologies. Therefore, such criticism
has to be addressed here. First, both the Esping-Andersen and the Hall
and Soskice typology are problematic, because they take the nation state
as the relevant unit of analysis. This is problematic because national
systems are allegedly undermined by homogenization from the inter-
national level and heterogenization from the regional one. Whether
the ensuing typologies of national differences provide an enduringly
accurate way of classifying countries is thus doubted, as the underlying
national arrangements, on which typologies are based, are allegedly
eroding (cf. Cerny 1997; Kitson 2005; Soederberg, Menz and Cerny
2005; Peck and Theodore 2007).
However, the notion that country differences erode can be countered
with a rich body of neo-institutionalist literature (Hall 1993, 1999; Hall
and Soskice 2001a; Kenworthy 1997; Kitschelt et al. 1999a; Pierson 1994,
1996, 2001a; Scharpf and Schmidt 2001b, c; Soskice 1999; Streeck and
Thelen 2005a; Whitley 1998; Wood 2001b). Using path dependency
and complementarity as underlying theoretical tools, empirical studies
reveal that convergence towards an Anglo-American variety of capital-
ism has indeed not taken place (Starke, Obinger and Castles 2008; Beyer
2002; Cerny, Menz and Soederberg 2005b; Crouch 2000: 33; Crouch
and Farrell 2002; Höpner 2001; Kitschelt and Streeck 2004: 3; Schneider
and Paunescu forthcoming 2012). Rather, studies show a surprising
‘unintended institutional fit’ (Aoki 1998: 235) of old institutions with
new governance modes, so that differences persist. Indeed, one can even
make the argument that, while all countries liberalize, the form that this
liberalization takes is different from one type of capitalism to another
(Martin and Swank 2012; Thelen 2012). So it is not so much that liber-
alization destroys capitalist varieties. It is also and possibly much more
the case that differences in capitalism shape trajectories of liberalization.
The end of the book will elaborate on this.
A second problem of the varieties approach is that national institu-
tions are punctured by diverging regional arrangements (cf. Crouch
2005a: 450; Crouch et al. 2001; Herrigel 1993; Hollingsworth 1998;
Piore and Sabel 1984; Whitley 1999; Crouch and Voelzkow 2009;
Crouch, Schröder and Voelzkow 2009). How can one speak of national
differences, when countries themselves are not internally coherent?
By claiming that systematic differences between as well as systematic
similarities within welfare and production regimes exist, this book does
not belittle the importance of regional intra-country variation. Instead,
it aims to show that the nation state is not the only, but one impor-
tant level of analysis, as a great share – for many purposes the most
important one – of the legislative body still emanates from this level
(Steinmo 2010: 222ff.).
A third problem is the (lack of) homogeneity of institutional arrange-
ments within nations. Individual institutions, such as the bargaining
system or labour market regulation, do not always come with the ideal-
typical model that they ‘should’ be assigned to (reflecting the embed-
ding welfare or production regime). For example, Denmark’s liberal
employment protection exists in a social democratic country. England’s
social democratic national health system exists in a liberal country.
However, as Esping-Andersen (1990: 2) notes, the salience of a typology
does not stand or fall with individual institutions, but with their com-
plex interplay. Nonetheless, this book will devote a section (2.3) to that
problem to understand its magnitude and how it can be dealt with.
Overall, academics have a paradoxical, one might even say cannibalis-

tic relationship to typologies such as Hall and Soskice’s varieties of capi-
talism or Esping-Andersen’s welfare regimes. While many criticize that
the overall picture that these typologies paint is too coarse, the very same
scholars often resort to a distinction between liberal and coordinated
market economies – or liberal and non-liberal welfare states – when they
analyse countries. The very fact that scholars resort to the same typolo-
gies that they criticize indicates that there must be something to them.
Chapter 2 shows that not only is this the case for valid empirical and
theoretical reasons. It also shows why these empirical and theoretical
reasons provide a good justification to combine the typologies.
2
Empirical Indicators and Existing
Typologies
2.1 Verifying diversity empirically
Typologies of welfare and production can only be combined when lib-

eral welfare regimes go along with liberal production regimes. In other
words, if a country scores high on welfare liberalism, it must also score
high on indicators of a liberal production system. In turn, high scores
for coordination should go together with indicators of a conservative or
social democratic welfare arrangement. This section will check whether
this is the case. As indicators for the make-up of a production regime,
I use the index by Hall and Gingerich (2004; also cf. 2009). I also use
a dummy variable for coordination and a variable for the degree of
corporatism as measured by Hicks and Kenworthy (1998). I correlate
these with Esping-Andersen’s scores for conservatism, socialism and
liberalism, as well as dummy variables for them.
The correlations in Table 2.1 on the following page provide a nuanced
picture.1 The degree of coordination and corporatism in the production
system correlates with the degree of conservatism in the welfare state.
Conservative welfare states seem to have coordinated production sys-
tems. However, the indicator for coordination is not correlated to the
degree of socialism or liberalism (first two rows). The surprise comes
with the dummy variables for the make-up of welfare states (last three
rows). All indicators of coordination correlate with the dummies for the
liberal and the conservative welfare regime; however, none correlates
with the dummy for the social democratic one.
What do these correlations mean? It is logical that coordination does
not correlate highly with indicators for a social democratic regime. This
is because social democratic countries take extreme values in terms of
the socialism of their welfare state, while they do not necessarily have
31
Table 2.1 Correlations between production and welfare regimes
Degree of Dummy – 0 Degree of

coordination for liberal corporatism
for different economy, for different
economies (Hall otherwise 1 societies
and Gingerich (Hicks and
2004: 14) Kenworthy
1998)
Degree of liberalism –.310 –.094 –.190

for different societies
(Esping-Andersen
1990: 74)
Degree of socialism .191 .310 .271
(Esping-Andersen
1990: 74)
Degree of conservatism .712** .433 .625**
(Esping-Andersen
1990: 74)
Dummy – 1 for liberal –.888** –.787** –.829**
welfare regime,
otherwise 0
Dummy – 1 for social .245 .289 .338
democratic welfare
regime, otherwise 0
Dummy – 1 for .627** .522* .514*
conservative welfare
regime, otherwise 0
N = 17 to 20 (data for Australia, Spain and Portugal sometimes missing).

* Correlation is significant at the 0.05 level.
** Correlation is significant at the 0.01 level.
the most extreme coordination. Correlations for the liberal welfare

regime are low, as many liberal countries with a low degree of coordina-
tion lack the correspondingly low scores for liberalism of the welfare
state. This is the case for the United Kingdom, New Zealand and Ireland.
In turn, some non-liberal countries such as Japan and Switzerland have
high coordination scores, but liberal welfare arrangements. Much of this
has been taken into account in Esping-Andersen’s later work (1999), on
which the dummy variables are based. Accordingly, these give a more
accurate picture and show stronger correlations. The social democracy
dummy does again not correlate strongly with coordination however,
as many conservative welfare states are more coordinated than many
Empirical Indicators and Existing Typologies 33
Table 2.2 Capitalisms structured into one framework
Liberal production Coordinated

system production system
High Social democratic

decommodification welfare state and
coordinated economy
Group-based Conservative welfare
decommodification state and coordinated
economy
Low Liberal welfare state
decommodification and liberal economy
social democratic welfare states. These correlations therefore yield the

scheme of Table 2.2, in which the degree of decommodification serves
as a proxy for welfare regimes.
As I will proceed to show in Chapter 4, the schema contains empty spots
for good theoretical reasons. Countries with liberal economies need com-
modified labour; countries with coordinated economies need more or less
decommodified labour. As Chapter 5 will show, the liberal policy style
of English-speaking countries prevented both coordination and decom-
modification; the high nationwide solidarity of Scandinavian countries
promoted both national coordination and strong decommodification,
while the group-based solidarity of continental European countries and
Japan promoted a conservatism that was both conducive to group-based
decommodification and coordination within the social groups in which
solidarity existed. Decommodification in these countries is group-based
because it only meant that workers would not have to work outside the
social confines of their social group, for example, they did not have to
take jobs below their skill level. However, as group-based decommodi-
fication leads to a different welfare state than in the social democratic
countries, we can distinguish three groups of countries. To illustrate this,
Figure 2.1 on the following page plots the degree of coordination in the
economy against the degree of socialism of the welfare state.
Here we see why in Table 2.1, there was no correlation between meas-
ures of coordination and measures of socialism. This is because the degree
of coordination of the production system, which also serves as a reason-
ably good proxy for the degree of conservativeness of the welfare state
(again cf. Table 2.1), was plotted against the degree of socialism in welfare
arrangements (cf. Hicks and Kenworthy 2003). Apparently, the relation-
ship between welfare and production regimes is more complicated than
2.00 Norway
Degree of welfare socialism (Hicks and Kenworthy 2003: 33)
Sweden
Denmark
1.00
Finland
New Zealand Belgium

Germany
Ireland Netherlands
0.00
United Kingdom
Austria
France
–1.00 Canada Switzerland Italy
Japan
–2.00 United States
0.00 0.20 0.40 0.60 0.80 1.00
Degree of economic coordination (Hall and Gingerich 2004: 14)
Figure 2.1 Production systems and welfare states in one grid
a simple linear one; an increment of coordination does not go along

with an increment of decommodification. It is thus doubtful whether
regressions are able to capture the picture of different welfare regimes and
production systems (also cf. Pierson 2000: 809; Shalev 1999).
Before using a qualitative analysis, I will however use Bruno Amable’s
(2003, also cf. Schneider and Paunescu forthcoming 2012) approach,
which has been widely acclaimed as the most appropriate method for
grouping individual countries into families (cf. Crouch 2005a). To use his
approach, I have accumulated a broad database, with variables that mirror
welfare and production regime arrangements (see Appendix A). I used this
dataset and submitted it to a principal factor analysis. In so doing, I leave
out variables that in themselves are designed to measure production or wel-
fare regimes, such as Esping-Andersen’s or Hall and Gingerich’s indicators.
The aspects that I have included can be seen in the factor loadings (see
Appendix B). The dent in the fourth factor of the scree plot in Figure 2.2
on the next page is good evidence that five factors provide a reasonable
trade-off between parsimony and accuracy for the data at our disposal.
They explain 71 per cent of the variance of the complete dataset.
15
10
Eigenvalue
0
1
3
5
7
9
11
13
15
17
19
21
23
25
27
29
31
33
35
37
39
Component Number
Figure 2.2 Scree plot of principal component analysis

Reprinted with permission.
After extracting the five factors, these were submitted to a measuring

of between-group linkages using the intervals of squared Euclidian dis-
tance. This should show which countries’ welfare and production system
variables are systematically close to the variables of other countries,
thereby possibly revealing a number of country-families by reflecting
both variables that mirror production and welfare regimes.
The ensuing dendrogram (Figure 2.3 on the next page) displays the cluster
analysis graphically. Based on the empirical indicators, country regimes
are ‘nested’ in each other. The broadest distinction runs between liberal
countries and all other ones – mirroring the Hall and Soskice typology,
but with the exceptions of Switzerland and Japan, which form their own
group. Within the non-liberal countries, the Scandinavian countries form a
sub-cluster, which thus contains all countries that Esping-Andersen labels
social democratic welfare states. Austria, France, Belgium, the Netherlands,
Portugal, Italy, Spain and Germany form the second non-liberal cluster.
These are all countries with a conservative welfare state, according to
Esping-Andersen’s typology. Within this group, the three Mediterranean
countries are a sub-cluster, possibly of extreme conservatism; and Germany
forms its own cluster within the conservative-coordinated group.
With the exception of Switzerland and Japan, which are continuous
outliers, the empirical clustering of countries thus reflects Hall and
CASE 0 5 10 15 20 25
Label Num
France 10
Austria 20
Belgium 16
Netherlands 9
Portugal 13
Italy 18
Spain 8
Germany 19
Sweden 11
Denmark 12
Finland 14
Norway 17
United Kingdom 3
New Zealand 5
Australia 1
Canada 4
Ireland 6
United States 2
Switzerland 7
Japan 15
Figure 2.3 Dendrogram of hierarchical cluster analysis

Reprinted with permission.
Soskice’s distinction between liberal and non-liberal countries on the

most general level. Within the non-liberal countries however, the major
distinction runs between countries with a conservative and a social
democratic welfare state. The cluster analysis therefore concurs with
Arts and Gelissen (2010: 574), who claim that the overlap between pro-
duction and welfare typologies is reflected by the empirical clustering of
indicators of the relevant countries.
However, a number of reservations have to be made against this kind
of empirical undertaking. Depending on which variables were included,
country clusters varied somewhat. Using slightly different variables, the
Mediterranean cluster and Finland composed separate subgroups. Other
hot candidates for outliers are Norway, Ireland and the United Kingdom.
However, there are also a number of recurring groupings, even if indi-
vidual variables are excluded or added. This provides evidence that there
are robust differences between and similarities within regimes. Thus, the
existence of a distinct Anglo-American cluster was usually confirmed.
In addition, a further consistent result is that Switzerland and Japan
formed separate clusters. Especially the continental European countries
can be divided into smaller groups. So while the major difference runs
between English-speaking and all other (non-liberal) countries, those
non-liberal countries can in turn be divided into two subgroups. In this

sense, the data resembles the grouping of the varieties of capitalism and
of the welfare regime approach. That the typologies are ‘nested’ in each
other in this way indicates that they can be combined. The next section
will look at established typologies to verify this.
2.2 Combining the country clusters of existing typologies
This section will show that most typologies depict a similar picture of
capitalist diversity, which is not only congruent with the cluster analy-
sis of the preceding section, but also with varieties of capitalism and
Esping-Andersen’s welfare regime typology. The problem is however,
that existing typologies either do not recognize that they fit with others
or do not consider the implication – namely that the most important
current typologies of capitalism can be unified into one encompassing
typology of capitalist diversity. This chapter will show why and how
existing typologies can be combined. Due to space constraints, I limit
myself to the most important typologies of capitalist diversity, to show
that all of these portray a similar picture. I do not consider the early
distinction by Wilensky and Lebeaux (1958) between ‘residual’ and
‘institutional’ welfare states. This is because Titmuss (1974) took up
this distinction, and his work has been refined by Esping-Andersen.
I will also not use Katzenstein’s (1985) distinction between liberal, sta-
tist and corporatist countries, as he refers to the beginning of the 1980s,
only looks at small states and because Schmidt (2000, 2002) largely
reproduced his typology. I will also not use typologies based on the
corporatism-literature (cf. Lehmbruch and Schmitter 1979, 1982; also
cf. Shonfield 1969) because these were refined by varieties of capitalism
through Hall and Soskice. The same is true for Albert’s (1992) distinc-
tion between a ‘Rhine model’ and an Anglo-American model of capi-
talism which has also been taken up by Hall and Soskice. I also leave
out Whitley’s (1999) distinction between different business systems,
as he does not provide a typology of national capitalisms, but of local
production systems. In Whitley’s (2007) later work, he does jump on
the bandwagon of differentiating national capitalism and distinguishes
between ‘arm’s length’, ‘dominant developmental’, ‘business corporat-
ist’ and ‘inclusive corporatist’ types of national capitalisms. But even
while this is reminiscent of a juxtaposition of a liberal model to differ-
ent coordinated models, Whitley never states what country belongs to
which of his types, which is why I cannot use his work here. As I am
interested in how other typologies connect to Esping-Andersen’s (1990)
welfare typology and Hall and Soskice’s (2001) varieties typology, and
since I have discussed these already in section 1, I will not discuss them
again here. Neither will I use the typologies based on ‘social systems of
production’ (Hollingsworth 1997a), as these are further refined by – and
contained in – Amable (2003). This leaves me with Vivien Schmidt’s
(2002), Robert Boyer’s (2004b), Bruno Amable’s (2003), Bernhard
Ebbinghaus’ (2006) and Crouch and Streeck’s (1997b) typology. I will
show how these existing attempts to classify capitalist diversity can be
integrated into one typology, together with Esping-Andersen’s welfare
regimes and Hall and Soskice’s varieties of capitalism. I will begin with
the typology that uses the smallest number of cases, which is Vivien
Schmidt’s.
2.2.1 Three capitalisms: Schmidt’s typology

Vivien Schmidt (2000, 2002) presents a typology based on case studies of
three countries: the United Kingdom, France and Germany. She analy-
ses how these countries reacted to globalization and a transfer of power
to supranational levels (the EU) given different economic vulnerabili-
ties, political institutional capacities, policy legacies, preferences and
discourses. She differentiates between monetary, industrial, labour, and
social policy to analyse these countries. For her three case studies, she
finds the following trajectories:
1. For the United Kingdom, she analyses the development from a crisis-
ridden country to an (until recently) successful market-led configura-
tion of the 1990s. She does this by tracing the efforts of the Thatcher
government. Being depicted as market-led, the United Kingdom is
seen as close to what varieties of capitalism considers a liberal market
economy, in which the state acts as arbiter over market-rules. She
presents the United States as a similar case.
2. For Germany, she describes the market-induced erosion of what was
once known as ‘Rhenish capitalism’. She still sees association-led
Germany as close to a coordinated market economy, where the state
promotes coordination between associations. She sees Sweden and
the Netherlands as similar cases.
3. For France, she describes the change from state-led to state-enhanced
capitalism. Even though the state still acts as a direct facilitator
towards business (different from its role in the United Kingdom and
Germany), its role is increasingly retrenched. She presents Italy as a
similar case, yet does not overlook Italy’s different inter-firm rela-
tions and state-led corporatist labour relations.
Schmidt (cf. 2002: Chapter 3) herself sees her approach as part of the
broader literature on varieties of capitalism. Yet, her crucial point is that
countries cannot be aligned to two types. Notably France constitutes its
own model, a point on which Hall and Soskice (2001b: 21) tentatively
agree. Yet, an enlarged notion of what constitutes a coordinated market
economy (being coordinated regardless of what actor coordinates, as
presented in section 1.2) could conceptualize France as coordinated,
since the state coordinates. This might be a ‘hierarchical direction rather
than joint-decision or unilateral action’ (Schmidt 2002: 144), but one
can still call it coordination in that it helps companies to cooperate in
prisoner-dilemma situations.
2.2.2 Four capitalisms: Boyer’s typology

Recent developments in the French ‘théorie de régulation’ (regula-
tion theory) point towards a convergence with varieties of capitalism.
Regulation theory has a long tradition. It builds on Marxist assumptions
and sees capitalism as inherently prone to crisis. Yet, in certain ‘accu-
mulation regimes’, it spots virtuous cycles of self-stabilization, such as
the mutual reinforcement of increasing demand and offer in Keynesian
economies (cf. Aglietta 1976; Lipietz 1987). Regulation theory analyses
how different countries regulate their economic system and how this
results in accumulation that is stable for a period of time. Robert Boyer
(cf. 2004b), arguably the most prominent proponent of the regulation
school, proposed a typology based on the different regulation regimes
that come with historically different growth trajectories. His typology is
based on five institutional domains: 1) the wage-labour nexus, 2) forms
of competition, 3) the monetary regime, 4) state-economy relations and
5) international trade embeddedness (Boyer 2004b: 13; 21). In analysing
these aspects, he depicts a fourfold typology of capitalist diversity.
1. Market capitalism: This model comprises all English-speaking coun-

tries. Production and welfare arrangements are market-based; the
state ensures that the market functions.
2. Meso-corporatist capitalism: This model comprises Japan and Korea.
Large and robust conglomerates reward their employees’ solidarity
with conglomerate-internal mobility and social protection.
3. State-driven capitalism: This model comprises those continental-
European countries that form the core of the European integration
drive. The state (or its equivalent on regional and local levels) assists
in production; public interventions ensure that economic projects
requiring large-scale coordination can be undertaken.
4. Social-partnership capitalism: This model comprises the Scandinavian

countries. The state and the social partners regulate in a way that
conforms to collectively held beliefs about egalitarianism, while also
aspiring to a high degree of competitiveness (for all four capitalisms,
cf. Boyer 2004b).
For a number of reasons, Boyer’s typology is a good extension of vari-

eties of capitalism and welfare state research. It combines a dynamic
dimension (analysing institutional change over time) with a static one
(classifying countries into a number of families for a given point in
time). It also analyses similar institutions as varieties of capitalism and
welfare state research (Boyer 2004a: 31, 2004b: 28, 2004c: 74f.) and
even tries to connect these (cf. 2004c: 17ff.). Largely, Boyer (2004c: 9;
also cf. Boyer and Hollingsworth 1997) stresses the same basic point
as welfare regime typology and varieties of capitalism, when he claims
that coordination (using a variety of institutions) can provide an eco-
nomically successful alternative to market arrangements. Is Boyer’s
typology thus already a combination of varieties of capitalism and
welfare state research? Indeed, Boyer (2004c) published a French book
whose title is telling: ‘Une théorie du capitalisme est-elle possible?’ (Is
a theory of capitalism possible?). But his approach has some problems.
Some cases, such as Germany, fit uneasily with his approach. Germany
can neither be regarded as one of Boyer’s meso-corporatist regimes,
since it does not possess the Japanese-style ‘keiretsu’ conglomer-
ates, with company-wide wage agreements (the formerly interlocked
German companies might have been considered as equivalent to these
in the past). Due to its tradition of corporatism, Germany can neither
be classified as Boyer’s ‘state-dominated’ capitalist variety, as this would
include strong state-imposed labour regulations and far-reaching state-
legislation on economic measures (cf. Boyer 2004c: 89). A similar point
can be made about the Netherlands, in which the state neither directs
nor influences the economy, as it does in France, Italy or Spain.2 It is
thus questionable whether ‘state-driven’ is a good label for continental-
European capitalism. It seems that Boyer had France in mind when he
invented this label and then generalized it to include all of continental
Europe.
A further arguable drawback of his approach is its ideological
undertone. Boyer describes the market-based mode of coordination
as inferior, due to alleged high unemployment, slow progression of
life-quality, uncertain profits and social inequality (Boyer 2004a: 60f.).
Even though at the time of writing around 2004, Boyer was right on
social inequality – the (unweighted) mean Gini index of coordinated

countries was 30.6, whereas that of liberal countries was 36.0 (World
Bank 2006) – he was wrong on unemployment and the wealth creation
capacity of different models. The mean unweighted unemployment rate
in the liberal economies in my sample was at 5.3 per cent, that of the
coordinated countries was at 6.9 per cent when Boyer wrote his text
(OECD 2005b: 238). GNI per capita in current prices and purchasing
power parity in liberal countries was at US$29.532, that of the coordi-
nated countries was somewhat lower, at US$28.971 (OECD 2006: 35).
Thus, Boyer’s claims lack empirical backing. If however he (normatively)
values equal income distribution as more important than GDP or unem-
ployment rates, he should make this obvious when he claims that non-
liberal models of capitalism are superior. It is not the aim of this book
to pursue a literal review of the workings of regulation theory in search
of undercurrent ideology, but examples can be given of Boyer citing the
American model as lacking social justice, without defining his under-
standing of social justice, which renders the term void of sense (Boyer
2004c: 101; cf. the problems on using the term ‘social justice’ without a
clear definition: Kersting 2003) or by citing the US market model as eco-
nomically unsuccessful in reducing production costs (Boyer 2004c: 96),
without actually delivering indicators to support this claim. Another
debatable drawback of regulation theory is that it predicts unpredict-
able change (cf. Boyer 2004a: 107, 2004c: 12f.). Varieties of capitalism
and welfare state research, contrarily, see themselves as capable of rather
deterministic forecasts (cf. Hall and Soskice 2001a: 57; Esping-Andersen
1990: 33).3 Instead of mutually replacing each other, regulation theory
could however complement these existing approaches. Table 2.3 on the
following page illustrates how.
Thus, many questions that exist in varieties of capitalism, welfare
state research and regulation theory can be answered by the respective
other approach. This is also true when we look at typologies that deline-
ate five groups of countries, such as Amable’s and Ebbinghaus’.
2.2.3 Five capitalisms: Amable’s typology

Amable (2003) builds on the insights of varieties of capitalism and wel-
fare state research to construct a typology of capitalisms based on the
analysis of five institutional domains. These are 1) product market com-
petition, 2) the wage-labour nexus and labour market institutions, 3)
the financial sector and corporate governance, 4) social protection and
the welfare state and 5) the education sector (Amable 2003: 14). He uses
dozens of indicators in each domain and performs a principal factor and
Table 2.3 Regulation theory, welfare state research, varieties of capitalism –

different theories, complementing explanations
Weakness of typology Solution by

complementary typology
Varieties of Lacks a dynamic framework to Framework for change

capitalism conceptualize institutional developed by regulation
change theory
Very rigid model, cannot Success of ‘intermediary’ and
explain performance of a ‘failure’ of pure cases can
number of countries possibly be explained by
welfare state research
Welfare state Cannot explain company Can possibly be explained

research performance by varieties of capitalism
Regulation Lacks a theory of firm Theory developed by
Theory behaviour varieties of capitalism
Cannot account for different Can be explained by
welfare state regimes welfare state research
cluster analysis on them, separately for each of the five domains. The
great point about Amable’s analysis is that he refrains from forming any
country-groupings ad-hoc or on a theoretical base. Instead, he performs
his statistical analysis with an open outcome. Nonetheless, his approach
leads him to distinguish five country-families that are reminiscent of
existing typologies.
1. Liberal capitalism: This model relies on the extensive use of markets

for the allocation of goods, capital and labour. The undisturbed
functioning of markets comes with a low degree of social and stra-
tegic interference by the state or other actors – such as associations.
Canada, the United Kingdom, the United States and Australia are in
this group; New Zealand and Canada are not analysed by Amable (for
all his country groupings, see Amable 2003: 178).
2. Social democratic capitalism: The small countries that use this model
tackle international competitive pressure by combining moderate
employment protection, high social protection and abundant offers
for retraining, as well as active labour market policy, with wage bar-
gaining that is solidary and favours innovation and productivity.
Finland, Sweden and Denmark are in this cluster.
3. European-integration capitalism: Countries of this model use more
employment protection and have a less decommodifying welfare state
than social democratic countries. Their financial system facilitates
long-term corporate strategies. Coordinated wage bargaining, retrain-

ing of the workforce and solidary wage policies exist, but are less exten-
sive than in the social democratic countries. Austria, the Netherlands,
Ireland, Belgium, Norway, Germany, France and Switzerland pertain
to this group. Amable (2003: 172) argues however, that Switzerland
and Norway could also be seen as a separate sub-group.
4. Mediterranean capitalism: What distinguishes ‘European-integration
capitalism’ from the social democratic countries is even more pro-
nounced in the Mediterranean countries. Thus, they have more
employment protection and less decommodification. They also have
low product market competition and little short-term pressure for
profits, due to a centralized financial system that can (could) devalue
national currencies. Limited qualifications obstruct a high-skill or
high-wage strategy. Greece, Italy, Portugal and Spain make up this
cluster.
5. Asian capitalism: This capitalism depends on the business strate-
gies of large corporations that work with the state and a centralized
financial system. Employment protection and retraining within
companies protects skills and provides a functional equivalent to a
strong welfare state. It also promotes a long-term perspective. Japan
and Korea constitute this group.
Even though Amable distinguishes five groups, one finds in his

typology that all groups apart from the liberal countries circumvent
markets in one way or another. In fact, Amable (2003: 176) claims
that ‘the distinctiveness of the market-based model explains why the
dichotomous classification liberal market economy – coordinated
market economy can sometimes be adopted as a first approximation’.
Accordingly, in his last chapter he distinguishes mainly between coun-
tries that rely on markets in governing their production and welfare
system and a (variable) number of country groups that rely on non-
market coordination. Intriguingly, the groups that he presents as alter-
natives to the liberal group are the same ones that one ends up with
when Esping-Andersen’s initial distinction is widened, as Ebbinghaus
(2001), Ferrera (1996), Leibfried (1992) and Lessenich (1994) sug-
gested. Therefore, instead of an alternative to Esping-Andersen and
the varieties typology, Amable is an addition. He simply places fine-
grained analysis over parsimony, by subdividing the larger groups of
existing typologies into subgroups.
The advantage of his typology is also its biggest problem: it is strictly
empirical and does not theorize the linkages between welfare states
and production systems. A typology that results from a combination of

varieties of capitalism and welfare state research must however help to
explain why certain welfare states come along with certain production
regimes. This is what the unified typology that is proposed here will do.
However, before this is done, we have to have a look at another fivefold
typology.
2.2.4 Again five capitalisms: Ebbinghaus’ typology

Bernhard Ebbinghaus (2006: 20) analyses ‘welfare regimes, produc-
tion systems, and labour relations, synthesizing these into a combined
heuristic tool to discuss the institutional affinities between the spheres
of protection, production, and partnership’. He does this to map
the institutional environment in which early retirement takes place.
Concretely, he differentiates 1) the welfare state, into a conservative-
corporativist, conservative-subsidiaristic, universalist, liberal-residual and
liberal-residual-familist model. He also analyses 2) the production system
of different countries, distinguishing between sector-coordinated, state-
coordinated, centrally-coordinated, firm-coordinated and uncoordinated-
liberal arrangements. Lastly, he also analyses 3) labour relations, where
he differentiates between cooperative social partnership, contentious,
cooperative neo-corporatist, voluntarist and cooperative firm-level labour
relations. He observes that types of welfare states, production systems and
labour relations systematically come about in pairs. In his words, there are
‘institutional affinities between particular regimes of protection, produc-
tion, and [social] partnership institutions’ (Ebbinghaus 2006: 76). That
welfare states and production systems come in pairs allows Ebbinghaus
to group Germany, the Netherlands, France, Italy, Sweden, Denmark, the
United Kingdom, Ireland, the United States and Japan into the following
five types of capitalisms:
1. Centre Countries: Germany and the Netherlands fall under this

regime-type, as they have a broadly conservative welfare state, which
is different from other conservative welfare states in being ‘corporativ-
ist’, instead of conservative, which describes the welfare state of Latin
countries. The production system of these countries is coordinated
by sectors. It is thus different from liberal countries. However, it is
also different from countries in which state-institutions (Latin coun-
tries) or firms (Japan) coordinate, or where coordination is centrally
organized, as in the Nordic countries.
2. Latin Countries: France and Italy fall under this regime-type. They
have a conservative welfare state that not only came late, but also
relies on the traditional intermediary institutions of family or

church. As this welfare state is subsidiary, it is different from the
conservative-corporativist one of the Centre countries. The produc-
tion system of these countries is state-coordinated, because they lack
strong firm-sectoral- or central-coordinating capacities. They display
contentious labour relations, in that their unions are organization-
ally weak but have strong veto power.
3. Nordic countries: Sweden and Denmark fall under this label with
their universal welfare state, their centrally coordinated production
system and their cooperative neo-corporatist labour relations.
4. Anglophone countries: the United Kingdom, Ireland and the United
States fall under this category. They all have a liberal-residual welfare
state, an uncoordinated (or one might say a market-coordinated) lib-
eral production system and voluntarist labour relations. Organization
in trade unions is not prohibited, but neither institutionally nor
legally supported. This sets these countries apart from all others.
While other country clusters are either variants of conservative wel-
fare states, of coordinated production systems or of countries with
cooperative labour relations, this group is different, as it uniformly
relies on markets in its welfare, production and industrial relations
arrangements.
5. Asian countries: Only Japan forms this group, which Ebbinghaus sees
as its own model due to a unique combination of a liberal-residual
familialist welfare state, a firm-coordinated production system and
cooperative firm-level labour relations (cf. this grouping Ebbinghaus
2006: 76).
Different from the typologies discussed so far, Ebbinghaus neither

relies on a unique theoretical tradition, as Boyer does with regulation
theory. Nor does he perform the extended empirical clustering of coun-
tries, which Amable engages in. Instead, he relies on the established
typologies of Esping-Andersen, varieties of capitalism and on Crouch’s
(1993) attempt to find regularities in labour relations. He does this to
study a specific policy in depth: early retirement. As he relies on existing
typologies, it might be less surprising than in the case of Amable’s,
Boyer’s or Schmidt’s typology that his country groupings are compatible
with existing ones.
Two aspects are striking however. First, Ebbinghaus finds that arrange-
ments in the welfare state tend to coincide with similar arrangements
in the production system and in labour relations within countries and
groups of countries. For example (and most clearly), liberal countries
have market arrangements not only in their welfare, but also in their
production system and in their labour relations. In turn, Japan’s firms are
not only responsible for economic coordination; they are also the locus
of cooperative labour relations and provide social services that the welfare
state provides in other countries. In stark contrast to this, the universal
nation-wide welfare state of the Nordic countries is flanked by equally
universal (central, as Ebbinghaus calls it) economic coordination in the
production system. In between, cooperation in the Centre countries is
consistently on a middle level, manifesting itself in a conservative wel-
fare state that keeps everyone where he is socially without redistributing
much, and cooperative social partnership that extends above the firm-
level but is not nationwide.
In this sense, one can argue that a certain ‘policy style’ within coun-
tries and country regimes not only permeates the welfare, but also the
production system and aligns both to a common principle. For example,
if a country uses market-arrangements in one sphere, other spheres seem
to be governed according to the same logic. We will come back to this
observation in Chapter 3, as it contains the seeds for a unified typology
of capitalisms. Second, what is interesting about Ebbinghaus’ typology is
that it reconciles the demand for parsimony (a small number of regimes)
versus that of exactness (fitting countries into a type where they fit). He
does so by arguing that countries can be subtypes of the same regime. In
that sense, Ebbinghaus follows Esping-Andersen’s classification in that
both Germany and the Netherlands on the one side and France and Italy
on the other side have a conservative welfare state. Yet, whereas the first
group has a conservative-corporativist welfare state, the second group has
a conservative-subsidiaristic welfare state. Similarly, all countries apart
from the Anglophone ones have some form of coordination, but they can
be subdivided into sectoral-, state-, central- and firm-coordination.
This offers a fruitful route out of the dilemma between parsimony
versus exactness, by stressing that proponents of both views have
a point. In fact, their views are not mutually exclusive, since more
exact divisions are ‘nested’ in broader ones. This insight will be used
to argue why different typologies are not mutually contradictory, but
show essentially the same picture, even though they delineate different
numbers of country families. The more fine-grained differentiations of
more exact typologies are the building blocks for broader distinctions.
Conversely, these broader distinctions of capitalist typologies can be
subdivided into finer groups.
The question then arises why Ebbinghaus’ typology is not already the
combination of varieties of capitalism and welfare state research, which
is envisaged here. First, Ebbinghaus only highlights one specific area in

detail: pre-retirement policy. Even though it is true that complementari-
ties exist between welfare and production regimes in the area of early
retirement (cf. section 4.6), there are also other areas where welfare
states and production regimes function according to a similar policy
style. These other areas are not Ebbinghaus’ focus of analysis though.
Additionally, he only uses half of the countries that are analysed here
and therefore leaves the question unanswered if the general image of
capitalist diversity that he portrays is simply a result of the (relatively
few) countries that he analyses. Therefore, one must look at more coun-
tries. This is not only what is done in this book, it is also what Crouch
and Streeck have attempted. The following section therefore looks at
their typology as the last one that will be reviewed here.
2.2.5 Ever more diversity: Crouch and Streeck’s typology

Even though Crouch and Streeck (1997b) use a dualistic distinction
between ‘institutional’ and market-oriented capitalism, they also argue
that ‘there is a need to proceed beyond crude distinctions between insti-
tutional and pure market economies’ (Crouch and Streeck 1997b: 7). In
their edited volume, they rally prominent authors who each presents
one country as a case of its own, even though common traits are not
overlooked. Interestingly, as if this were a matter of course, all authors
analyse aspects of production and welfare arrangements to describe
capitalist diversity.
Dore (1997) contrasts Japan to Anglo-American practices of corporate
governance, in which the firm is seen as property of the shareholders
and has to be treated according to their legitimate property rights. In
Japan, the firm is rather a community of people, much like a nation.
Loyalty and moral obligations play an important role. This also stabi-
lizes relationships of companies with their suppliers and banks.
Wolfgang Streeck (1997b) inquires about ‘German Capitalism: Does it
Exist? Can it Survive?’ He stresses Germany’s distinctiveness as well as
the success of its economic configuration in combining external com-
petitiveness with lack of social conflict and high but egalitarian wages.
He explains Germany’s economic success through the rich institutional
infrastructure available to small- and medium-sized firms. An organ-
ized labour movement that takes a cooperative approach to industrial
relations by negotiating binding agreements also explains why German
companies are successful. Streeck claims that German producers are
competitive because they are strongly socially embedded. Further aspects
that explain Germany’s success are an ‘enabling state’ (Streeck 1997b: 38),
a social market economy and, such as Dore (1997), Streeck (1997b: 39)
stresses the role of a ‘traditionalist’ culture with high savings rates, a
preference for quality and security, as well as technical knowledge being
highly valued. Even though he sees the German model as possibly being
swept away by external pressures, he describes it as desirable due to its
economic and social outcomes.
Pontusson (1997) presents Sweden as a case that is distinctly different
from Germany, even though the two share many characteristics when
compared with Anglo-American economies. However, a predominance
of large firms, a more encompassing (national) wage bargaining system
and a larger public sector, combined with a more inclusive welfare state
are key differences between Sweden and Germany. Even though the two
models produced similar results concerning social outcomes, Pontusson
claims that they diverged since the mid-1970s – with Germany presented
as the more successful case (cf. Pontusson 1997: 56).
Boyer presents France as an essentially state-driven variety of capitalism.
He explains that ‘the distinguishing mark of the French model has to do
with the intervention of central government exercising, first, a strategic,
then a permanent, function with the object of piloting a modernization
programme which seldom reflected compromise actually achieved between
management and unions’ (Boyer 1997: 78). He therefore explicitly con-
trasts this state-driven variety of capitalism with a market-oriented,
corporatist and social democratic variety.4
Regini (1997: 108) holds Italy to be a distinctive case, where the weak
regulatory capacity of the state leads to ‘voluntaristic ad-hoc regulation’
in social networks. He argues that this brings a structural advantage
for small firms based on two main factors. 1) Regulation such as taxa-
tion and employment protection can be enforced more easily in bigger
companies and thus reduces their flexibility. 2) Other aspects, which
the state is unable to enforce in any case, bring an uncertainty and
instability that is problematic for big firms, while small firms can more
flexibly adapt to this uncertainty. He thus argues that the same lack of
regulation that is problematic for large firms is an advantage for small
firms, as it allows them to flexibly rearrange production to fulfil volatile
demand.
Andrew Graham (1997: 119) elaborates on how the United Kingdom
found its way back to ‘orthodox capitalism’ based on ‘small firms, free
trade and free markets’. He finds that this sets the United Kingdom
apart from Germany, Japan, but also from the United States. The change
of economic policy in the United Kingdom between 1980 and 1995
included a switch to supply-side politics as well as privatization and
deregulation of public infrastructure, welfare and production. Graham

criticizes this by outlining the usual arguments that are put forward
against neo-classical policy, notably that such policy erases the social
foundations, on which capitalism rests. For the present purpose, it
is sufficient to note that Graham sees little in the governance of the
English production system and welfare state that is not aligned to
market principles.
Faced with saturated markets and legislation that rendered price-
fixing agreements impossible, firms in the US opted for horizontal and
vertical integration, as Hollingsworth (1997b) describes in his chapter.
Legislation and a huge domestic market historically gave rise to a system
of mass production. However, this system needed to adapt as it came
under increasing pressure in high-quality and high-profit markets by
competition from Japan and Germany, not to talk about countries with
significantly lower wages. Hollingsworth describes how this led to the
reactivation of a ‘dormant’ institutional structure that uses informal
company networks in clusters, state-sponsored research and the flex-
ibility that the American regulatory system offers, to excel in industries
that bring innovations and new ideas to market use. As this structure
exists alongside the established American low-skill/low-wage equilib-
rium, Hollingsworth depicts a double nature of American capitalism.
Since both production modes rely on little state-regulation, the United
States is an economy in which external institutional constraints are of
little importance.
Colin Crouch (2005a, b) further refined the approach of privileging
exactness over parsimony by assigning an individual variety of capital-
ism to every country. He argues that a ‘labelling’ approach assigns coun-
tries to a limited number of regimes, whereas an ‘analysing’ approach
focuses on individual specificities. I argue that, just as we need maps
of different scales for our spatial orientation, we also need to combine
Crouch’s two approaches. Whereas it can be useful to have a world
map to get the big picture, a small-scale city map will be more useful
when trying to find one’s way through Paris’ wicked one-way streets.
Analogously, different levels of analysis are necessary depending on the
questions asked. If we want to know if Sweden is more like Denmark
or more like the United States, we need another (broader) framework
than we need when we try to answer the question whether important
differences exist between Denmark and Sweden, in which case we need
an approach that stresses exactness over parsimony. However, there
are also a number of outlier countries that obstruct smooth typologies,
which I will now deal with.
2.3 Dealing with outlier countries
If this book were a play, divided into the five classical acts, then this
chapter would make for a great second act, in which the escalation of the
conflict takes place after the principal actors (varieties of capitalism and
welfare state research), the supporting actors (other typologies) and their
relation to each other have been introduced. This chapter thus illustrates
the breadth of the problem (lets the conflict materialize). This is neces-
sary because, as the last chapter showed, capitalism comes in a consider-
able variety. Yet, while some countries provide emblematic cases, and
even those are debatable, more often than not countries are difficult to
clearly place into a regime-type. It is relatively easy to see that Germany
is the typical case of a coordinated market economy and of a conserva-
tive welfare state, whereas the United States and the United Kingdom are
often cited as key examples of liberal market economies and of liberal
welfare arrangements. To challenge this simplicity and to show that
almost all countries exhibit more complicated institutional structures
than typologies can fully account for, Table 2.4 on the following page
sums up how countries’ institutions diverge from ‘their’ regime-type.
Given this much diversity that sets countries apart from ‘their’
regime, a slight depression can befall every scholar that tries to classify
countries. Indeed, the institutions that set countries apart from ‘their’
regime-type are a powerful argument that regimes, on which typologies
rely for classifications, do not have a deterministic impact on countries’
institutional arrangements. It is therefore certainly true that a crude
usage of typologies can blind scholars towards important national spe-
cificities (cf. Crouch 2001, 2005a, 2005b). This illustrates the necessity
of Crouch’s ‘analysing’, as opposed to a ‘labelling’ approach. Indeed,
whether the ‘labelling’ approach still makes sense at all can be ques-
tioned when almost every country is an exception in some way. I claim
that for a number of reasons however, the high complexity and diver-
gence of institutional arrangements is less of a problem than it might
seem at first. The following paragraphs show why this is the case.
The number of outlier countries is already reduced if we follow Esping-
Andersen’s (1999: 88) dictum that ‘one programme does not define a
regime’. This implies that even if a country shows some divergence
from ‘its’ regime type, it has to be checked whether this is an isolated
exception of one institution or a rule that changes the structure accord-
ing to which the production or welfare regime of a country functions.
In this sense, divergent institutions can provide functional equivalents
to what could be seen as ‘typical’ institutional set-ups. Therefore, that
Table 2.4 Incoherencies in welfare and production regimes
Incoherencies in welfare arrangements Incoherencies in production arrangements
Australia Historically: Regulated labour markets, strong labour courts Some coordinated wage bargaining (cf. OECD 2004: 145)
pursued decommodification until the 1990s (cf. Castles 1996;
Castles, Gerritsen and Vowles 1996; Castles and Mitchell 1993)
Austria
Belgium Aspects of contestation5 alongside corporatist structures,
some equity-based financing (cf. Pontusson 2005a: 207)
Canada Universal health care system (cf. Esping-Andersen 1999: 18)
Denmark Low employment protection combined with high wage
replacement rates (flexicurity)
Finland Latecomer in the Scandinavian cluster Strength in telecommunications (cf. Crouch 2005a: 444)
France Childcare arrangements close to the Scandinavian model Low unionization rate, associations contestational,
(cf. Marc and Zajdela 2005) state takes up coordinating role (cf. OECD 2004)
Germany Some movement away from the insurance system Strong local and sectoral variation of production
(for example, with the ‘Hartz IV’ legislation) system (cf. Casper 2002; Herrigel 1993, 1996). General
demise of ‘German model’ (Streeck 2009)
Ireland Centralized bargaining since the 1980s, now some
neo-corporatism (cf. Amable 2003: 17; Soskice 1999: 124)
Italy Mediterranean welfare state, low redistribution of income Union structure contestational (cf. OECD 2004)
(cf. Ferrera 1996; Leibfried 1992; Lessenich 1994) Strong regional variations in production systems perforate
a weak national model (cf. Piore and Sabel 1984)
‘Hindering’ role of the state (cf. Regini 1997)
Japan Welfare state plays a marginal role, very low public provision Weak unions on an economy-wide level, which stems
of social welfare in terms of share GDP spent, social protection from a weak organized labour movement in general
in conglomerates. Ambiguous ranking by Esping-Andersen as (cf. OECD 2004)
‘liberal’ (1990: 74) and ‘conservative’ (1999: 90f.) Does not fit into empirical grouping of countries
Does not fit into empirical grouping of countries (cf. Figure 2.3) (cf. Figure 2.3)
(continued)
Table 2.4 Continued
Incoherencies in welfare arrangements Incoherencies in production arrangements
Netherlands Contradicting ranking by Esping-Andersen: considered to Firm-ownership dispersed, equity-based financing

be ‘social democratic’ in 1990: 74, yet considered to have (cf. Amable 2003: 17f.; Pontusson 2005a: 207)
many conservative traits in 1999
New Historically: regulated labour markets, strong labour courts
Zealand pursued decommodification until the 1990s (cf. Castles 1996;
Castles, Gerritsen and Vowles 1996; Castles and Mitchell
1993; Menz 2005b)
Norway Resources from raw materials prevent austerity Seen as partially liberal by Amable (cf. 2003)
Portugal Mediterranean welfare state, often neglected by welfare Contestational union structure (cf. OECD 2004)
state research and varieties of capitalism
Spain Mediterranean welfare state, also lacking analysis Contestational union structure (cf. OECD 2004)
by welfare state research and varieties of capitalism
Sweden Much equity-based financing (cf. Pontusson 2005a:
207), strength in radical innovation in health
sector (Crouch 2005a: 444)
Switzerland Unregulated working-market (cf. Amable 2003: 17) with Pluralist unions (cf. Amable 2003: 17). Shareholder-
little employment protection (cf. Kenworthy 2006: 77) oriented corporate governance system (Börsch 2007:
Does not fit into empirical grouping of countries 174). Does not fit into empirical grouping of countries
(cf. Figure 2.3) (cf. Figure 2.3)
United Public health system, high public social spending.6 Mixture Historically early collective bargaining and exceptionally
Kingdom of dominant liberal- and weak social democratic welfare strong labour movement (cf. Crouch 1993: 84; 95)
arrangements (cf. indicator Hicks and Kenworthy 2003)
United Forerunner of some aspects of social policy with the Large military sector shielded from market. Until
States ‘New Deal’ in the 1930s 1914: investment banks structure economy and cartels
Company-networks in which coordination takes place
(cf. Hollingsworth 1991)
Ireland and Australia have some coordinated wage bargaining is not a

problem as long as associations do not coordinate the economy or push
for a more extensive welfare state; that is, as long as an institution is
isolated. The same is true for a universal health system in Canada or in
the United Kingdom, for ‘flexicurity’ in Denmark, equity-based financ-
ing in the Netherlands or Sweden and organized company networks in
the United States.
A second reason why typologies make sense in spite of country
peculiarities is that countries often use different institutions to reach
functionally equivalent outcomes. For example, France’s conflict-prone
associations cannot coordinate the economy. But the state and elite
networks coordinate the economy in a functionally equivalent way.
Companies in France therefore still excel in industries that need large-
scale coordination, such as aeronautics and defence. In addition, labour
markets, though decentralized, still have significant job protection
because of state regulations on recruitment and dismissal (cf. Schmidt
2002).
Third, there are also a number of historical outliers, which have
reverted back to ‘their’ type over time – not for incidental reasons, as
Chapter 5 will show. Australia and New Zealand for example had a very
egalitarian income distribution – for liberal countries. Welfare regime
and labour market arrangements were relatively regulated and labour
courts decommodified workers. The United States in turn was a forerun-
ner of social policy in the 1930s; Britain historically had a strong labour
movement and was a welfare forerunner for some time with its 1942
Beveridge report. Germany was going through a period of liberalization,
which might have led it away from being the archetypical example of
a coordinated market with a conservative welfare state (Streeck 2009) –
examples abound. However, the crucial point is that coordinated
arrangements could not survive in liberal countries in the long run,
even though they did survive in countries with conservative or social
democratic welfare states. Why is it for example that New Zealand and
Australia – from the beginning of the 1980s onwards – modelled reforms
on the United Kingdom and the United States (cf. Casey 2006: 7f.; Menz
2005b)? It was not for pure economic necessity. If it were, Australia or
New Zealand could also have adopted the Japanese, Swedish or German
model of economic governance, each a successful example of how to
organize a capitalist economy in its heyday. Yet their path led these
two countries back to liberal traditions. I will argue in Chapter 5 that
this is because country regimes embody certain policy styles, to which
member countries tend to return. A similar argument can be made for
Finland, which found its way into the Scandinavian cluster or for Spain
and Portugal, which established a conservative welfare state.
This does not mean that a country is bound to stay in its regime type
forever. Countries can fit uneasily with their regime type for long peri-
ods of time. Such cases cannot be treated as mere exceptions to the rule.
However, there is reason to believe that, as Schumpeter (2003 [1943]: 12)
said: ‘Social structures, types and attitudes are coins that do not readily
melt’ so that what stands behind welfare and production regimes will per-
sist over time and even revert countries to their regime-types. As mentioned
above and as Chapter 5 will show, there seem to be ‘hegemonic belief sys-
tems’ or ‘policy-paradigms’ (cf. Lehmbruch 2001: 41; Hall 1993) that span
country-regimes and work towards aligning individual countries to their
regime-type, even if this does not preclude divergence during significant
time periods. For example, when Germany cut unemployment benefits in
2005, this was a divergence towards its conservative regime logic, because
it ended the principle of status-protection. However, such moves triggered
widespread popular protests in Germany, as they go against normative
orientations that were previously in accordance with institutional arrange-
ments and now work to realign the country towards a more conservative
welfare regime. Such protests against the principle of status-stabilization
would be unlikely in liberal countries such as the United Kingdom or the
United States. This does not mean that institutional changes never go
against regime-typical policy. It does however mean that change indeed
is less likely to occur when it has to take the hurdle of being normatively
undesirable. Chapter 5 will show in how far this aligns countries to their
regime type in production and welfare arrangements.
Then there are also the Mediterranean countries of Italy, Spain and
Portugal that do not fit easily into the continental group. However,
evidence that there are significant structural similarities between them
and the conservative regime type is provided by the hierarchical cluster
analysis in section 2.1 and theoretical arguments by Esping-Andersen
in section 1.1. Yet, if we further want to subdivide the group of conti-
nental countries, the Mediterranean countries would be a good candidate
for a separate subgroup. But especially where to put the Netherlands,
Switzerland or Japan (cf. Aoki 2000) is highly – and rightly – contested. At
least Switzerland and Japan do not fit into the groups that the hierarchi-
cal clustering (cf. Figure 2.3) indicates. They are also recurring outliers in
Chapter 1. So even though some outlying cases provide less of a problem,
the remaining ones bring up some methodological implications as to
what a typology can (and cannot) accomplish and how it should therefore
be constructed. The following section elaborates on these aspects.
2.4 A methodological interlude – how to build typologies
As seen above, almost no country is typical of its welfare or production

regime. This is obviously a problem when trying to integrate welfare
and production regimes into a unified typology. This section therefore
clarifies why, in spite of the peculiarities of individual countries, a uni-
fied typology of production and welfare regimes can – and should – be
constructed. Notably, such a typology is possible as long as three aspects
are borne in mind: 1) regimes must be ideal-typical; 2) the number of
regimes must remain variable; 3) the possibility has to be envisaged that
a number of countries cannot be classified at all, but are to be understood
as an assemblage of disparate models.7 Drawing on another example that
we readily categorize (flowers), this section shows how a unified typol-
ogy of capitalist diversity is possible, as long as we keep these caveats in
mind.
1) The difference between ideal type and real cases

If we scrutinize flowers of the same kind, we see that each is unique,
even though it is part of a broader type. Every flower will vary in size,
shape, colour and a number of other attributes. Accordingly, every
empirically observable flower will show a slight alteration towards
its ideal-typical conception. Yet we readily group flowers into catego-
ries, even though no two flowers are alike. The same can be argued
for country-typologies and individual countries. Even though it is
impossible to deny that every country is unique, neither can one
refute that systematic similarities between countries exist, as the
previous sections illustrated. If we had no regimes to determine some
regularity in the first place, we would not be able to distinguish what
is unique about individual countries either. Thus, far from arguing
that there are no differences within country-regimes, I argue that it
is well-developed regime types that permit recognition of national
peculiarities in the first place. For this to apply, countries must sig-
nificantly cluster into regimes, so that these are discernible as mean-
ingful entities. This however is the case when empirical indicators
of welfare and production arrangements are analysed together (cf.
section 2.1).
2) The undetermined number of cases

Notwithstanding that we treat flowers as a group, by distinguishing
them from animals and humans for example, we still find it useful
to further split up this group to achieve a more fine-grained analysis.
Thus, we distinguish roses from orchids or pansies and even divide the
group of ‘roses’ into different subgroups. We have no problem in doing
so as long as we bear in mind on what analytical level we operate and
as long as we do not forget that any flower is at the same time unique
and pertaining to a group. The same thing should be true for how we
deal with typologies. Different typologies group capitalist countries
in a varying number of categories. However, these categories are not
mutually exclusive but rather complement each other, as section 2.2
showed.
3) The problem of unclassifiable cases

Even though we find it useful to treat the category of flowers as an
entity, we also divide it into individual cases and are even able to
distinguish two flowers of the same kind. We are aware that categories
we have already constructed might not always suffice. As new flow-
ers are found, we are able to subsume them under existing groups.
Alternatively, we create a new group if we find existing ones inad-
equate. Thus, the fact that flowers cannot readily be classified leads
us to extend our systems of classification, instead of abandoning
them.
Again, the same should be true for capitalist diversity and its typolo-
gies. The fact that individual cases are sometimes not readily classifiable
is no reason to not classify those countries that can be subsumed under
groups and develop new groups for cases that we cannot account for.
Yet, there is one important possible difference between the category of
‘flowers’ and of capitalist countries. We can only use categories when
there are systematic and stable similarities between them; only then
can we draw out the differences within a group. The necessary condi-
tions are without doubt given for the example of ‘flowers’, as we can
define all flowers as having a head and a stem and we can expect that
this will not change. The previous chapters showed however, that
a similar common lowest denominator can be reasonably assumed
for production and welfare regimes. Thus on one hand, all countries
use the market to allocate goods, all have private ownership of the
means of production and all have a certain share of goods and serv-
ices that is not allocated by the market. Additionally, diversity also
seems to be stable. Pontusson (2005b: 185) claims that ‘whatever the
particular map of the varieties of capitalism approach we employ –
say Esping-Andersen’s threefold typology of “welfare regimes” or Hall
and Soskice’s twofold typology – we do not observe many instances of
fundamental “regime-change”, but we do observe many instances of
a specific institution being reformed or realigned’.8 No doubt, ample

evidence indicates that welfare states and production regimes change.
But they hardly converge on one regime type. Instead, the regime
differences that Esping-Andersen has delineated with data from the
1980s are usually argued to persist (cf. Starke, Obinger and Castles
2008; Scruggs and Pontusson 2008; Castles and Obinger 2008; Hall
and Gingerich 2009). And as Chapter 5 will show, the basic principles
behind regulation in different regimes were already discernible at
the end of the nineteenth century. Thus, similarities and differences
of capitalisms seem to be significant and stable enough to justify a
typology based on them.
More concretely, it is the relatively parsimonious models of Esping-

Andersen and Hall and Soskice that have become the most widely used
and still seem to withstand the test of time. Their country distinctions
can also be further subdivided, but unless differences within groups are
to be emphasized, for example when comparing the German and the
French welfare state, we have no pressing reason to do so. At the same
time, Germany is more similar to France than either of the two is to the
United States. Thus, the argument is that differences between regimes
are greater than differences within them. The criticism that a typology
of capitalism has to be treated with caution is well taken, yet it seems
that we are able to exert that caution in typologizing, since we routinely
exert it when classifying other objects, such as flowers.
3
This chapter integrates the empirical findings of the past chapters and
the image that emerges from the literature into a unified typology of
capitalisms. It thus takes up the thread from section 2.2. Notably, it
shows how the different typologies of the preceding chapters amount to
a common picture of capitalist diversity. So based on what the existing
literature agrees on, what does a unified typology of capitalist diversity
look like?
Complications (that the previous chapters dealt with) notwithstand-
ing, virtually all scholars see Ireland, New Zealand, Australia, Canada,
the United States and the United Kingdom as pertaining to a liberal
or market-based regime. This is the case regardless of whether scholars
looked at the welfare state or the production system. But how many
other regimes should be distinguished from this? The most parsimoni-
ous classification is that of varieties of capitalism, which labels Austria,
Germany, Belgium, the Netherlands, Switzerland, Sweden, Denmark,
Norway, Finland and Japan as coordinated market economies. Esping-
Andersen (1999) then simply splits Sweden, Denmark, Norway and
Finland out of the group of coordinated market economies and calls
these the social democratic regime, whereas he sees other coordinated
market economies and the mixed cases in the varieties typology as con-
servative. However, Esping-Andersen (cf. 1999) is not certain where to
assign the Netherlands and Japan.
What is crucial however is that Esping-Andersen distinguishes welfare
states, but still arrives at a similar grouping of countries as varieties of
capitalism, which distinguishes production systems. The only difference
is that he splits the group of coordinated market economies into two sub-
groups, one with a social democratic and one with a conservative welfare
state. If we further split Japan out of the group of coordinated-conservative
58
A Unified Typology of Capitalisms 59
countries, we arrive at Boyer’s (2004b) typology, who adds an additional

‘meso-corporatist’ variety of capitalism. However, the remaining coun-
tries (the group of ‘conservative’ or coordinated countries) are largely the
same as what Boyer calls ‘state capitalism’ in his typology.1 Amable (2003)
then simply adds a (fifth) ‘Mediterranean capitalism’, taking Spain, Italy,
Portugal and Greece out of what Boyer calls state-capitalism, and what
Amable himself calls ‘Continental European Capitalism’. However, he also
mentions that Switzerland and Norway might require a sixth type of capi-
talism (Amable 2003: 172). Strikingly, Ebbinghaus (2006) also places every
country in a group that is compatible with the prior typologies (apart from
seeing France as being part of Latin capitalism, which Amable does not).
This sounds complicated, yet it gives rise to a simple scheme, in which
fine-grained typologies only split up broader country groups of coarser
typologies, while all agree on what countries are different from others.
Figure 3.1 shows how the different typologies relate to each other.
Capitalist countries
Coordinated Market Economies Liberal Market

AT/GER/NO/BE/JA/FIN/DK/SWE/NL/CH Economies Varieties of
UK/US/CA/AUS/ capitalism
Intermediate: FR/IT/ESP/POR/TUR IRE/NZ
Conservative welfare states Social demo- Liberal welfare

AT/GER/BE/FR/IT/ESP/POR cratic welfare states Welfare state
states UK/US/CA/AUS/ research
Ambiguous when compared to grouping 1990: SWE/FIN/DK/ IRE/NZ
NL/JA NOR
State Capitalism Meso- Social-partner Market

AT/GER/BE/NL/FR/IT/ESP/ corporatist Capitalism Capitalism
Regulation
POR Capitalism SWE/FIN/DK/ UK/US/CA/AUS/
theory
JAP NOR IRE/NZ
(NOR partially)
Continental Mediter- Asian Social Liberal

European ranean JAP/KOR democratic CA/UK/US/AUS Amable’s
CH/NL/IRE/BEL/ GR/IT/ SWE/FIN/DK typology
NOR/GER/FR/ POR/
AT ESP
Centre Latin Asian Nordic Anglophone Ebbinghaus’

GER/NL FR/IT JAP SWE/DK UK/IRE/US typology
Figure 3.1 Typologies and their congruence

Source: Amable (2003: 173); Esping-Andersen (1999: 77ff.); Hall and Soskice (2001a: 20f.);
Boyer (2004b: 20f.); Ebbinghaus (2006: 76).
Figure 3.1 sums up what is intriguing about existing typologies – and

capitalist diversity more generally. The fine-grained typologies simply
refine categories of broader typologies, instead of fundamentally disa-
greeing on where to put countries. Thus, the different typologies do not
contradict each other (also cf. the table in Arts and Gelissen 2010: 575f.).
Instead, if Esping-Andersen’s typology of three worlds of welfare capitalism
is broadened in a way that his most pertinent critics suggest (cf. Ferrera
1996; Leibfried 1992; Lessenich 1994), then virtually the same classifica-
tion as Amable’s (2003) results from this.2 Pontusson (2005a: 17) uses a
similar distinction, arguing that coordinated market economies can be
split up into a continental European and a Scandinavian subgroup because
‘classificatory schemes overlap’. Basically, he argues that the broad Hall
and Soskice distinction into liberal and coordinated market economies is a
starting point. From there on, one can subdivide ‘the Nordic (social demo-
cratic) welfare states from their continental (conservative) counterparts’
(Pontusson 2005a: 27). However, Pontusson (compare especially for 2005a:
204) also has some reservations about France, Japan and Italy.
Thus, all typologies follow the argument of Ebbinghaus and Manow
(2001a: 311) that ‘there is more variation in welfare regimes and industrial
relations systems than matches the two varieties of capitalism’. However,
the difference between different typologies, delineating a varying number
of country-types, is not one of fundamental disagreement, but instead
each typology ‘reflects a trade-off between parsimony and complexity
in describing the unique features of individual cases’ (Jackson and Deeg
2006: 21). Strikingly, different typologies show the same basic picture,
only with different degrees of detail. For example, even though the typol-
ogy of Ebbinghaus distinguishes five capitalisms, he mentions that these
could also be summed up in two: ‘when we juxtapose [Center, Nordic,
Latin and Asian capitalism] with Anglophone liberal capitalism […] these
different subsystems seem to fall in the same conceptual box’ (Ebbinghaus
2006: 77, also compare how Amable falls back on two varieties of capital-
isms towards the end of his book). As another example, a tentative typol-
ogy by Hancké, Rhodes and Thatcher (2007b: 24f.) has a broad group of
conservatively coordinated countries. But then they separate countries out
of this group where the state is heavily involved in the economy. They fur-
ther separate these countries into those with a compensating state (Italy
and Spain) and those where the state interferes directly in the economy
(French ‘étatisme’). Vivien Schmidt (2000, 2002) also simply splits the
group of coordinated-conservative countries apart, since she sees France,
Italy, South Korea, Taiwan and even Japan as part of state-enhanced capi-
talism, while she sees Germany and Austria as coordinated.3
A Unified Typology of Capitalisms 61
The basic point behind this is that we can split the same image of
capitalist variety in two, three, four or five different varieties, yet these
models do not contradict each other. The typologies only differ in
that they split broader groups into finer ones (or by aggregating finer
groups into broader ones). Instead of fundamentally disagreeing on
the underlying picture of capitalist diversity, the different approaches
and typologies are ‘nested’ in each other like Russian Matryoshka dolls.
Open a country regime in one typology and you find the smaller coun-
try regimes of other typologies in them.
Not only do the different typologies fit into each other. They also fit
well with the empirical cluster analysis of section 2.1. As the typologies,
the cluster analysis showed that the broadest distinction runs between
the liberal countries and all other ones. Japan and Switzerland are two
exceptional cases in the cluster analysis. They are equally exceptional
in the typologies, where they are classified ambiguously or not at all.
The cluster analysis, as the typologies, then splits up the coordinated
countries. Existing typologies likewise agree that the four Scandinavian
countries are different from all other coordinated economies. Within
the group of coordinated non-social democratic countries, the cluster
analysis singles out Germany, which is unsurprising because most
typologies also assign it a special place as the archetypical coordinated
market economy and as the archetypical conservative welfare state.
Further, the cluster analysis splits the three Mediterranean countries out
of the group of coordinated countries, which in turn is what Amable’s
typology does and what has been suggested as a further differentiation
of Esping-Andersen’s typology. In this sense, we are confronted with a
strikingly homogeneous image: typologies of welfare and production
see the world similarly; both classify countries similarly to the empirical
analysis of section 2.1. The ever-repeated result is that on the broad-
est level, it makes sense to distinguish liberal countries from all other
ones. Then, one can subdivide non-liberal countries into those with
conservative and social democratic welfare states. Within the countries
with a conservative welfare state, one can further see Japan and south-
ern European countries as a special case of this broader category. In this
sense, both empirical data and existing typologies support an effort to
integrate existing typologies, as there is no fundamental disagreement
on country families. The only question is how many subgroups non-
liberal countries should be divided into.
Until this point, I tried to show that different typologies can be
integrated into one unified typology, since neither existing typologies
nor empirical data disagree fundamentally on how countries should
be categorized. As existing typologies indicate, all coordinated coun-

tries are different from all liberal ones, which means that a twofold
typology makes sense. But within the group of coordinated countries,
those with a social democratic welfare state are different from those
with a conservative one, which suggests a threefold typology. Luckily,
as the preceding chapters have shown, a two- and threefold typol-
ogy are not mutually exclusive. Instead, we can distinguish between
1) conservatively coordinated, 2) social democratically coordinated
and 3) liberal countries. This implies that on the one side, countries
with a conservative and a social democratic welfare state both have
more in common with each other than either of them has in common
with liberal countries. But it also implies that the Scandinavian social
democratic countries are differently coordinated from countries with
a conservative welfare state. As later chapters will show, this is not
only because of different welfare states. Also, conservative economic
coordination is different from the more inclusive social democratic eco-
nomic coordination. Chapter 5 will show along these lines that certain
policy styles influence both the welfare state and the production system
according to liberal, conservative or social democratic doctrines, so that
both the welfare state and economic coordination follows one of the
three principles within each regime. But even apart from this, it is not a
coincidence that specific welfare and production regimes coincide. It is
well documented that liberal production arrangements stabilize liberal
welfare states – and vice versa, while coordinated production systems
stabilize social democratic and conservative welfare states – and are
stabilized by them. The following chapter shows how this happens,
accounting for the concurrence of welfare states and production sys-
tems, which the preceding sections unearthed.
4
How Complementarities Stabilize
Three Capitalisms
As mentioned earlier, varieties of capitalism and welfare state research

have distinct research agendas. Put simply, they ask different questions.
Whereas varieties of capitalism analyses how firms coordinate with
their environment, welfare state research focuses on how individuals
are provided for, if not by the market. The two typologies do not ana-
lyse the same actors and do not ask the same questions. So they should
not arrive at a similar categorization of countries. But they do, as the
preceding chapters showed. The puzzling question is why. This chap-
ter sheds light on this question. It argues that conservative and social
democratic welfare arrangements support coordinated production sys-
tems and that these in turn support conservative and social democratic
welfare arrangements. Liberal production and welfare arrangements
are also mutually supportive. In other words, it is not by accident that
certain production and welfare regimes come together; instead, they
exert a causal influence on each other, which is why they come in pairs.
Complementarity thus not only exists within the production system,
as stressed by varieties of capitalism, but also between the production
system and the welfare state. Peter Hall (2007: 40) alludes to this by
mentioning that ‘social policy is a crucial adjunct to coordination’. But
no comprehensive attempt has yet been made to show how types of
welfare states and production systems support each other. This chapter
is such an attempt. It shows how liberal welfare states support liberal
production systems and are supported by them. It also shows how coor-
dinated production systems support conservative and social democratic
welfare states and are supported by them. I begin by showing how
liberal welfare states support liberal industrial relations and vice versa,
while coordinated industrial relations are supported by conservative
and social democratic welfare states.
63
4.1 Welfare regimes and industrial relations
More closely scrutinized than most other institutions of production sys-

tems, industrial relations are widely acknowledged to influence welfare
arrangements (cf. Esping-Andersen 1999: 8). Strong unions and collec-
tive bargaining, both hallmarks of coordinated capitalism, equalize the
primary income distribution, levelling the ground on which egalitarian
welfare states are built. Thus, there is a clear and strong relationship
between coordinated industrial relations and egalitarian welfare states:
Comprehensive, universalistic welfare states almost invariably go

hand in hand with centralized, nation-wide and co-ordinated bar-
gaining (Denmark, Norway, Finland and Sweden); residual welfare
states with decentralised and weak unionism (the United States
Canada, and now also the UK). […] ‘Liberal’ welfare states, especially
since the late 1970s early 1980s, exhibit low levels of union coverage,
centralization, and co-ordination (Australia being the sole exception).
The Nordic social democracies score generally high on all three meas-
ures, although there has been some decline in both centralization
and co-ordination since the 1980s, especially in Sweden. Continental
Europe presents a more mixed picture however. Coverage rates are
generally high, but there are important variations in centralization
and co-ordination. At one end, we find Austria, Germany, and the
Netherlands with strong ‘neo-corporatism’; at the other end, France
and Spain with a weaker capacity for co-ordinated collective action.
In between are Belgium and Italy.
(Esping-Andersen 1999: 17, 120;
also cf. Scharpf and Schmidt 2000a: 15)
Indeed this is largely the image that empirical data showed earlier (cf.
Figure 1.11). However, it would imply that the strength of organized
labour correlates with the form of the welfare state. Table 4.1 uses two
indicators of the historical strength of organized labour to test this: 1)
the power of a unified union movement and 2) the long-term prevalence
of collective bargaining (cf. Martin and Swank 2001: 899). For a measure
of welfare arrangements, I use Esping-Andersen’s original (1990) scores
on regime types and dummy variables for his welfare regimes.
As Table 4.1 illustrates, countries with a high union power density
were either non-liberal or social democratic. However, indicators of a
conservative welfare state do not exhibit a high correlation with indica-
tors for organized labour, which is logical. Countries with a conservative
How Complementarities Stabilize Three Capitalisms 65
Table 4.1 Correlations between historical industrial relations and welfare

regimes
Generic indicator of: Generic indicator of:

central union power collective bargaining
density – average for prevalence – average
1979–95 (Martin and for 1979–95 (Martin
Swank 2001: 899) and Swank 2001: 899)
Degree of conservatism .124 .112

for different welfare
states (Esping-Andersen
1990: 74)
Degree of liberalism –.713** –.417
1990: 74)
Degree of socialism .538* .574*
1990: 74)
Dummy – 1 for liberal –.472* –.595**
welfare regime,
otherwise 0
Dummy – 1 for social .657** .679**
democratic welfare
regime, otherwise 0
Dummy – 1 for –.089 .016
conservative welfare
regime, otherwise 0
N = 18 (data for Spain and Portugal missing).

welfare state on average neither supported organized labour as much as

social democratic countries, nor did they constrain labour as much as
liberal countries. The clearest correlation therefore exists between cen-
tral union power density and Esping-Andersen’s indicator for liberalism.
The following scatterplot in Figure 4.1 shows this graphically.
The negative connection between central union power density and wel-
fare liberalism is obvious. The following paragraphs argue that this is not
coincidental. Instead, powerful mechanisms connect organized industrial
relations and strong welfare states. But before showing these mechanisms,
more data must substantiate the correlation behind them. The two factors
that measure industrial relations above are rather out-dated (pertaining
Fitted values
Central Union Power Density – Average
for 1979–1995 (Martin and Swank 2001: 899)
AT
1.5
SWE
NOR
1
DK
.5
FIN
BE
IT
0
IRE
NL AUS
NZ UK
–.5
GER CA
JA CH
US
–1
FR
0 5 10 15
Liberalism (Esping–Andersen 1990: 74)
Figure 4.1 Central union power density and welfare liberalism
to the period between 1975–95). This is also true for Esping-Andersen’s

indicators, which measured welfare arrangements in the 1980s.
But do strong welfare states still coincide with strong organized
labour? I check this by correlating Hicks and Kenworthy’s (2003) welfare
state indicators of socialism (as opposed to liberalism and conservatism)
and OECD data about the share of trade union members of all workers,
the dominant level of wage bargaining, the strength of co-determination
and the coverage of collective wage bargaining.
The emerging pattern as shown by Table 4.2 on the following page is
mixed. Esping-Andersen’s degrees of liberalism and socialism correlate
well, significantly and in the expected direction with the share of trade
union members. The degree of conservativeness correlates with all
other three indicators of industrial relations. The liberal dummy also
correlates with the three latter indicators, again significantly and in
the expected direction. The social democratic dummy also gives a good
estimation of union members. The Hicks and Kenworthy socialism indi-
cator correlates with the share of union members and with the strength
of co-determination, whereas the indicator of conservatism correlates
with the remaining two indicators. In this sense, there is evidence that
organized industrial relations come with a social democratic or – to a
lesser degree – a conservative welfare regime.
Table 4.2 Correlations between current industrial relations and welfare regimes
Trade Union Index of dominant level of Co-determination index, Coverage

Members as bargaining: 0 company or made of indices: employees of collective
percentage individual; 25 company/ appoint board members; wage
of total plant; 50 industry; 75 employee participation bargaining –
workforce – industry and central; 100 required by law; works data for
data for central. Data for 2004 – council’s decision rights – 2000
2000 (EIRO (data for 2005), URL: data for 2002 (OECD Survey (OECD
(OECD http://www.eiro.eurofound. 2003, URL: www.oecd.org/ 2004: 145)
2004: 145) eu.int/ and OECD dataoecd/58/27/21755678.
2004: 151) pdf, page 47f.)
Degree of conservatism for different .032 .569* .472* .580*

societies (Esping-Andersen 1990: 74)
Degree of liberalism for different –.634** –.438 –.372 –.470
Degree of socialism for different .680** .209 .463 .328
Dummy – 1 for conservative –.343 .265 .433 .403
welfare regime, otherwise 0
Dummy – 1 for liberal welfare –.360 –.483* –.756** –.701**
regime, otherwise 0
Dummy – 1 for social democratic .855** .246 .363 .306
welfare regime, otherwise 0
Degree of socialism as opposed to .776** .387 .495* .462
liberalism (Hicks and Kenworthy
2003: 33)
Degree of conservatism (Hicks and .006 .519* .341 .504*
Kenworthy 2003: 33)
67
N = 17 to 20 (data for Spain, Portugal, Ireland sometimes missing).

Yet the important question remains: Why is this so? What connects
organized industrial relations to strong welfare states? One answer is
that if welfare states provide more than poor relief, the question arises
who should be eligible and how schemes should be administrated.
Organized unions have a privileged position to judge workers’ needs,
so it makes sense to let them administer welfare schemes. When they
take this crucial position in the governance of the welfare state, they
can use it to press for their demands. They can press for a strong welfare
state and economic coordination. Therefore, a good explanation for the
correlation between coordinated industrial relations and strong welfare
states is the underlying third variable of a strong and organized labour
movement, which historically not only pushed for coordinated indus-
trial relations, but also for extensive welfare states. This so-called power
resources approach will be further discussed in section 4.7.
Another mechanism that connects strong welfare states with coor-
dinated industrial relations is that the welfare state can provide social
benefits, making it difficult for companies to poach workers from each
other by using company-based social benefits. That this element of
competition between companies falls away can make it easier for com-
panies to cooperate, also by negotiating wages collectively, leading to
organized employers and a coordinated wage bargaining system, which
further removes wage competition between companies and thereby
again lowers the threat that companies poach workers from each other,
which in turn makes it easier to cooperate. This is why employers
can even cling to extensive welfare arrangements; they make strategic
cooperation between companies easier (cf. Thelen 2000; Martin and
Swank 2001).
However, it is not necessary to discriminate which of these mecha-
nisms links organized industrial relations to strong welfare states,
because they are not mutually exclusive and most likely work together.
While it must remain unclear how strong each mechanisms is, it is
obvious that there are good correlations between indicators of organ-
ized industrial relations and strong welfare states, as well as plausible
mechanisms that can explain the correlation.
4.2 Welfare regimes and corporate governance
Not only are strong welfare states and organized industrial relations
systematically linked; a similar claim can be made for welfare regimes
and certain types of corporate financing. Why is this? The high-income
stratum of a population has a greater capacity to absorb financial risks. It
69
250.0
percentage of GDP – Data for 2004, 2000 for Belgium

Overall size of the stock market in US dollars as a Switzerland
200.0
(World Bank 2006)
150.0 United States
Canada Australia
Sweden United Kingdom
Netherlands
100.0 Finland France Spain
Japan Belgium
Denmark Ireland
Norway
50.0 Germany Italy New Zealand
Austria Portugal
0.0
20.0 22.0 24.0 26.0 28.0 30.0

Percentage share of income or consumption for the
richest 10 percent of the population – Data stretching
from 1993–2000 (World Bank 2006).
250.0
Switzerland
percentage of GDP – Data for 2004, 2000 for Belgium
Overall size of the stock market in US dollars as a
200.0
(World Bank 2006)
150.0
United Kingdom
Canada United States
Sweden Netherlands Australia
100.0 Finland France Spain
Japan
Denmark Belgium Ireland
50.0 Italy New Zealand

Norway Germany
Austria Portugal
0.0
34.0 36.0 38.0 40.0 42.0 44.0 46.0

Percentage share of income or consumption for the
richest 20 percent of the population – Data stretching
from 1993–2000 (World Bank 2006).
Figure 4.2 High-income strata and market capitalization

can therefore invest in riskier and more lucrative assets, which is why it
shows a preference for stock-investments, whereas middle-income groups
prefer credit- and thus bank-based investments (cf. Deeg and Jackson
2006: 14). The availability of long-term bank-based versus short-term
stock market-based capital, on which respectively coordinated market
economies and liberal market economies rely, could thus depend on the
stratification of society. Since social democratic and conservative welfare
regimes have a more egalitarian income structure (cf. Figure 1.3), the
stock market in these countries is ‘dried out’ (cf. Vitols 2001a). Inversely,
welfare regimes with a high-income stratum (usually accompanied by
a low-income one), should show a greater prevalence of equity-based
corporate financing. This in turn should lead to shareholder-oriented
corporate governance in liberal welfare regimes as more financing is
available through shares than through banks (Deeg 2001; Höpner and
Jackson 2001; Lütz 2003; Vitols 2001b). But do countries with a rich
high-income stratum really have more equity-based financing? The
graphs of Figure 4.2 on the preceding page plot how much of a country’s
income and consumption goes to the richest 10 or 20 per cent against
the size of the stock market as a measure of a country’s GDP.
There is no apparent link. Two of the societies with the highest
share of income for the richest 10/20 per cent of the population (the
United States and the United Kingdom) also have high rates of market-
capitalization. Yet, other societies in which the upper 10/20 per cent
of the population receive a high share of incomes (Portugal and New
Zealand) score rather low. Then there is – as always – the outsider
Switzerland. There might be a hidden bias however, because it is also
overall economic development that influences market capitalization,
possibly more than income distribution. Yet, even when controlling
for GNI/person, no significant correlations emerge (not shown here).
One might nonetheless not fully want to discard the claim that a dis-
proportionately rich high-income group leads to a greater prevalence of
equity-based financing, as there might be variables that hide the effect.
However, it is definitely not an obvious connection.
4.3 Welfare regimes and skills creation
Possibly the most conceptualized link between welfare and produc-

tion systems runs between generous unemployment insurance and
economic coordination. This is because in the interpretation of the
literature on varieties of capitalism, social democratic and conserva-
tive welfare regimes support industry- or firm-specific skills, which in
turn are crucial for the functioning of a coordinated market economy

(Estevez-Abe, Iversen and Soskice 2001; Amable 2003: 96).
More specifically, strong unemployment insurance motivates
employees to invest in firm- and industry-specific skills because it guar-
antees income maintenance for workers even if technical change or an
economic downturn in the firm/industry for which skills are specific
renders skills useless. In this perspective, unemployment benefits are
insurance for workers, who in their absence shy away from learning spe-
cific, non-transferable skills, which reduce their employability. Welfare
arrangements that insure against loss of income therefore allow an incre-
mentally innovating industry to accumulate a high degree of industry-
and firm-specific skills, despite the potential fear of employees about not
possessing skills that offer them a wide range of employment possibili-
ties when they lose their job – at least this is how the story goes.
Welfare state costs that render employees less dependent on market
outcomes in this sense constitute a subsidy for incrementally innovat-
ing industries. Conversely, even if a liberal country were inclined to
invest in long-term skill creation strategies, this will not succeed, given
non-accommodating welfare arrangements (cf. Lloyd and Payne 2002).
This is because workers would refuse to learn specific skills, which are
necessary in incrementally innovating industries, if they are not insured
against job loss. In coordinated market economies however, busi-
ness associations have an interest to protect generous unemployment
insurance, since they want to protect the high skills equilibrium that
is crucial for their production system (Martin 2005: 62; Wood 2001a:
252; critically, cf. Pontusson 2005b). The welfare state must therefore
replace high wages and align unemployment and pension benefits to
them – exactly the opposite of what liberal welfare states with their
emphasis on restrictive poor relief do. Thus, skills creation also interacts
with the wage structure and decommodification. Section 4.5 discusses
this further.
Upholding an equilibrium of high specific skills is allegedly also sup-
ported by an inflexible labour market, namely through strong employ-
ment protection. For firms that teach and depend on industry-specific
skills, a non-flexible labour market is crucial. It assures them that invest-
ments in the skills of their workers will not be rendered useless because
other firms poach their workers, after they have invested into their train-
ing. In this perspective, strong employment security is not only seen as a
desire of employees for protection, but also as resulting from the employ-
ers’ interest in establishing long-term working contracts (cf. Wood 2001b:
378). Changing to another firm also becomes less tempting as collective
bargaining levels wages (see section 4.1). But do employment protection

and economic coordination indeed go together? Figure 4.3 plots the degree
of coordination for different countries, as measured by Hall and Gingerich
(2004), against the strictness of a country’s employment protection laws,
as measured by Porta et al. (2004) and by the OECD (2004). If coordinated
countries indeed protect their workforce, employment protection laws
should correlate negatively with a dummy variable for the liberal welfare
regime. Table 4.3 (after the following page) checks for this.
The scatterplots of Figure 4.3 and the correlation analysis show a
relatively clear picture. There are strong and significant correlations
between the degree of coordination of an economy and the degree of
employment protection. The scatterplots show similarly clear pictures.
All coordinated countries offer a high degree of employment protection.
There is clear clustering into a group of liberal and a group of coordi-
nated countries. However, Japan poses an exception, which is especially
true for the index by Porta et al. (2004), where it is ranked as having a
low degree of employment protection. One should also keep in mind
that the indices only measure the formal protection of employment, not
de facto (possibly informal) employment patterns. Japanese companies
combine weak formal employment protection laws with flexible con-
glomerate-internal labour markets. Therefore, employees can be sure
that their company-specific skills will serve them a lifetime, as their
company will try to reemploy them within its conglomerate-structure.
The opposite exception is Denmark, with its small-sized companies,
where unemployment protection is low but benefits are high, so that
employees can find new employment in another company without
having to fear an unemployment/poverty trap due to their specific
qualifications. The Mediterranean countries provide another exception,
as strong employment protection does not lead to a high skills/high
wage equilibrium.1
Apart from these exceptions, both the theoretical arguments and the
data are relatively straightforward. The conservative and social demo-
cratic welfare regimes, with their high job protection and wage replace-
ment rates, provide welfare state support for the high-skills/high-wage
equilibrium that is typical for coordinated market economies. Conversely,
by ensuring labour flexibility, the liberal welfare state deprives its pro-
duction system of the long-term oriented skills needed for incremental
innovation, but provides it with the flexible labour that is necessary
to compete in volatile markets. An alternative and more straightfor-
ward explanation for the co-occurrence of employment protection
and coordination is that countries that reject market arrangements by
73
Portugal
Employment laws index to measure the protection of 0.80
Netherlands France
employed workers (Porta et al. 2004: 1362f.)
Spain Sweden Finland Germany
Norway Italy
0.60 Denmark
Belgium Austria
Switzerland
0.40
Ireland
United Kingdom
Canada
0.20 United States
New Zealand Japan
R Sq Linear = 0.455
0.00 0.20 0.40 0.60 0.80 1.00

Degree of coordination for different economies
(Hall and Gingerich 2004: 14)
Portugal
3.50
Spain
restrictions on the ability of employers to lay of or fire
Employment protection (composite index of legal
3.00 Sweden
France
regular employees) – Data for 2003
Germany
2.50 Norway
Belgium
Italy
(OECD 2004: 117)
Netherlands Finland
Austria
2.00
Denmark
Switzerland Japan
1.50
New Zealand
United Kingdom Ireland

1.00 Canada
United States
R Sq Linear = 0.564
0.50
0.00 0.20 0.40 0.60 0.80 1.00

Degree of coordination for different economies
(Hall and Gingerich 2004: 14)
Figure 4.3 Employment protection and coordination

Table 4.3 Correlations between employment protection and coordination
Employment protection Employment

(composite index of legal Law Rigidities,
restrictions on the ability index by Porta
of employers to lay off et al. (2004)
or dismiss regular
employees) – Data for
2003 (OECD 2004: 117)
Degree of coordination for .751** .421

different economies (Hall
and Gingerich 2004: 14)
Dummy – 0 for liberal .752** .458*
economy, otherwise 1
N = 19 to 20 (data for Australia sometimes missing).

coordinating their economies, also hesitate to expose workers to the

vicissitudes of markets and therefore favour strict employment laws.
This highlights the importance of cultural values in the construction of
welfare and production regimes, which Chapter 5 discusses further. It is
an alternative explanation for the correlations observed here; however,
it is not mutually exclusive to the complementarities outlined so far,
as it is possible that countries established a combination of economic
coordination and worker protection for historical or cultural reasons
and then complementarities stabilized this combination.
4.4 Welfare regimes and employer associations
As mentioned above, not only unions, but also employers can have
an interest in a strong welfare state. As Wood (2001a: 250f.) states,
‘employers will look to public policy to maintain and reinforce the vari-
ety of institutional competitive advantage upon which they rely’. While
collective bargaining limits wage-competition, the welfare state can
level social benefits (cf. section 4.1). The resulting absence of poaching
by company-related social benefits can level the ground for company-
cooperation and economically efficient coordination. As already
outlined, a decommodifying welfare state can also promote a high
skills equilibrium of the workforce (cf. section 4.3). Thus, to the degree
that companies see coordination as beneficial, and to the degree that
they need a workforce with firm- or industry-specific skills, they have an
incentive to support generous and encompassing welfare arrangements

that let workers acquire these skills. By analysing regulation of labour
markets and welfare arrangements in Sweden and the United States,
Swenson (2002) shows that employer associations were indeed not the
natural enemies of welfare state expansion. Even though they rarely
initiated social policy, they willingly gave in to labour’s demands for
a more encompassing welfare state, if (and this is indeed an important
condition) this could be presented as increasing their competitiveness.
Due to the above- and below-mentioned arguments, this is more the
case in coordinated market economies than in liberal ones. Therefore,
not only should labour be more powerful in pushing for welfare state
expansion in coordinated countries, but employers should also be more
willing to give in to these demands. Martin and Swank (2001: 894) thus
argue ‘that greater degrees of business organization – the centralization
and concentration of representational power in national associations
and the level of coordination across business interest associations and
firms – should, all else being equal, promote support for social policy
initiatives’.
To understand whether strong welfare states go together with highly
coordinated employers, Table 4.4 on the following page uses Martin and
Swank’s indicators to estimate how much employers are centralized,
coordinated and how much they cooperate. It correlates this with public
social expenditure. To check not only how much, but also how money
is spent, it also includes the share of means-tested social transfers on all
social transfers; it also includes Esping-Andersen’s decommodification
index to get an idea of the effectiveness of welfare arrangements.
As Table 4.4 shows, employer centralization and coordination corre-
late with all measures of welfare state extensiveness, apart from means-
testing, to which they are negatively (as expected) but non-significantly
correlated. Enterprise coordination correlates with all welfare state
indicators. This shows that organized employers go along with more
social spending and decommodification. One possible explanation
for this is that employers in coordinated market economies are not
only more organized, but also support – or at least do not successfully
prevent – extensive welfare arrangements. Contrary to this, the weak
organization of employers in liberal market economies goes along with
low social spending, which allows for the conjecture that employers
in liberal market economies are less apt in supporting – or more apt in
suppressing – a strong welfare state. Beramendi and Rueda (2008) give
another reason why this might be so. They claim that the less social
inequality there is at the top, the more employers are similar to each
Table 4.4 Correlations between employer organization and welfare state

characteristics
Public social Percentage Decommodification

expenditure share of (Esping-Andersen
(cash benefits, means-tested 1990: 52)
provision of social transfers
goods and on all social
services and tax transfers in
breaks with per cent – data
social purposes) – for second half
data for 20012 of 1990s3
Employer .580* –.244 .630**

centralization –
average for
1979–1995
(Martin and
Swank 2001: 899)
Employer .655** –.435 .751**
coordination –
average for
1979–1995
(Martin and
Swank 2001: 899)
Enterprise .517* –.559* .558*
cooperation –
average for
1979–1995
(Martin and
Swank 2001: 899)

other so that – as they already have the same earnings – they can also
agree to pay similar wages. This means that a relatively egalitarian distri-
bution of wealth among employers leads to collective bargaining so that
‘levels of inequality in a country are the most important determinant of
coordination in the labor market’ (Beramendi and Rueda 2008: 3). The
more the welfare state levels gains at the top (one would assume with
high taxation), the more employers are willing to give in to coordinated
bargaining and organize accordingly.
Another explanation for the concurrence of strong employer organi-
zations and welfare state strength may be cultural. When countries
circumscribe the market by public spending, they might similarly mobi-

lize to contain it by inciting employers to organize more. In such soci-
eties, which are critical towards markets, organized employers might
have found ways to make do with strong welfare state spending. They
might even have learned to use it to their advantage when they could
not change it. This is discussed more in Chapter 5. While causality is
hard to pin down, it is beyond doubt that ‘for historically tractable
reasons, the institutions through which employers coordinated their
activities were often empirically deeply entangled with the genesis of
the institutions now associated with the more egalitarian models of
capitalism’ (Thelen 2012: 144). Yet, there are more concrete reasons
why strong welfare states and strong coordination go together. Notably,
another link between organized production systems and welfare states is
that both subsystems decommodify labour strongly or weakly.
4.5 Varieties of capitalism and decommodification
Scholars have long wondered why those welfare states that traditionally
provided the least protection for workers were often the ones that also
liberalized (that is, commodified labour) the most (Pierson 2001b: 435;
also cf. King and Wood 1999). This section argues that this happened
because liberal production systems are complementary to commodify-
ing welfare arrangements, while coordinated production systems are
complementary to decommodifying labour arrangements.
One reason is the ‘productivity whip’ of strong welfare states. Decom-
modifying welfare states are expensive. The more expensive the welfare
state, the more an individual worker has to pay to finance it, especially
in conservative welfare states, which are financed through payroll taxes.
Almost by definition, payroll taxes increase with the cost of the welfare
state. High payroll taxes make low-wage employment unattractive,
especially if the welfare state provides a high reservation wage. In this
sense, strong welfare states promote high-productivity jobs. Conversely,
they crowd out low-productivity employment in the private sector,
which most often relies on commodified labour (Scharpf and Schmidt
2000b, c; also cf. Chapter 6). Thus, conservative and social democratic
welfare states tend towards highly productive employment, avoiding
the low-skills/low-wage equilibrium of liberal market economies, which
relies on the commodification of labour. But they also have an inherent
tendency to produce unemployment, as only highly productive work-
ers can find a private-sector job. Liberal market economies are different.
Their welfare states do not make labour expensive. They therefore have
more private sector jobs, especially in the price-sensitive service-sector

(cf. section 1.1). But these do not rely on long-term employment rela-
tions as much as manufacturing jobs in incrementally innovating
industries do. Instead, they rely on labour being readily available in a
commodified way. In this sense, a liberal production system needs a
welfare state that commodifies labour, while a coordinated production
system needs a welfare state that decommodifies labour.
In a more fundamental way, it can also be argued that cooperation
between labour and capital, which is characteristic of coordinated mar-
ket economies, can only take place if both sides are equally strong. As
certain authors stress that the power of labour is structurally lacking
behind that of capital in capitalist countries (Offe 1984), the welfare
state needs to raise the two partners to eye-level and thus allow for
deliberation and bargaining between them.4 For the individual worker,
this can involve a significant degree of decommodification. This link is
well-developed in the literature. Hall and Soskice (2001b: 50f.) mention
that
[u]nemployment benefits with high replacement rates, for instance,

can improve the ability of firms to attract and retain pools of labour
with high or specific skills. Disability benefits and early retirement
benefits can allow firms that operate production regimes requir-
ing employee loyalty to release labour without violating implicit
contracts about long-term employment. […] For this reason, there
should be a correspondence between types of political economies
and types of welfare states. And that appears to be the case. Virtually
all liberal market economies are accompanied by ‘liberal’ welfare
states, whose emphasis on means-testing and low levels of benefits
reinforce the fluid labour markets that firms use to manage their
relations with labour.
Thus, the literature on varieties of capitalism acknowledges that

decommodifying welfare arrangements, as provided by social demo-
cratic and conservative welfare states, are an important precondition
for the equilibrium that companies in coordinated market economies
rely on, whereas strong welfare states are inimical to the functioning of
liberal market economies. That employees want to be protected against
the vicissitudes of markets seems self-evident. However, not only
labour, but also business sometimes favours decommodifying strate-
gies. Companies that rely on a highly qualified labour force will favour
a system that provides incentives to acquire the needed qualifications.
This is more the case for an unemployment insurance that replaces high
wages. Yet it is not the case for a means-tested system of poor relief
(also cf. section 4.3). Therefore, liberal countries, which already had the
most commodifying welfare states, might have become even more com-
modifying because this was complementary to their liberal production
system, which relies on cheap and readily available labour. Whereas
the competitiveness of liberal production regimes depended on the re-
commodification of workers, coordinated economies have an interest
to uphold some decommodification, not only for social reasons, but to
preserve their competitive advantage (cf. Vogel 2001; Wood 2001b).
Another way how decommodifying welfare states go together with
the organized employer associations that coordinated market econo-
mies rely on is based on control. Liberal, means-tested unemploy-
ment insurances give out low benefits that are usually tax-financed
and therefore state-controlled. Mares (2001a; b) argues that business
associations are tempted to ‘trade’ such a liberal tax-financed unem-
ployment insurance (in which they have no control) against a more
expensive scheme, that they can control somewhat.5 However, if there
are no corporatist business associations that are organized enough to
exert some control over unemployment insurance in the first place, as
in liberal production systems, the chance for more than means-tested
unemployment insurance remains meagre.6 To check whether – for
one reason or another – economic coordination and decommodifica-
tion indeed go together, I correlate Esping-Andersen’s (1990) decom-
modification index with Hall and Gingerich’s coordination-index and
a dummy variable for coordination.
Table 4.5 Correlations between decommodification and coordination
Degree of coordination for Dummy – 1 for liberal

different economies (Hall economy, otherwise 0
Decommodification .683** –.795**

(Esping-Andersen
1990: 50)
Decommodification .499* –.654 **
(Scruggs and Allan
2006: 61)

While all correlations are significant, this tells us nothing about

where the different welfare and production regimes situate themselves.
Also, it tells us nothing about the relationship between production and
the decommodification of individual workers. I therefore look again at
the average production worker from section 1.1, who still has to provide
for his family and now, to make matters worse, has become dependent
on income transfers. To check whether countries that are more coordi-
nated also decommodify their workers more, Figure 4.4 plots the net
income wage replacement rate that an average worker can expect (this
will serve as a rough estimate of decommodification) against the degree
of coordination for different countries.
As can be seen, uncoordinated countries provide low and medium
replacement wages; Switzerland, France, the Netherlands and the
Scandinavian countries provide high replacement rates, but are only
coordinated to an intermediate degree. In turn, the highly coordinated
countries only provide medium replacement rates. Apart from some
outsiders and even though there is again no linear and simple relation-
ship, there is nonetheless significant clustering according to regime types
when plotting this welfare state and the production system variable
against each other. This suggests that the theoretical arguments linking
welfare and production regimes – in this case by differing degrees of
decommodification – have at least some empirical backing.
Average net income replacement rate for fully ensured single
worker with APW wage and couple with two children and one
80 Denmark
Switzerland
APW wage – Data for 1999 (Allan and Scrugge 2002)
Netherlands
Sweden
Norway
70 France
Canada Germany
Finland Austria
Belgium
60 United States
Japan
50
Italy
New Zealand
Ireland
40
United Kingdom
30
0.00 0.20 0.40 0.60 0.80 1.00

Degree of coordination for different economies (Hall and Gingerich 2004: 14)
Figure 4.4 Replacement wage and coordination

This runs the risk of painting a picture where employers welcome

coordination and decommodification. That is seemingly at odds with
a power resources view, which contrarily stresses that both coordina-
tion and decommodification result from a strong labour movement,
and not from employers (cf. Korpi 1985, 2006). As I stressed previously
however, the two explanations must not be mutually exclusive. A power
resources view explains how decommodification and economic coordi-
nation came about (by a strong labour movement). A view that high-
lights complementarities between strong welfare states and coordinated
production systems adds to this by explaining how employers turned
constraints into beneficial constraints (Streeck 1997a) by using arrange-
ments to their advantage, which may have been forced upon them as
the result of power struggles in the first place.
4.6 Varieties of capitalism and retirement policy
The hallmark of coordinated capitalism is that it enables companies

to cooperate in game-theoretic situations, where each company has a
short-term interest to defect from cooperation, even though working
together is in the long-run interest of firms. However, coordinated
arrangements not only help companies, but also labour and capital to
cooperate. In practice, companies in coordinated market economies
are in a position to offer stable employment to their employees, for
which they can expect loyalty in return. But when companies con-
template laying off workers, they run the risk of violating the stable
employment that their employees expect of them – and based on
the promise of which they have learnt firm-specific skills. One way
to shed labour ‘without violating implicit contracts about long-term
employment’ (Hall and Soskice 2001b: 50) is to send employees into
early retirement. Therefore, early retirement often becomes the lowest
common denominator, on which unions and employers can agree to
‘mop-up “excess labour” via distinctive social policy regimes’ (Hall
2001: 59; also cf. Ebbinghaus 2001: 93f.).7 Workers will not accept this
however, if it decreases their disposable income steeply, which is why a
high wage replacement rate of the pension system becomes crucial for
the functioning of pre-retirement schemes that support coordinated
capitalism. Figure 4.5 on the following page shows whether the con-
comitant relationship between the average pension replacement rate
and economic coordination exists.
There indeed seems to be a strong connection between economic
coordination and the generosity of pension benefits. But the connection
retirement net-earning for men with average wage – Data 100

programs replacement rates as percentage of pre-
Pension replacement rate-Net mandatory pension
Austria
90 Spain Italy
Netherlands
Portugal
80
for 2003 (OECD 2005a)
Finland
Germany
70 Sweden
Switzerland
Norway
France
60 Belgium Japan
Canada
Denmark
United States
50
United Kingdom
New Zealand
40
R Sq Linear = 0.508
Ireland
30
0.00 0.20 0.40 0.60 0.80 1.00

Figure 4.5 Pension replacement rate of prior wage and coordination
is not linear; instead, countries cluster into liberal/ungenerous and

coordinated/generous ones. But the wage replacement rate of manda-
tory pension programs is a poor indicator of early retirement policy and
therefore should be treated with caution. The image that emerges in
the literature is that many coordinated market economies have indeed
used the public pension system to shed labour. But some countries, such
as Sweden and Japan, have rather used re-employment, job-rotation
and part-time work to reduce excessive labour, whereas American and
British companies used company-internal pension plans to attract and
retain skilled workers (Ebbinghaus 2001: 96). It is therefore questionable
whether economic coordination and decommodification by the pension
system really have to go together or whether there are functional equiva-
lents to a public pension system. On the following page, I check this by
using Esping-Andersen’s indicator for the degree of decommodification
by the pension system. To provide more recent data, I again include the
pension replacement rate of mandatory pension programs for 2003. I
correlate both with the degree of economic coordination and a liberal/
coordinated dummy variable.
Indeed, the indicators for a strong pension system correlate with indi-
cators of economic coordination. Countries that have more coordina-
tion tend to have more generous and decommodifying pension systems.
Table 4.6 Correlations between pension system decommodification and

coordination
Degree of Pension replacement

decommodification rate – net mandatory
by the pension pension programs
system replacement rates as
(Esping-Andersen percentage of
1990: 50) pre-retirement net
earning for men
with average wage –
data for 2003 (OECD
2005a)
Degree of coordination for .573* .713**

different economies (Hall
Dummy – 0 for liberal .541* .697**
economy, otherwise 1

It is important to warn against a functionalistic misinterpretation of this

however. The mere correlation between coordination and decommodi-
fication as well as the wage replacement rate of the pension system does
not imply that coordinated employers cheerfully welcomed extensive
welfare schemes in the area of pension financing. It can also mean that
employers in coordinated countries were less able or willing to prevent a
decommodifying pension system and only afterwards learnt to live with
it, or even used it to their advantage.
There is yet another plausible reason why generous public pensions
go together with coordination, while liberal pension arrangements
come with liberal production systems: ‘Strong public systems will tend
to crowd out private savings, thus depriving securities markets of an
important source of capital’ (Deeg and Jackson 2006: 19). The more
that people have to save for their own pensions, the more investment
capital there is for companies that use a liberal shareholder value
approach. To illustrate this, in 1995 33 per cent of UK and 25 per cent
of US stocks were owned by pension funds, whereas this was only the
case for 1 per cent of German and 2 per cent of Japanese stocks (Jackson
and Vitols 2001: 180). Companies with equity-based corporate financ-
ing focus on their shareholder-value and are therefore more short-term
oriented and in this sense liberal (Deeg 2001; Lütz 2003; Vitols 2001b).
Private pension savings therefore support liberal corporate governance.

By providing capital, private pension funds also decrease the need for
cross-company financial linkages, which stabilize company networks
in coordinated market economies, making companies less vulnerable
to take-overs by competitors (cf. Beyer 2002; Höpner 2001; Höpner
and Jackson 2001). Therefore, to the degree that private pension funds
increase the importance of equity-based corporate financing, the likeli-
hood of liberal shareholder-oriented corporate governance and hostile
inter-company relations increases.
Conversely, tax-financed or social security-contribution financed
pay-as-you-go pension systems dry out capital that pushes towards a
shareholder value orientation. Again, this does not mean that such
a link was ever consciously designed. The more sensible interpretation
is that historical compromises in one field had effects that were neither
foreseen nor even necessarily desired, but nonetheless linked certain
welfare to certain production arrangements.
4.7 Varieties of capitalism and power resources
This section stresses what is probably the most fundamental link between
strong welfare states and economic coordination. Namely, it argues that
a strong organized labour movement explains both. If organized labour
occupies a privileged position in governing the economy, it will use
this to also push for a strong welfare state and to resist its retrench-
ment. This view extends the famous claim of Walter Korpi (1985, 2006)
that the welfare state relies on the ‘power resources’ of labour. It argues
that the stronger the labour movement, the more it forges welfare and
production arrangements (cf. Hicks and Swank 1992; Lehmbruch 1984;
Martin and Swank 2001; Streeck 1992b). The following reasons can
account for this.
If a welfare state limits itself to means-tested poor relief, then chances
are high that these will be financed by taxes, which the state controls
(Mares 2001b: 55). However, more extensive welfare arrangements involve
complex questions of eligibility and necessitate knowledge of workers’
status. Such more complex arrangements are therefore more likely to
be administrated by unions (cf. section 4.1), which in turn strengthen
them. Where unions administered a system of unemployment insurance
(Ghent System), rising unemployment increased their membership, even
if it was already high in the first place (Rothstein 1992; also cf. Kitschelt
et al. 1999b). This argument follows the logic of a genie that escaped its
bottle. Once freed, it follows its own goals – and does anything not to get
back in (also cf. Olson 1984); in more scientific language, we can call it a
principal-agent problem.
Iversen and others (Cusack, Iversen and Soskice 2007; Iversen and
Soskice 2006) argue that the organized labour movement is also con-
nected to a strong welfare state through the political system. Notably,
organized labour pushes for a political system of proportional represen-
tation, which tends to enact more redistributive policy than majori-
tarian political systems. Concerning welfare state expansion, there
are therefore good arguments why organized labour (and organized
employers in coordinated market economies – cf. section 4.4) breed
social democratic or conservative welfare states. However, in liberal
market economies, there is no organized labour movement to push for
coordination or a strong welfare state in the first place.
The argument is quite straightforward concerning the extension of
the welfare state; it is more complicated for its retrenchment. Whereas
welfare state expansion consists of distributing manifest benefits that
incur diffuse costs (a higher contribution rate), welfare state retrench-
ment means cutting manifest benefits for diffuse gains (cf. Pierson
1994). Once an institution exists, it becomes a power resource in its
own right; the welfare state can therefore become its best protector,
by forming a group that relies on it and is therefore willing to pro-
tect it (also cf. Streeck 2001b: xix). Therefore, the logic ‘strength of
the labour movement determines welfare state size’ cannot simply
be turned on its head to explain welfare state retrenchment. Instead,
Pierson (1994) argues that it is the strength of groups that have gained
a vested interest in preserving the welfare state, rather than the
strength of the labour movement in general, which explains welfare
state retrenchment. However, the crucial point is that welfare state
retrenchment was resisted in policy fields where those that were con-
cerned by retrenchment could organize resistance. This should make
it more difficult to retrench the welfare state in coordinated market
economies. Consequently, Swank (2002: 41) argues that countries with
an organized labour movement were better able to mount organized
resistance against welfare retrenchment. Indeed, labour and even busi-
ness associations use power that they have in the production system
to resist, or at least adapt, welfare state retrenchment (cf. Amable
2003: 96; Anderson 2001; Crouch 2001: 115; Martin 2005: 62; Swank
2001: 208f.). An extension of Pierson’s initial argument showed that
the more labour coordinated the economy, the more it could prevent
welfare state retrenchment, because ‘[b]lame avoidance strategies of

obfuscation and decremental cutbacks might fool untutored and atom-
ized voters in fragmented, pluralist polities, but they are unlikely to
fool trade union confederations or social democratic parties’ (Myles
and Pierson 2001: 332).
Table 4.7 Correlations between organized labour and welfare state

characteristics
Public social Percentage Decommodification

expenditure share of (Esping-Andersen
(cash benefits, means-tested 1990: 52)
provision of social transfers
goods and on all social
services and tax transfers –
breaks with social data for
purposes) – data second half
for 20018 of 1990s9
Central union .435 –.232 .644**

power density –
average for
1979–95
(Martin and
Swank 2001: 899)
Collective .578* –.583* .623**
bargaining –
average for
1979–1995
(Martin and
Swank 2001: 899)
Trade union .464* –.224 .646**
members as
percentage of
total workforce –
data for 2000
(OECD 2004: 145)
Coverage of .741** –.505* .521*
collective wage
bargaining – data
for 2000
(OECD 2004: 145)
Historical power .339 .092 .213
of left parties.10

This explains why countries where organized labour coordinated

the economy retained strong welfare arrangements, even when these
were more extensive to begin with. If the labour movement takes up a
privileged place in coordinating the economy, it uses this to also prevent
a liberal model of either the welfare state or the production system.
Indeed, the strength of organized labour goes hand in hand with indica-
tors of welfare state strength. Table 4.7 on the preceding page measures
public social expenditure and, to check not only how much is spent, also
includes the share of means-tested social transfers on all transfers as well
as Esping-Andersen’s decommodification index. These measures are cor-
related with different indicators for the strength of organized labour.
Indeed, almost all indicators for the strength of organized labour
correlate with indicators for extensive welfare arrangements. This is
not due to the historical power of left parties in different countries, as
this variable correlates with none of the indicators for extensive welfare
states. Thus, it seems as if an organized labour movement indeed goes
along with broader and more encompassing welfare arrangements.
Therefore, to the degree that organized labour gains power in governing
the economy, as it does in coordinated countries, it seems theoretically
and empirically plausible that it will push towards a more expansive,
non-liberal welfare state. An additional hypothesis to explain why
welfare state extensiveness and a strong labour movement go together
might even be more straightforward and complementing the one
mentioned here. Societies where a strong labour movement enjoys
legitimacy might also grant more legitimacy to strong welfare states.
Not only would it be the case then that power resources of the labour
movement help to expand – or prevent the retrenchment of – the wel-
fare state. Additionally, it would also be reasonable to assume that in
countries that tolerate a strong labour movement, there is less pressure
towards retrenchment in the first place and more public support for
welfare state enlargement. However, the correlations above are no proof
of causation. Thus, it remains unclear whether low power resources of
unions in Anglophone countries bring about weak welfare states, if
weak welfare states undermine the power of unions, or both. In general,
using correlations to argue for complementarities has some weaknesses
that the next section addresses.
4.8 Résumé: Are complementarities the answer?
Summing up many of the above-mentioned arguments, Manow (2001c:

156) stresses that social policy can ‘force economic agents to search
for high-skill/high-wage production strategies [and] allows firms to

specialize and gain comparative advantages over less coordinated com-
petitors’. In this perspective, which I have advocated here, welfare costs
become less a matter of solidarity but of upholding the comparative
institutional advantage of a production system. The last section ana-
lysed how strong welfare states and coordinated production systems
support each other; the main ways were the following:
1. If welfare benefits exceed a minimum level, the question arises who

should profit and how. Trade unions have a privileged position to
acquire information on workers; they are therefore apt to administer
extensive welfare schemes. To do so, they need to be organized how-
ever. If trade unions organize, this can incite employers to organize
as well. Extensive welfare arrangements can therefore induce the
associational structure that coordinates an economy (cf. Manow
2001b, c). Accordingly, welfare liberalism is correlated negatively
with centralized union power (r = –0.71).11
2. Once an organized labour movement does govern the production
system, it may use this privileged position to push for a strong
welfare state and veto its retrenchment. To the degree that welfare
arrangements are seen to support competitive advantage, business
associations can do the same (cf. Mares 2001a, b). These exten-
sive welfare arrangements make coordination easier because high
and uniform social benefits by the welfare state make it easier for
companies to cooperate, as poaching by company-centred welfare
schemes becomes more difficult when these are swamped by public
benefits. An extensive welfare state thus not only results from strong
coordination, but it also levels the ground for coordination between
companies (cf. Martin and Swank 2001; Swenson 2002; Thelen
2000). Accordingly, welfare state decommodification correlates with
economic coordination (r = 0.68).12
3. Equity-based pension systems and a rich top-income stratum
increase investments in stock markets. The stock market offers firms
capital if these adopt a shareholder value-oriented corporate govern-
ance, which is characteristic of liberal market economies. Therefore,
corporate financing, an important aspect of production systems, can
result from the stratification and pension system of a society, which
welfare arrangements influence (cf. Deeg and Jackson 2006; Vitols
2001a). However, this link does not show up in simple correlations.
4. Employees rationally hesitate to learn specific skills, because these
limit their employability to certain firms or industries. But coordinated
market economies rely on specialized skills. Conservative and social

democratic welfare states therefore offer generous wage replacement
rates and employment protection to reduce the risk of learning specific
skills. Skills, which determine what production system a country can
develop, therefore depend on welfare arrangements (cf. Iversen and
Soskice, 2006; Soskice and Iversen, 2001). A welfare state that promotes
inflexible labour market arrangements pushes workers and employers
into long-term working contracts. This renders investments in skills
more rational and allows an economy to embark on a high-skill/high-
wage equilibrium, which is characteristic of coordinated market econo-
mies. In turn, it deprives liberal market economies of the flexibility that
they need (cf. Estevez-Abe, Iversen and Soskice 2001). Accordingly, the
strength of employment protection correlates with economic coordina-
tion (r = 0.75).13
5. Generous pre-retirement programmes allow employee layoffs with-
out violating loyalty that is necessary for the trust-based relationship
between management and workforce that coordinated economies rely
on. Extensive pre-retirement programmes as part of the welfare state
therefore support coordinated market economies (cf. Ebbinghaus
2001; Hall 2001). Accordingly, the pension replacement rate of coun-
tries correlates with economic coordination (r = 0.71).14
6. The high costs that social democratic and conservative welfare regimes
impose on companies enforce specialization in highly profitable pro-
duction, which usually entails more coordination than mass produc-
tion. Conversely, expensive welfare arrangements hinder production
that relies on cheap and/or flexible labour. A costly welfare state
thus promotes coordinated production. Accordingly, public social
expenditure correlates with economic coordination (r = 0.64).15
These complementarities explain why liberal welfare states go along

with liberal production systems and why conservative as well as social
democratic welfare states go along with coordinated production sys-
tems. However, it remains unclear which mechanisms are the most
powerful in linking certain production and welfare systems.
And in spite of these apparent complementarities, one has to dig
deeper. This is because complementarities can explain a system’s stabil-
ity over time. But the really interesting question is ‘what happens when
we move from an analysis of institutional effects to one of institutional
causes’ (Pierson 2001c: 7; also cf. Streeck 1999, 2002). In other words, it
is fine to show that indicators of the production system and of the wel-
fare state correlate and that in light of existing theories, it is plausible
that they do so because certain welfare states and production systems

are complementary. But it is one thing that complementarities may
explain why coordinated production systems stabilize and are stabi-
lized by social democratic and conservative welfare states, while liberal
production and welfare arrangements stabilize each other. But this tells
us nothing about where the ensuing threefold combination of 1) social
democratic/coordinated, 2) conservative/coordinated and 3) liberal
capitalism comes from in the first place. To answer this, we need to
turn away from functional explanations and get into the history of how
capitalist countries and their configurations emerged. The next chapter
does this to show that historical trajectories that lead to economic coor-
dination also lead to stronger welfare states.
5
Diversity’s Source: Three Policy
Styles, Three Capitalisms
This chapter argues that, beyond complementarities, there is a deeper

reason why empirical indicators and existing typologies suggest a
unified typology of production and welfare regimes. The argument is
that a culture and social structure that lead to a liberal welfare state
also lead to a liberal production system. In turn, a culture and social
structure that stand behind a coordinated production system also
stand behind a social democratic or conservative welfare state. In
short, types of production regimes exist for the same reason as types of
welfare regimes. This chapter shows this for each of the 20 countries.
Before, however, I will show how this argument connects to existing
approaches.
Varieties of capitalism analyses why in ‘liberal market econo-
mies, firms rely more heavily on market relations’ (Hall and Soskice
2001b: 27). Welfare state research analyses to what degree ‘a person
can maintain a livelihood without reliance on the market’ (Esping-
Andersen 1990: 21f.). At first sight it might therefore appear puzzling
why both typologies group countries in a similar way. But underlying
both typologies is the question how much societies rely on market
arrangements – respectively in their economy and their welfare system.
This, in a nutshell, explains why we can fuse varieties of capitalism
and welfare state research and why we should fuse them: each typol-
ogy analyses one side of the same coin. Varieties of capitalism analyses
how much societies use markets in the organization of their production;
welfare state research looks at how much societies use markets in the
organization of the distribution of what has been produced.
Thus, the degree to which societies allow markets to govern produc-
tion and distribution could explain why typologies of welfare states and
production systems can be combined. Either societies allow the market
91
to govern production and distribution, or neither. This would explain

why market-based production systems come with market-based welfare
states, while non-market based production systems come with non-
market based welfare states. Several research traditions suggest that this
conception, according to which common social values stand behind
the welfare state and the production system, provides the potential to
combine typologies of production and welfare regimes. Svallfors (2010:
245 – emphasis added) claims it is well established that ‘[s]upport for
equality, redistribution, and state intervention is strongest in the social
democratic regime, weaker in the conservative regime, and weakest in
the liberal regime’. Van Kersbergen (1995: 229) argues that ‘it is not so
much politics or politically generated institutional settings that mat-
ter in explaining cross-national qualitative differences in welfare-state
regimes, but rather it is cultural variables’. Accordingly, scholars claim
that popular attitudes about what is considered fair distribution are the
reason Why Welfare States Persist (Brooks and Manza 2007; also cf. 2006).
If cultural values influence welfare arrangements, they might influence
the production system similarly. Notably, Goodin et al. (1999: 5) argue
that societies prioritize differently the goals of assuring individual liberty,
social stability and nationwide solidarity. Such priorities, which stand
behind the three welfare regimes, might explain how the production sys-
tem and the welfare system are organized; they might span institutional
subfields. For example, Dobbin (1994: 3) argues: ‘As modern industrial
policies were devised, extant principles of social and political order
were generalized to the economic sphere’. Becker (2007: 280) similarly
mentions that ‘historically evolved but contested political goals such as
(more) income equality, welfare for everybody [and] participatory rights’
form the basis of different varieties of capitalism. Lehmbruch (2001: 41)
makes a similar argument, mentioning that economic regulation results
from overarching ‘hegemonic belief systems’, which he understands as
‘the cognitive, normative, and instrumental beliefs of elite decisions
makers’. What Lehmbruch calls a ‘hegemonic belief system’, Peter Hall
(1993: 279) calls a ‘policy-paradigm’, according to which
policymakers customarily work within a framework of ideas and

standards that specifies not only the goals of policy and the kind of
instruments that can be used to attain them, but also the very nature
of the problems they are meant to be addressing. Like a Gestalt, this
framework is embedded in the very terminology through which
policymakers communicate about their work, and it is influential
precisely because so much of it is taken for granted.
Diversity’s Source 93
This concept, which structures how societies organize themselves,

has various designations, ‘such as trope, frame, ideology, worldview,
typification, collective unconscious, habitus, collective representa-
tion, toolkit, and episteme’ (Dobbin 1994: 18f.) or ‘ideological setting’
(Kuhnle 1981: 135). Possibly the most useful term is introduced by
Colin Crouch (1993: 296), who refers to widely held views on how
production and distribution should be organized as a ‘policy style’.
He argues that ‘“[s]tyle” may seem a trivial variable compared with
the balance of power; an epiphenomenon if there ever was one. But it
is not to be written off in this way’, because when societies deal with
problems, they do not work out each time in the most abstract way
according to which principle these should be resolved. Instead, they –
more or less consciously – use principles of organization that worked
in the past.
This means that policy makers learn ways of dealing with social chal-
lenges and then apply established principles of organization to new
challenges. If, for example, policy makers see the market as a good way
of governing pre-industrial relations of production, they will tend to
see the market as a good way of regulating newly emerging industries,
at least till a crisis convinces them of the opposite – and possibly not
even then (cf. Dobbin 1994). Extend this argument, and it provides
the reason why welfare and production arrangements regularly follow
similar principles: If policymakers in a country see the market as a good
mode of governance, they will want to use it in governing not only
their welfare, but also their production system – and vice versa. To use
a metaphor, countries do not have multiple personalities; it is unlikely
that they use one policy style in one field and a completely different
and inconsistent policy style in another.
Adding such a concept to the literature on varieties of capitalism and
welfare state research would be a much-asked-for theoretical advance;
the possibility of doing so is alluded to by a number of scholars. Peck
and Theodore (2007: 739) claim that varieties of capitalism are embed-
ded in ‘relatively enduring sociocultural environments – “national
cultures.”’ Culture in this sense is seen as collectively held ‘basic values
and beliefs’ (Huntington 1993: 40), which stem from a common his-
tory, and thus from the collective experiences of a nation, which lead to
policy prescriptions along a ‘socially learned way of life of a people and
the means by which orderliness and patterned relations are maintained
in a society’ (Hamilton and Woolsey 1988: 71). It usually implies a ‘tacit
dimension’, meaning ‘propositions and opinions shared by a group and
so obvious to it that they are never fully or systematically articulated’
(Hirschman 1977: 69). With such a view in mind, Castles (1993c: xvii)
maintains that
[a] shared culture and language may facilitate the transmission of

policy ideas amongst the political elite and policy demands amongst
the political masses within a particular group of nations, but impede
their transmission beyond those cultural and linguistic boundaries.
A shared core of historical experience may make some policy issues
particularly salient and downgrade the importance of others. The
logic of institutional structure may facilitate certain types of policy
solutions and make others inconceivable. A common legal tradition
may shape the subsequent trajectory of the permissible bounds of
both public and private behaviour. Imperial ties long since severed
and patterns of imperial settlement may shape the institutional
legacy and influence the sequence of economic, social and political
development of countries spread far across the globe.
I build on these numerous, yet hitherto largely unconnected, ideas.

As mentioned above, I propose that countries exhibit dominant ideas,
which organize their production and distribution. To the degree that
these ideas are similar between countries, their institutions should also
be similar. For example, if in a country the view predominates that the
government should interfere in the production system to help those
without a job, there should also be a widespread believe that the state
should interfere through the welfare system, after people have lost their
job. To illustrate this, Figure 5.1 on the following page combines data
from the International Social Survey Programme (ISSP) 2006 and 2009,
showing that the more people in a society agree (or strongly agree) that
the government should provide a job for everyone (hinting to a will-
ingness to interfere with the production system), the more people also
say that it should help the unemployed after loss of a job (hinting to a
similar willingness to interfere via the welfare state).
For the (relatively few) countries for which data is available, we can
discern the familiar distinction between liberal countries (lower left
corner), which eschew market interference through the welfare state
and the production system. Then we have countries that support inter-
ference through the production systems and through the welfare state.
Sweden, Finland and Denmark are a subgroup with comparatively high
welfare support; Germany and France have relatively high support for
intervening in the economy. But all non-liberal countries support inter-
ference in markets both through the welfare state and the production
% very or somewhat: Government should provide

decent living for the unemployed (ISSP 2009)
90
DK
PT
80 FI ES
SE NO
CH
70 JP
FR DE
60
AU
50
US
NZ
40
40 60 80
% definitely or probably: Government should provide
job for everyone (ISSP 2006)
Figure 5.1 Agreement: government should provide job and unemployment

compensation
system more than all liberal countries support either of the two. What is
particularly striking is the high correlation (r=.75) between values about
market mechanisms in the welfare state and in the production system.
However, using survey-based evidence about values does not get us
very far, for we can never know whether values result from or result
in welfare and production institutions. I therefore do not argue that
culture is an ‘ultimate’ independent variable. It is itself structured by
material circumstances, such as Castles claims in the paragraph above
(also cf. Dore 1997: 27f.; Rothstein 1998: 135). Therefore, I prefer the
concept of ‘policy style’, which is less demanding about the assump-
tions necessary to sustain it. A policy style can be seen as resulting from
culture, which in turn results from material conditions, which in turn
result from culture, and so on. It is not important to find an ‘ultimate’
independent variable. Cultural values are influenced by institutions.
But even while, in the very long run, culture changes, there are, over
very long stretches of time, policy styles, which can be reasonably
explained through culture, even if this culture in turn results from insti-
tutions over long periods (also cf. ‘metatraditions’ in Sorge 2005). The
more important point is, rather, that such a concept of ‘policy style’ is
crucially lacking in the explanation of varieties of capitalism, welfare

states, and their co-occurrence, even though it is often alluded to.
Notably, varieties of capitalism has been criticized for ‘failing to explain
how the institutional equilibria that it delineates arose in the first place’
(Pontusson 2005b: 172), and for failing to ‘consider the origins of differ-
ent forms of coordination and different models of capitalism’ (Hancké,
Rhodes and Thatcher 2007b: 37). While being careful about ‘ultimate
causes’, I think these issues can be solved.
The paradox is that even though the varieties of capitalism approach
is criticized for being unclear about the origins of variety, it is wide open
for the concept of a policy style. Peter Hall mentioned that ‘the more
basic causal factor [behind regulation] seems to be differences in the
attitudes or orientations of the relevant political actors, with roots deep
in national history. The great advantage of such analysis is that they are
able to capture a range of attitudinal variables, often cultural in charac-
ter, that have especially broad effects over the political economy’ (Hall
1999: 137; also cf. 1993; Hall and Soskice 2001a: 13f.; Zysman 1994:
279). More specifically, Hall (2001: 53) argues that ‘European value
systems, reflected in the strength of Social Democratic and Christian
Democratic parties are inimical to the allocation of resources entirely by
competitive markets’. Such arguments can and have to be applied to the
study of capitalist variety. Concretely, there is reason to believe that dif-
ferent religions led to different policy styles, which in turn influenced
how the welfare and production system of countries developed. Just as
Max Weber (1965 [1905]) argued that capitalism itself resulted – filtered
through a number of factors – from the Protestant ethic, scholars men-
tion that (with the exception of Japan) three variants of Christianity
led to three varieties of capitalism and welfare states. Scattered over the
literature, one can find claims that:
1. A Calvinist ethic, prevalent in Anglo-American countries, promotes

a culture of individualist striving on markets. It thereby not only
supports a liberal welfare state, but also a liberal production system.
2. Catholicism, prevalent in continental Europe, separates society in
different groups. This not only promotes a conservative welfare
state (keeping everyone within his social group); it also promotes a
production system in which coordination stops at group-cleavages,
leading to sector-based coordination.
3. Lutheranism, prevalent in the Scandinavian countries, promotes
nationwide solidarity, which not only promotes a social democratic
welfare state, but also a nationally coordinated economy.
In this view, which builds on the work of Sigrun Kahl (2005, 2009),
Manow and van Kersbergen (Manow 2009; Kersbergen and Manow
2009; also cf. Iversen and Soskice 2006; Cusack, Iversen and Soskice
2007), religiously influenced cultures, filtered through class coalitions
and voting systems, not only influenced the welfare, but also the pro-
duction system, aligning both to a common principle within a country,
but also within a regime type. This argument also builds on the claim of
Arts and Gelissen (2010: 571) that ‘welfare states with similar patterns of
resource mobilization, dominant ideologies, and political coalitions will
have relatively similar institutional structures’. It takes this view one
step further, arguing that welfare states with similar patterns of resource
mobilization, dominant ideologies, and political coalitions will not
only have similar welfare states, but also similar production systems.
To substantiate this claim, the following chapters trace the influence of
material and cultural conditions in three groups of capitalist countries.
For each of the 20 countries discussed, I show how material and cul-
tural conditions aligned the production and welfare system to a similar
principle. I use a similar approach as Steinmo (2010) in that ‘in each of
the narratives I emphasize different factors when trying to explain how
each system evolved. This is simply because different factors mattered
in each of these cases’. Factors such as social cleavages and religion play
a role in the development of every country. Since I will cite a different
amount of evidence from different times in the different countries, I do
not aim for an exhaustive welfare and production system history for
any country. Rather, I try to illustrate in how far each country is similar
to others in its regime due to similar influences. I start with the liberal
countries, showing how each followed the ideal of liberalism, both in
its welfare and in its production system regulation.
5.1 Liberal capitalism
It is commonplace among social scientists to argue that ‘attachment to

free competition is not first and foremost a question of economic effi-
ciency, but of political culture, one might say political faith’ (D’Iribarne
1997: 162). No group of countries illustrates this better than the liberal
ones. The United States, the United Kingdom, Australia, New Zealand
and Canada (I will talk about Ireland in more detail later) share one
political culture, leading to a liberal policy style. Scholars argue that even
though the ‘policy foci of the social policy and political economy lit-
eratures differ, the ultimate source of the distinctiveness of the English-
speaking countries in both domains is seen as stemming from the same
source: policies that seek to accommodate market imperatives’ (Castles

2010: 636). If indeed these countries have a strong desire ‘to accommo-
date market imperatives’, this would explain why they all share both a
liberal welfare state and a liberal economy.
But why would these countries be more prone to accommodate
market imperatives, both in their welfare and production system? There
are good reasons why this could be the case. Calvinism and Reformed
Protestantism influenced all of them strongly (Ireland only indirectly,
by adopting British regulations). These religions promote the view that
success on markets should determine social status. Even the quite crucial
question what status your soul is to have – heavenly ascendancy or hell-
ish doom – is seen as being legitimately determined by market success
according to the Calvinist religions that predominate in these countries
(cf. Weber 1965 [1905]). Kahl (2005: 107) pointedly summarizes this: ‘suc-
cess in a worldly occupation and wealth became an absolute sign that
one was saved by God from the start, while […] strict and systematic
requirements about what constitutes a life that increases the glory of
God (e.g. personal responsibility, individualism, discipline, and asceti-
cism) made poverty appear to be the punishment for laziness and sin-
ful behaviour’. Simply put, if poverty is God’s punishment, then why
help the poor? If market-success is God’s sign of being chosen, then
why circumvent the market by a coordinated market economy or the
welfare state? The culture of these countries was therefore sceptical of
state-intervention in markets, regardless of whether production or dis-
tribution was concerned (cf. Kahl 2009: 285). Thus, these countries are
liberal in the sense that they value free exchange between (formally)
consenting adults, while state interference is seen as coercive (Goodin
et al. 1999: 41).
If indeed such a preference for unhindered market exchange exists in
these countries – for reasons that have to been shown for every country
individually, then it would be surprising if this only showed in the crea-
tion of the welfare state and not also in the creation of the production
system. The following chapters show that indeed, a distinct policy style
marks these countries, which not only influences their welfare state, but
also their production system. I will show this for each country, starting
with the birthplace of liberal capitalism, the United Kingdom.
5.1.1 The United Kingdom

All liberal countries were inspired by Britain’s liberal thought, most of all
Britain itself. Adam Smith (1993 [1776]: 22) famously argued that: ‘It is
not from the benevolence of the butcher, the brewer, or the baker, that
we expect our dinner, but from their regard to their own self-interest’. In
this conception, the economy works best under a liberal policy, where
the government merely sets the rules, so that the market can func-
tion (but cf. Smith 1979 [1759]). Closer to the area of welfare, Thomas
Robert Malthus (1766–1834) famously argued that poor relief raises
the number of paupers and thereby lowers the average living standard.
Jeremy Bentham (1748–1832) and (the young) J. S. Mill (1806–1873)
prominently advocated liberalism by arguing that the greatest happi-
ness for the greatest number could be achieved by a utilitarianism that
leaves much leeway for individual liberty. David Ricardo (1821) in turn
applied this to economic regulation. He maintained that individuals
and countries should pursue their self-interest through peaceful trade.
This backbone of British social thought was reinforced by the Calvinist
idea that ‘God helps those that help themselves’ (Weber 2002 [1905]:
115; also cf. Kahl 2005; 2009). Calvinism also brought the belief that
poverty resulted from ‘idleness and intemperance’, so that it was under-
stood as an individual, not a social problem (Ogus 1982: 161).
The United Kingdom was not only the first country to demand
that all citizens shall be free to compete on markets, unhindered by
privileges for the nobility or clergy. It also internationalized this liberal
view of how social relations should be organized. Due to its religious,
philosophical and social conditions, which all militated in favour
of individual liberty and state-abstention, the United Kingdom led
Western countries in its commitment to liberalism in the nineteenth
century (Mann 1993: 286ff.). For Goodin et al. (1999: 40), this is there-
fore where and when the liberal regime started. Its main aspects were
the same as now: free markets and residualist welfare policies. The
Speenhamland poor relief system assured that wages rose with family
size and the price of bread, to ‘make work pay’. On the other hand,
the Poor Law Amendment Act made the conditions under which one
received welfare aid so miserable, that only the truly desperate would
ask for help (cf. Polanyi 1944: 77ff.). The United Kingdom not only has
this liberal tradition in welfare matters. The same longstanding tradi-
tion of liberal thought influenced its production system, which led to
the most radical industrial innovation, the industrial revolution itself.
Scholars claim that ‘the Protestant ethic, the strength of the individual-
ist doctrine, a stable constitutional and legal system, and an open social
structure which allowed for upward mobility’ supported the innovation
and entrepreneurialism that started the industrial revolution in England
(Ogus 1982: 156). In this liberty-cherishing environment, the insight
arose early that institutionalized social policy above the level of poor
relief is incompatible with a market economy (cf. Polanyi 1944: 78).

Instead, so-called friendly societies organized relief, retarding the onset
of state-sponsored welfare schemes. These friendly societies mirrored
widespread popular resistance to ‘subject the affairs of the societies to
increased government interference. […] The evidence of British history
suggests that Victorian ideals of individualism and voluntarism could
remain strong among the people at large (as well as conservative elites)
without preventing the coming of a modem welfare state’ (Orloff and
Skocpol 1984: 734).
Britain began to build its welfare state in 1907.1 Typical for liberal
welfare policy, its 1911 unemployment insurance provided ‘flat contri-
butions to ensure that better-paid workers would not foot the bill for
their high-risk colleagues. [B]enefits were held to minimum subsistence
levels’ (Alber 1981: 154). In 1942, the Beveridge report provided a blue-
print for what was to become the modern English welfare state. Benefits
were – in line with the liberal tradition – ‘set at levels that would not
undermine a “willingness to work”’ (Lavalette and Penketh 2003: 69).
There were some exceptions to a strictly liberal policy though, such
as the 1945 National Insurance Act. Also, the National Health Service
Act of 1946 provided mostly free and tax-financed universal health
services. From the Second World War until the 1970s, social democratic
policy made its greatest inroads in Britain, following the worldwide
trend towards Keynesianism and the conviction that societies are more
productive when their populations are healthy, educated and out of
poverty. Still even at this time, Britain tried to leave as much as possible
within the scope of individual responsibility (cf. Schmidt 2000: 236f.).
Frank Dobbin (1994: 4) finds that a similar liberal and individual-
ist ideology that influenced the United Kingdom’s welfare state also
influenced its production system. The principle behind UK industrial
policy was to have individuals compete uninhibitedly. This is not only
contrary to conservative or social democratic welfare states, but also
contrary to coordinated production systems. UK law had typically
emphasized ‘strictness of property and contract rights [over] the protec-
tion of the weaker members of the community. The nineteenth century
saw the elaboration of the classical law of contract […] without regard
to the “fairness” of their content’ (Ogus 1982: 158f.). In this liberal
environment, Britain’s once fairly encompassing working movement
never integrated into a centralized organization capable of coordinating
the production system; it remained ‘a forum from which unions might
lobby parliament’ (Crouch 1993: 89; also cf. Zolberg 1986: 415ff.). That
Britain’s liberalism made it impossible for the labour movement to unite
and to organize influenced the welfare state, as ‘failed political struggles

for coordination in the early twentieth-century set into place a system
of pluralist business and labor organizations – that reinforced cleavages
in the British economy – and contributed to the emergence of a liberal
welfare state regime’ (Martin and Swank 2012: 191). In other words,
liberal thinking precluded a unified and organized labour movement,
which not only influenced the production system in a liberal way, but
also the welfare state.
When the international oil crisis of the 1970s shook the UK, it not
only reacted with liberalization of its welfare, but also of its production
system.2 As Martin and Swank (2012: 88) mention, ‘British employers
remained fragmented and pluralistic in character, and institutional
capacity for reaping the benefits of coordination remained weak.
With the ascent of neoliberalism and a succession of market-oriented
governments since the late 1970s, these historical patterns have been
reinforced and significantly deepened’. Notably, the Thatcher govern-
ment not only cut welfare schemes and instead used tax relief for
private pensions and health-care, but also promoted ‘an “enterprise
culture” to change Britain towards a self-reliant society’ (Schmidt 2000:
239). That the United Kingdom promoted the market in welfare and
production arrangements is unsurprising when one considers its long-
standing ‘ideological belief in the superiority of markets as a means
of allocating resources’ (Alcock and Craig 2001: 135; also cf. Rhodes
2000: 19). Indeed, it would be strange if a deeply ingrained belief in
the market was limited to welfare policy, while the production system
was somehow shielded from it. This still seems true. Colin Crouch
(1993: 338) claims that ‘the essential individualism of British liberalism
remains a persistent theme in the United Kingdom’. King and Wood
(1999: 25) claim that structurally similar to the welfare state, where
individuals have to rely on themselves, firms are ‘embedded in unco-
ordinated organizational contexts’. The United Kingdom still embraces
‘voluntarist industrial relations systems’ (Rhodes 2000: 20) and thus
free market principles not only in its welfare, but also in its production
system (Martin and Swank 2012: 189f., 207).
Regardless of whether we look at welfare or production arrangements,
we find the same liberal principles governing both. Even the 2008 finan-
cial crisis was perceived by many in the United Kingdom not as result-
ing from too little, but from too much regulation, so that the United
Kingdom still routinely responds to crises with a strengthening of lib-
eral market regulation (Moschella 2011: 93; Busemeyer and Trampusch
2011: 13; Konzelmann, Fovargue-Davies and Schnyder 2010).
In this sense, an individualistic view of economic freedom influenced

not only the United Kingdom’s welfare, but also its production system,
aligning both to a similar principle of liberalism. What is more, the
two systems moved in tandem. Liberal reforms of the welfare state were
accompanied by liberal reforms of the production system and vice versa.
Castles claims that liberal thought, which influenced British wel-
fare and production policy, also influenced countries that were once
Britain’s colonies, often because their inhabitants demanded the same
liberal rights that inhabitants of the United Kingdom enjoyed earlier
than inhabitants of any other nation. New Zealand, Australia, the
United States and Canada ‘largely owe their national identity to settle-
ment by the UK’ (Castles 1993b: 27, also cf. Sachs 2000: 37). These other
countries adopted British values of liberalism and individualism (Alford
1967: 71). The following sections show how liberal thought and institu-
tions moved from the United Kingdom to other liberal countries, not
only influencing their welfare, but also their production systems.
5.1.2 The United States

American settlers that came from England soon demanded similar civil
rights as individuals in England enjoyed after the Glorious Revolution.
They wanted to own property and trade freely on markets, uninhib-
ited by feudal regulations that favoured noblemen and clergy. Later
American settlers escaped repressive states to enjoy these same market-
freedoms once they were established. This led to a population that was
critical of state interference and equated free markets with freedom
(cf. Grønbjerg, Street and Suttles 1978: 14f.; Mann 1993: 486). Similar
to the United Kingdom, Protestantism influenced the American under-
standing of individual responsibility (cf. Mann 1993: 636). Poverty was
seen as God’s punishment and thus not be treated as a social problem.
Following the United Kingdom, the United States penalized the poor
with workhouses when they asked for help (Kahl 2005: 108). This
basic conviction, that poverty is an individual, not a social problem,
remained stable over time. It continuously made Americans reluctant
about social policy (Birch 1955: 44). The strength of liberal values had a
strong effect on both the organization of the economy and the welfare
state. Specifically, Orloff and Skocpol (1984: 729f., 733f.) explain the
late onset of public pensions and social insurance in the United States
(and the United Kingdom) by
the unusual strength and historical persistence of liberal values in

these nations. [Americans] emphasize the freedom and responsibility
of individuals so much that virtually any kind of collective provision

for individuals’ needs becomes suspect as ‘socialism’ or ‘tyranny’.
[T]he long-delayed U.S. acceptance of the modern welfare state can
be attributed to the persistent faith of Americans not only in rugged
individualism but also in voluntary associations rather than govern-
ments as instruments for the pursuit of collective purposes.
Goodin et al. (1999: 58) give an almost identical account, arguing

that ‘the basic ethos of US social policy has long been dominated by
rugged laissez-faire individualism and liberal notions of self-help’. Few
would question that the United States is a laggard in terms of welfare
state development. By 1930, it still had no public ‘old-age pensions,
maternity benefits, child allowances or health insurance schemes’, as its
political climate ‘was not favourable to measures which could be held to
weaken individual responsibility’ (Birch 1955: 27).
There were times however, when even the United States could not
uphold this state-critical policy style. Notably, the Great Depression
left a lasting impression that one could work hard and still not succeed
(cf. Steinmo 2010: 169). This blurred the crucial distinction between
the deserving and the undeserving poor, which had historically marked
US social policy. As a result, Roosevelt introduced national contributory
insurance for old age and compulsory unemployment insurance (cf.
Kuhnle and Sander 2010: 76; Scheurell 2003: 34). Even though this was
a watershed for the United States, it remains remarkable that it came
so late, with such low benefits and left a large part of the population
unprotected (Goodin et al. 1999: 59; Steinmo 2010: 175). Additionally,
politicians and employer organizations resisted social insurance on the
grounds that it would undermine ‘market mechanisms and the incen-
tives to individuals to make provisions for their own needs’ (Clarke and
Piven 2001: 27). Accordingly, the United States remained a laggard in
public welfare spending and coverage. It hardly regulated wages and
let its unemployment insurance drift away under the pressure of new
needs and inflation after 1935, so that benefit levels and coverage rates
remained low and stigmatizing. Overall, social policy remained ‘firmly
embedded within the poor law tradition’ (Myles 1998: 351). However,
the United States did develop a hidden welfare state, which functioned
not through direct transfers, but through less observable tax breaks,
which subsidize private welfare provision (Steinmo 2010: 152ff.). This
is striking: welfare arrangements lack legitimacy in the United States to
such a degree, that they have to be delivered ‘in stealth’, to be publicly
acceptable.
In spite of some non-liberal exceptions, scholars therefore widely

argue that liberal beliefs dominated social policy. Castles and Mitchell
(1993: 94) claim that the United States ‘eschewed interventionist wel-
fare activism as a basic tenet of its political belief structure’. Others
argue that ‘individualism, and personal salvation have historically
played central roles in shaping public welfare’ (Scheurell 2003: 28).
Hollingsworth (1997b: 134) claims that ‘Puritan traditions emphasiz-
ing hard work and achievement reinforced the belief in the USA that
one could “get ahead” by hard work and individual initiative’ as the
reason why ‘the market mentality became so pervasive in the USA, why
Americans […] have championed individualism over collective respon-
sibility’. Similar to the United Kingdom, if a policy style of favouring
market individualism affected the welfare system so strongly, might it
not also have influenced the production system?
Indeed, the United States exhibits the same liberal policy style in the
regulation of its production system as it does in the regulation of its
welfare state. Not only might early civil rights, which the United States
granted its citizens, have led to ‘fewer incentives for organizing along
class lines’ (Hollingsworth 1997b: 134). As Grønbjerg, Street and Suttles
(1978: 4f.) argue, the same liberal policy style that minimized the US
welfare state also influenced the production system, in that the United
States saw the market as self-adjusting towards an optimal equilibrium.
The United States therefore limited state interference to ‘minor changes
that are supposed to return the economy to its more “natural” state – for
example, the passing of antitrust legislation to restore “true” competi-
tion’. Tellingly, at the end of the nineteenth century, when other coun-
tries closed their ports to foreign imports, the United Kingdom and the
United States stubbornly adhered to free-market liberalism, upholding
free trade (Mann 1993: 289ff.).
Thus, the United States ‘entered the twentieth century with industrial
policies designed to guard economic liberties by preventing restraints of
trade and enforcing price competition’ (Dobbin 1994: 3). Indeed, when
economic associations sprung up at the end of the nineteenth century,
American courts declared collective price agreements as illegal with
the Sherman Anti-Trust Act. In this environment, ‘associations gener-
ally failed to stabilize output and prices, because the strong tradition
of individualism sanctioned the freedom of American firms to behave
as they wished’ (Hollingsworth 1997b: 135). In an effort to strengthen
free market competition, the Clayton Anti-Trust Act also forbade inter-
locking directorships and to acquire the stock of competing companies
(cf. Steinmo 2010: 170; Hollingsworth 1997b: 141). An overarching
structure of associations could not arise, because the United States saw
‘economic self-determination as integral to economic order’, which is
the opposite of representation through associations (Dobbin 1994: 20;
also cf. Mann 1993: 486f.; Scheurell 2003: 28). Again, social- and eco-
nomic policy moved in tandem. When, following the Great Depression,
the United States strengthened welfare provisions, it also increased eco-
nomic coordination (Steinmo 2010: 177). But this was only a temporary
exception in both fields (cf. Block 2006). Dobbin (1994: 10f.) sees the
free-market regulation of the United States as a durable policy style, so
strong that it even allows for predictions.
What kinds of policies will the United States adopt to govern a new
industry? Place your money on policies that enforce price competi-
tion […]. One needs to know nothing about interest groups, micro-
economic patterns, or state organizational resources to make these
broad predictions. One need only grasp the logic underlying current
policies to be able to guess what future policies will look like, because
policies in different countries follow fundamentally different logics.
Analogous to the US welfare state, which is invisible, by operating

through taxes, there is ‘relatively little explicit coordination’ in the econ-
omy, but some indirect coordination through tax incentives and courts,
as well as through anti-discrimination policies (Steinmo 2010: 159).
Openly coordinating the economy is publicly illegitimate and – similar
to welfare policy – has to operate in stealth. Public (and elite perception)
seems to be that market arrangements are generally superior. Tellingly
and similar to the United Kingdom, many in the United States saw the
2008 financial crisis as ‘a result of too much rather than insufficient
regulation and supervision’ (Konzelmann, Fovargue-Davies and Schnyder
2010: 23f.; also cf. Dobbin and Jung 2010: 59).
Again, regardless of whether we look at the welfare, or the produc-
tion system, we find the same liberal doctrine behind both (also cf.
Hollingsworth 1997b: 144). It might therefore be no surprise that when
the United States cut its welfare programs in the 1980s, it also deregu-
lated corporate governance, corporate financing and its industrial rela-
tions system (Hacker and Pierson 2010; Dobbin and Jung 2010). The
welfare and production system moved in tandem, following a similar
overarching policy style. Indeed, it would seem strange that a country
saw markets as a legitimate means to allocate resources in one field, wel-
fare, but would eschew markets in another field, production. All being
former colonies of Britain, a similar argument can be made not only for
the United States, but also for the remaining countries that now form
the group of liberal capitalisms. Ireland takes up a somewhat ambiguous
position however, which the next section elaborates on.
5.1.3 Ireland
With the United States and the United Kingdom, we leave the two
countries behind that are unambiguously liberal. Ireland is a hybrid of
two main influences, liberal from the United Kingdom, conservative
from the Catholic Church. This shows in both its welfare and produc-
tion system. When Ireland became an independent dominion in 1922
it had inherited the United Kingdom’s social policy: the liberal Poor
Law and policy in ‘public health, housing and educational provision’
(McLaughlin 2001: 228; Kiely 1999: 2; cf. Crouch 1993: 311; Burke 1999:
11ff.). Even after its formal independence, Ireland still looked to the
United Kingdom as a guiding model (Cousins 1997: 228f.; Ebbinghaus
2006: 58).
However, it also diverged from the United Kingdom. Ninety per cent
of its population was Roman Catholic. The Catholic Church claimed
the poor had to be taken care of and that families and communities,
but not the state, should provide welfare (Burke 1999: 26). The Irish
state refrained from active welfare provision to satisfy both influ-
ences (McLaughlin 2001: 228f.; Kahl 2005: 113f.). Ireland was not only
liberal; it was liberal in a ‘deeply conservative’ way (Burke 1999: 26).
This strange mixture, together with poor economic performance, led
to a ‘dearth of social legislation and provision’ (Kiely 1999: 2; Conroy
1999: 44). Indeed, it was not until 1933 and the Great Depression that
Ireland introduced means-tested unemployment insurance. The British
Beveridge plan influenced the country. But the Catholic Church blocked
universal health insurance, even for mothers and children; its subsidi-
arity principle spoke against it (Burke 1999: 27ff.). This conservative
legacy lasted into the 1970s. When Ireland joined the European Union,
it even tried to derogate from the EU principle that men and women
doing the same work should be paid equally (Conroy 1999: 40).
Ireland regulated its production system with a similar mix of con-
servatism and liberalism as its welfare state. Similar to social policy,
Irish economic regulation was ‘dominated by an emphasis on free trade,
limited government spending and taxation, modest state intervention
in productive sectors, and parity’ (Doherty 2001: 139). Catholic teaching
promoted ‘a society in which socio-professional groups organize them-
selves into corporations and collaborate in decision making, and in
which the independence of the family and voluntary organizations from
the state is guaranteed’ (McLaughlin 2001: 226, citing Peillon, 1982: 96).
Different from purely liberal countries, this led to a slow but more thor-
ough acceptance of unions. The 1946 ‘Trade Union Act licensed unions
for the purposes of collective bargaining’ (McLaughlin 2001: 233). Within
this relatively cooperative arrangement, Ireland used a typically liberal
policy of low taxes to attract foreign investment after 1956 (Doherty
2001: 141f.). The country became richer and in the 1970s developed
social insurance schemes above subsistence levels, causing welfare
expenditure to rise steeply (McLaughlin 2001: 235). Together with the
oil price shocks of the 1970s, this resulted in an unemployment rate
of 19 per cent and a public debt of 131 per cent of GNP (Roche 2007:
396). In 1987, Ireland started to counter its economic problems partially
by typical liberal measures, such as social cuts. Untypical for a liberal
market economy however, it secured public and especially working-class
consent for the social cuts, by negotiating them with trade unions and
employer associations (Doherty 2001: 147ff.; McLaughlin 2001: 239ff.;
O’Donnell 2004: 54f.; Allen 2000). Scholars therefore mention that
an economy that is generally described as a classic liberal market

economy has adopted some coordination strategies rather typical of
coordinated market economies. […] While the overall organization
of the Irish economy might still, by and large reflect liberal market
economy characteristics, wage restraint is now coordinated centrally
by political and economic actors.
(Kelly and Hamann 2008: 144)
Roche (2007: 397ff.) contends that unions mainly supported wage

restraint because they feared being crushed like their UK counterparts
under Margaret Thatcher. Even though this led to wage coordination
until the financial crisis of 2008, the workplace relationship between
capital and labour remained antagonistic, conflictual, and in this
sense liberal (Doherty 2001: 135f.). Also, macro-coordination was not
accompanied by workplace cooperation (O’Donnell 2004: 64). Typical
for Anglo-American countries, an extensive system of labour courts
and commissions decides on working conditions in a legalistic fashion
(cf. Teague 2009: 502ff.). During the financial crisis, Ireland’s national
wage bargaining system broke down because the unions would have
had to agree not only to wage restraint but also to wage cuts, which they
could not (cf. Baccaro and Howell 2011: 30ff.).
Overall, conservative (Catholic-inspired) and liberal (UK-inspired)
influences brought a curious mix, which, similar as its welfare system,
puts Ireland’s production system in between the conservatively coor-

dinated and the liberal countries (also cf. Roche 2007: 401). Thereby,
Ireland illustrates the basic point made here: It is not that all countries
of one regime are the same. Instead, similar structural conditions not
only influenced production, but also welfare arrangements, in this
case aligning both to a similar policy style of liberalism, with some
conservatism.
5.1.4 Canada
Canada also has a liberal policy style that influenced both its welfare
and production system. However, similar to Ireland, Canada’s liberalism
is more diluted than that of the United Kingdom or of the United States.
From the second half of the nineteenth century until the 1930s, Canada
was a laissez-faire state. It not only punished the ‘undeserving poor’ in
the style of the English Poor Law. Also, the state limited itself to provid-
ing minimal public goods necessary for the production system. The idea
behind this was that ‘the state – whether federal or provincial – should
not interfere in matters naturally governed by the invisible hand of the
market’ (Brodie 2003: 11). Similar to the United States, where welfare
legislation hardly proceeded until the Great Depression; the state’s
involvement in the welfare and production system was limited (McBride
2005: 107).
I begin by the principles that governed Canada’s approach to welfare
legislation, then I will look at the production system. Scholars often
classify Canada as a latecomer in terms of social legislation. Indeed,
it often followed the United States. For example, Canada established
industrial accident insurance in 1930, the same year as the United
States and much later than other countries (Germany: 1884, France
and Italy: 1898, Sweden: 1901). It introduced means-tested old-age sup-
port in 1927 (United States: 1935, Germany: 1889, France: 1910, Italy:
1919, Sweden: 1913) and unemployment insurance in 1940 (United
States: 1935, Germany: 1927, France and Italy: 1914, Sweden: 1934, for
all dates, cf. Kudrle and Marmor 1981: 83). Timing and content often
followed the United States, so Canadian social legislation was marked
by low benefits, market and family reliance, as well as general means-
testing (Birch 1955: 177; Myles 1996: 116; Lightman and Riches 2001:
46; Brodie 2003: 9).
Canada’s first minimum wage legislation and unemployment protec-
tion in 1935 was ruled as unconstitutional. In 1940, it started a national
unemployment program, basically emulating the British scheme (Birch
1955: 193f.). After 1943, it drew inspiration from the Beveridge report
and introduced many welfare measures that were roughly similar to the
United Kingdom’s until the 1970s (cf. Lightman and Riches 2001: 51).
In 1945, it started to pay mothers a small universal monthly benefit;
in 1952, every person over 70 got a monthly pension that was almost
enough to live off (cf. Lightman and Riches 2001: 51f.; Myles 1996: 119).
By 1966, Canada had introduced a contributory pension scheme, a social
assistance scheme and a national health insurance, which was ‘remark-
able even by European standards for the virtual absence of private provi-
sion of health care services’ (Myles 1998: 349; also cf. Brodie 2003: 15).
After the Second World War, Canada also relaxed the means-testing of its
schemes (Myles 1996: 122f.). During this time, not only was the welfare
state extended, but the economy was also more regulated: ‘[t]he new
governing philosophy […] held that economic activity should be pub-
licly regulated in order to maximize economic stability and the collective
well-being of Canadians’ (Brodie 2003: 14). After this movement towards
more public welfare provision, which was similar to the United States and
the United Kingdom, around the 1980s, and again following (this time
the neoliberal turn in) the United States and Britain, Canada started to
retrench its welfare programs. Brodie (2003: 24; also cf. Myles 1998: 346f.)
claims that this ‘asserted the primacy of the market’ over the notion of
collective responsibility. Values centred on individualism and market
justice replaced ideas of social solidarity (Lightman and Riches 2001: 45).
Thus, not only the creation, but also the retrenchment of Canada’s welfare
schemes followed the United States and the United Kingdom.
Canada’s similarity to the United States is often seen as stemming
from similar structural characteristics. Scholars claim that the ‘social
backgrounds of the two countries were essentially similar’; both had
unexploited open land and were territorially as well as socially mobile
societies (Birch 1955: 177). Some even claim the two countries have
‘most similar societies’ (Kudrle and Marmor 1981: 88). Differently from
the United States however, French settlements and French non-liberal
thought influenced Canadian thinking. So ‘while the predominant
social doctrine in both countries was that of laissez-faire, it seems to
have been less rigorously adhered to in Canada’ (Birch 1955: 178).
Basically, Canada followed the liberal approach, but less strictly than
the United States or the United Kingdom (cf. McBride 2005: 107; Myles
1996: 121). It had more extensive welfare arrangements than the United
States before the 1980s and its subsequent retrenchment was less severe
than in the United States under Reagan or in the United Kingdom under
Thatcher. Similar to its welfare system, Canada also followed such an
attenuated liberalism in its production system.
Before 1930, the Canadian state was hardly involved in the produc-
tion system. When welfare legislation was introduced during and after
the Great Depression, Canada also started to regulate its production
system (cf. Brodie 2003: 14, 18). Again, it oriented its policy towards the
United States. Block (2006: 25) mentions that: ‘The industrial relations
systems in the United States and Canada are like siblings; they have
similar origins and reflect similar values’. For example, Canada used the
US 1935 National Labour Relations Act ‘as the model for labour legisla-
tion’ (Block 2006: 26; Rogow 1993: 70). Both countries eschew the idea
that collective bargaining should influence workplace conditions or
that unions should in principle represent employees (Block 2006: 27;
Rogow 1993: 68). Hacker and Pierson (2010: 161) see the US influence
as so strong, that they label it a ‘contagion effect’. Indeed, not only did
Canada follow the United States in retrenching its welfare system in the
1980s. It also followed suit in the deregulation of its production system
(cf. Brodie 2003: 14, 18).
Yet, even though Canada’s social policy and production system regu-
lation broadly followed the liberal policy of its regime, Canada’s liberal-
ism, not only in its welfare, but also in its production arrangements, was
always attenuated compared to the United States (cf. McBride 2005). Not
only is Canada’s coverage of collective agreements twice as high (Rogow
1993: 34; Venn 2009: 17ff.). Also, it regulated its financial system more.
Konzelmann, Fovargue-Davies and Schnyder (2010: 12) argue that US
liberalism is influenced by Hayek’s ideas, according to which the state
merely sets rules for the economy, unconcerned about their outcomes
in terms of social inequality or economic coordination. Canada never
followed such a pure liberalism, so that for example dispersed owner-
ship of companies – and therefore hostile takeovers – never became as
prevalent as in the United States or the United Kingdom (Konzelmann,
Fovargue-Davies and Schnyder 2010: 17f.).
Kudrle and Marmor (1981: 98) claim that this attenuated liberalism in
comparison to the United States, which shows in Canada’s welfare and
production system, is due to a stronger influence by French settlers, so
that Canada is ‘more open and supportive to the notion of state action
for national and provincial development’. Nonetheless, the same lib-
eral ideology that one can find in the United States (and in the United
Kingdom) influenced Canada’s welfare and production system, align-
ing both to a similar principle. Second however, in both subsystems,
liberalism was attenuated in comparison especially to the United States.
Canada – similar to other liberal countries – follows a liberal policy style
in its welfare and production system, but in an attenuated version.
The same is true for the remaining two liberal countries: New Zealand
and Australia.
5.1.5 New Zealand

New Zealand was founded in 1840, when the British crown concluded
the Treaty of Waitangi between the indigenous Maori and European set-
tler. In Maori culture, social rights belonged to social groups, not indi-
viduals. In that regard, Maori culture was opposed to British liberalism
(Davey 1999: 85; Cheyne, O’Brien and Belgrave 2008: 20). In addition,
New Zealand is a small and therefore rather homogeneous country, so
solidarity was easier to organize. Different from Ireland, New Zealand did
consequently not diverge from liberal capitalism in a conservative, but
in a social democratic way. But the United Kingdom and later the United
States strongly influenced its social policy from the beginning on:
British social policy was dominated by ideas that stressed individu-

alism and the absence of state control over the economy. British
capitalism was largely being given a free hand to industrialise, and
constraints on economic activity were being reduced or eliminated.
Guilds and trade unions were outlawed, […] free trade was adopted
as a principle almost of divine revelation. These ideas were quickly
transported to New Zealand.
(Cheyne, O’Brien and Belgrave 2008: 21;
similarly, Davey 1999: 91)
Even though British liberalism thus influenced New Zealand, the

country was often a forerunner in welfare legislation. Since its settler
population was very mobile and male, care of the elderly became a
problem, since there were simply no families that could care for them
(Davey 1999: 86; Kuhnle and Sander 2010: 68). The country therefore
introduced a basic pension in 1898 as a means tested entitlement, pre-
ceding the United Kingdom by a decade. This started a commitment
to welfare provisions, which lasted until the 1970s. New Zealand’s
Social Security Act of 1938 is often depicted as the beginning of New
Zealand’s modern welfare state (Boston 1999: 3; also cf. Davey 1999:
87). It provided assistance against illness, unemployment and widow-
hood, which exceeded stigmatizing poor relief. Some scholars even see
New Zealand’s schemes of this period as an ‘attempt at universal welfare
provision’ (Higgins 1999: 273; also cf. Goldsmith and Kingfisher 2003:
330). However, benefits remained largely means tested and in this sense
liberal. A small universal superannuation was an exception to this, so
were family benefits and tax-financed medical and hospital care (Davey
1999: 87). Pensions were universal, flat-rate and income-related but not
means tested. While New Zealand is broadly liberal, some acknowl-
edged that it had social democratic universalist schemes in health and
family assistance (Boston 1999: 8; Higgins 1999: 274f.).
The same broadly liberal policy style, with social democratic elements,
is evident in the production system. While New Zealand historically
modelled its industrial relations system on the United Kingdom’s, it was
more interventionist. In 1894, New Zealand’s Industrial Conciliation
and Arbitration Act established so-called Arbitration Courts. These state-
run courts ‘set wages at socially defined levels rather than leaving them
to market forces or free collective bargaining’ (Schwartz 2000: 72f.). In
practice, if unions and employer associations failed to agree on sector-
wide wage increases, the courts imposed minimum working conditions
and approximately proportional wage gains for all earners (cf. Schwartz
2000: 78f.; Harbridge and Walsh 2002: 198). The courts’ minimum
wages were high enough to support a family of four, thus their designa-
tion as ‘social’. Court rulings also extended wage agreements to indus-
tries and occupations, sowing the seeds for compulsory trade union
membership, which was enacted in 1936. Different from other welfare
states, New Zealand influenced the primary wage distribution to achieve
its welfare goals, famously becoming known as a ‘wage earner welfare
state’ (cf. Davey 1999: 86; Arts and Gelissen 2001: 286; Boston 1999: 9;
Cheyne, O’Brien and Belgrave 2008: 23; Castles 1996; Castles, Gerritsen
and Vowles 1996).
It was not only with compulsory union membership and mandatory
wage increases that New Zealand regulated its production system more
than other liberal countries. The country also controlled ‘exchange
transactions, domestic credit, interest rates, incomes and prices, import
licenses and tariffs’ (Davey 1999: 87). Taken together, this implied more
economic coordination than in the United Kingdom or the United
States (McAllister and Vowles 1994: 387). Again, when New Zealand
showed a greater commitment to welfare measures, it also showed a
greater commitment to intervene in its economy (Schmidt 2000: 237).
However, when economic growth declined and unemployment rose
in the 1970s, not only did support for extensive welfare state arrange-
ments decline (Harbridge and Walsh 2002: 199f.). Lowered support
for the welfare state reflected a general ‘growing ascendancy of market
liberalism’, mainly imported from the United Kingdom and the United
States (Boston 1999: 3; Davey 1999: 94; Higgins 1999: 272). Starting in
1984, New Zealand not only dismantled welfare arrangements, but also
‘concentrated on financial and trade liberalization’ (Schwartz 2000: 94).

The Labour Relations Act of 1987 stipulated that, with the consent of
unions, individual enterprises could opt out of collective wage agree-
ments. While its impact remained limited, the 1991 Employment
Contracts Act halved union membership. A general believe had ‘driven
deregulation in other sectors of the economy since 1984. The belief in
the market as the mechanism to deliver the best outcomes was extended
to employment relations’ (Harbridge and Walsh 2002: 202). The end of
compulsory union membership massively increased individually bar-
gained employment contracts. As in other countries, when welfare was
retrenched, the production system was deregulated; the two systems
moved in tandem, following an overarching policy style and its change
(cf. Kesting and Nielsen 2008: 37ff.).
In the 1990s, New Zealand ‘completed the transformation to a full mar-
ket model by replacing arbitration with purely private contracts between
individuals and firms, and by replacing socially defined wages with mar-
ket wages’ (Schwartz 2000: 88). Again, similar reforms took place in New
Zealand’s welfare system (Harbridge and Walsh 2002: 208f.). The tax rate
was capped at 33 per cent. Education, health and income maintenance
became increasingly means tested and unemployment benefits were cut
(Boston 1999: 10f.; Harbridge and Walsh 2002: 209f.). New Zealand’s
social policy now reflected the idea of ‘private responsibility for economic,
social, and physical well-being, mirroring in many ways the discourse
of welfare reform in the United States, from which it got its inspiration’
(Goldsmith and Kingfisher 2003: 338; Davey 1999: 88; Menz 2005b: 59f.).
While New Zealand cut back the welfare state, it did the same with organ-
ized industrial relations (Schwartz 2000: 71). The Employment Relations
Act of 2000 attenuated some liberalization of the industrial relations sys-
tem. But by then, the preceding legislation had abolished ‘the century-old
wage-setting system, replacing it with contractual arrangements between
employers and employees which also stripped the unions of their tradi-
tional powers and status’ (Davey 1999: 89).
Overall, a common judgement is that in the 1930s, New Zealand created
a welfare state with some social democratic traits, which it then cut back
while it also liberalized its economy (Schwartz 2000: 82; Cheyne, O’Brien
and Belgrave 2008: 17). Since reforms were predicated on ‘a belief in the
market as the mechanism for delivering the best outcomes’ (Harbridge
and Walsh 2002: 199), it is not all that surprising that welfare and produc-
tion arrangements were affected, since both interfered with market forces.
New Zealand, like other liberal countries, oriented its production and
welfare system to the United Kingdom and the United States. Again, the
two subsystems moved in tandem. When one was liberalized, the other
was liberalized. A similar phenomenon occurred in Australia.
5.1.6 Australia
Australia was a British colony until 1901, starting as a penal colony
where Britain sent its outcasts. In Australia, these increasingly demanded
the same civil rights that citizens in England had become accustomed
to. They demanded the right to own and trade property on formally free
markets. Later, as in other liberal countries, Australia’s ‘British heritage
of the majority of free settlers infused the nature of the markets, chari-
table organizations and informal support systems’ (Bryson 2001: 66).
This influence of British liberalism aligned Australia’s welfare and pro-
duction system to a common principle. Yet, Australia is in some ways
exceptional in the liberal regime. In 1910, it had the world’s first labour
government. It introduced a federal means-tested, non-contributory old
age pension as early as 1908 and invalidity-pensions in 1910 (cf. Kuhnle
and Sander 2010: 68). Public pensions were generous and tax-financed;
they catered to about 30 per cent of an age cohort and redistributed
income (Bryson 2001: 75). In 1912, Australia was also the first country
to introduce maternity benefits. In these aspects, it was a social policy
forerunner (cf. Castles and Mitchell 1993: 93f.).
Similar to New Zealand, Australia instituted a ‘wage earner’ welfare state.
This started with the 1907 Harvester Judgment, establishing the right of
male workers to a ‘living wage’, high enough to support a family of four
(Schwartz 2000: 76; Bryson 2001: 67). As in New Zealand, arbitration
courts set wages at socially defined levels (Schwartz 2000: 72f.). The effect
of this approximated sectoral collective bargaining, since employees who
did the same work received the same wage, regardless of their employer. It
also reduced pressure to redistribute through welfare measures, since pri-
mary income was already fairly evenly distributed. Even though Australia’s
(and New Zealand’s) welfare states were smaller and more targeted than
Europe’s, workers were more protected and decommodified than in the
United Kingdom or in the United States (McAllister and Vowles 1994:
387). Australia relied on social wages even more than New Zealand. In fact,
they constituted Australia’s main social policy (cf. Castles and Mitchell
1993: 93f.). However, even though this created a relatively egalitarian sys-
tem, it is not fundamentally different from the minimum wage typical for
liberal countries, even if Australian social wages were generally on a higher
level. Conversely, workers and employers could not be brought to pay the
payroll taxes that European social security schemes rely on; they perceived
them as pay cuts (Bryson 2001: 73).
Similar to most liberal countries, welfare state development took off

after the Great Depression. In the 1940s, Australia introduced pensions
for widows, as well as sickness and unemployment benefits (Bryson
2001: 73). After the Second World War, Australian citizens enjoyed more
social protection than those of any other liberal country (Birch 1955:
236). By the 1970s, Australia still tried to move its welfare state in a more
social democratic direction. It introduced social assistance in 1973, tied
pensions to the inflation rate and abolished means-testing for every-
one over age 70. It also developed a national health system in 1976,
albeit not as universal as the United Kingdom’s (Bryson 2001: 74ff.).
Whereas from the 1940s onwards, the British Beveridge report influ-
enced Australia, beginning in the 1980s, the tide turned, as ‘neo-liberal
economic theories initially from the USA’ started to dominate the politi-
cal discourse (Bryson 2001: 66). Inflation had shrunk the neglected
public pensions, which now had to be complemented by private ones
(Schwartz 2000: 80).
When Australia started to retrench its welfare state, it also liberal-
ized its financial markets (Bryson 2001: 78; Konzelmann, Fovargue-
Davies and Schnyder 2010: 18ff.). At the same time, it deregulated its
industrial relations system and encouraged bargaining at individual
and company levels (Bryson 2001: 76). Since labour and employers
were fairly centralized, this could be negotiated with them (Schwartz
2000: 71). Overall, Australia followed the United States in liberalizing,
but was not as radical as New Zealand, the United Kingdom or the
United States itself. Its corporate governance system also retains some
stakeholder elements (Mallin 2010: 299ff.). Similarly, in the welfare
state, there was clearly a ‘greater society-wide commitment to equality
and security […], even though both antipodal countries also shared
Britain’s underlying liberal emphasis on individual responsibility’
(Schmidt 2000: 238).
Strikingly, when Australia strengthened its welfare state, it also regu-
lated its production system more. When it liberalized its welfare state
in the 1980s, it also liberalized its production system. Production and
welfare policy moved in tandem, as one overarching policy style influ-
enced both. This overarching policy style saw markets as a legitimate
way to govern social and economic relations.
5.1.7 Summing up liberal capitalism

The United States, Ireland, Canada, New Zealand and Australia inher-
ited a tradition of liberal thought from the United Kingdom. Given that
they share a policy style that, going back to Adam Smith, advocated that
‘social atomization is the prerequisite of perfect competition’ (Smith

1979 [1776]: 232f.), these countries have broadly similar arrangements
in their welfare and production system. Castles (1993a: 30) refined this
idea in his ‘Families of Nations’ approach:
Because these nations were linked to Britain […] by ties of power,

trade and cultural transmission of ideas, their patterns of economic
and political development – the preconditions for their subsequent
strategies of state intervention – were decisively different from those of
nations that embarked on modernization in a later era. In other words,
quite irrespective of whether the formative influences were historical
sequences of events or culturally transmitted ideas and attitudes, the
consequences of either or both shaped behaviour patterns and/or
became the directing principles of institutional arrangements.
For this reason, British liberalism influenced how these countries

thought about welfare and production, aligning both to a similar
policy style. States in this group therefore have a tradition of seeing
markets as a legitimate means of regulating transactions. They tried
to ensure that everyone has the freedom to earn and spend as he or
she pleases. So they not only abolished what remained of their guild
system (Crouch 1993: 314f.). They also saw public welfare systems as
essentially undesirable, even if they had to concede that they offer an
unavoidable way of introducing human rights and equality of chances
(as opposed to equality of outcomes) into economic and social life.
While these countries agree that poverty somehow has to be dealt
with, social policy or economic coordination as a desirable end in itself
is largely foreign to their dominant policy style. All these countries
experimented with social policy and economic coordination after the
Great Depression. But during the 1980s, their liberal tradition made it
easy for them to not only cut welfare arrangements, but also to liberal-
ize their production system in a way that was inconceivable in non-
liberal countries (Castles 1993b: 6, 10). Scholars therefore claim that
the ‘liberal principle of non-intervention into the market has put its
mark on social policy and labour relations’ (Ebbinghaus 2006: 76; also
cf. Goodin et al. 1999: 243). At first, an orientation towards Britain was
responsible for this liberal bias in both institutional subfields. Starting
in the 1970s, US neoliberalism inspired these countries (Hacker and
Pierson 2010: 160).
This does not mean that liberal welfare arrangements invariably bring
liberal production arrangements or vice versa. But how both subsystems
mirror the same liberal principles is as striking as their ensuing institu-

tional homogeneity. And this is not only true for their static equilibria,
but also for their direction of change. When countries liberalized their
welfare system, they also liberalized their production system. When
countries liberalized their production system, they also liberalized their
welfare system. This is actually not all that surprising. If a country
exhibits a strong preference for liberal policy in one field, why should
this general way of thinking not also affect another field? For exam-
ple, a Protestant market ethic should influence welfare and produc-
tion arrangements by its central doctrine. In this sense, countries do
not suffer from ‘multiple personality disorder’. Their arrangements in
one sphere tend to be complemented with coherent arrangements in
another sphere, as both reflect an underlying policy style, which in turn
results from a country’s culture and social structure.
The liberal countries therefore now reap the benefits that are typical
for following the prescriptions of neoclassical economics. Tellingly, all
six of them rank under the top 16 countries with the highest economic
freedom, as measured by the economic freedom index 2011 (Heritage
Foundation website, data from 2011). Thus, the main benefit of these
countries is that investors encounter few inconveniences. They cannot
get the same production cheaper elsewhere, as poor countries cannot
(not yet?) specialize in computer and biotechnology or financial services.
Apart from their industrial sector, the laissez-faire attitude of these coun-
tries allows a dynamic private service sector that thrives because taxes,
and especially payroll taxes, are low. Production is centred on goods that
require flexibility and benefit from free market arrangements.
This brings a high private sector employment rate. The flipside of
this is a high wage-spread and a workforce that has to be very flexible.
In summary, the positive attitude of these countries towards markets
affected their welfare and production arrangements. This is why all
liberal market economies also have liberal welfare states. Countries that
are less purely liberal than the United States or the United Kingdom
also document this through their production and welfare system. It is
in both subfields that Ireland diverges in a conservative way, while New
Zealand was more social democratic. This coincidence of welfare and
production regulation is why varieties of capitalism and welfare state
research should not be treated as two separate typologies. Rather, the
same variable underlies both; they both measure how much countries
rely on the market. However, while there is considerable heterogene-
ity between the liberal countries, it is even more so for conservatively
coordinated countries.
5.2 Conservatively coordinated capitalism
Notwithstanding important differences, Austria, Belgium, France,

Germany, Italy, the Netherlands, Portugal, Spain and – to a lesser
degree – Switzerland and Japan belong to this regime-type. Different
from liberal countries, these nations have social solidarity; different
from social democratic countries, it does not extend to the national
level, but is bound to social groups. Each of these countries has strong
social cleavages, which divide society and unite members of social
groups. For example, parliaments in these countries were often ‘strictly
divided along status-group lines, the overriding concern of each of the
individual chambers which composed such assemblies was to protect
and, if possible, extend group-specific privileges’ (Ertman 1997: 20f.;
also cf. Hinze 1970a [1930]). As Miller (1999: 31) argues, the social
segmentations within these countries affect the scope of redistribution,
because ‘distributive justice presupposes a bounded world within which
distributions take place: a group of people committed to dividing,
exchanging and sharing social goods, first of all among themselves’.
These countries not only segmented social security along their (often
occupational) cleavages, which gives them a conservative welfare state.
They also organized their production system along occupational lines,
which makes them coordinated. Thus, through both their welfare state
and their production system, these countries maintained ‘an organic-
hierarchical social order inherited from the past’ (Myles 1998: 344). As
the following sections will show, they are conservative, both in their
welfare state regulation and in their economic coordination.
State-building was complicated in these countries as a yet weak state
had to emancipate itself from a strong church. All these countries – except
Japan – are directly influenced by Catholicism. This distinguishes their
policy from liberal or social democratic countries, since ‘Catholic politi-
cal mobilization was a response to the threat of liberalism and socialism’
(van Kersbergen 1995: 31). On the one side, ‘social Catholics loathed the
treatment of workers as commodities’. On the other side, these countries
rejected state involvement, as they worried that this might replace the
family as the basic building block of society (van Kersbergen 1995: 187).
Flora and Alber (1981: 43) therefore argue that in ‘the religiously mixed
and Catholic countries the tradition of private charity and the principle
of subsidiarity, giving priority to the responsibility of smaller collectivi-
ties, remained strong’. Catholic social thought was also characterized
by the notion of ‘almsgiving without discrimination between and judg-
ments about the poor [and] relieving poverty without systematically
enabling people to overcome it’ (Kahl 2005: 114). Following this doc-
trine, the main thrust of Catholicism was to stabilize social structures.
It fulfilled this role for about 1,000 years during the Middle Ages. The
central belief behind trying to keep everyone where he is, is that indi-
viduals should not try to find fulfilment in this world anyways, as it is
only a transitory stage to afterlife, where true fulfilment waits.
Additionally, these societies’ strong social cleavages meant that
‘nation-building had been rendered difficult by religious and cultural
heterogeneity’ (Crouch 1993: 336f.). Long and expensive wars desta-
bilized these countries – with the exception of Switzerland (Knudsen
and Rothstein 1994: 203). Apart from Japan, they were all involved in
the continuous struggle for power between European nation states. In
this continuous social turmoil and under the influence of the Catholic
Church, the overarching aim of these countries was to stabilize the
social order (cf. Mann 1993: 281). Palier (2010d: 37f.) argues that this is
an enduring trait, so that stability and security is still what populations
and elites in these countries aim for. Indeed, the dominant policy style
of these countries never tried to fundamentally change the income dis-
tribution, but to secure people’s position within it, making these coun-
tries deeply conservative in trying to guarantee social peace. Fittingly,
Goodin et al. (1999: 55) describe this regime type as essentially con-
servative, because: ‘Its basic goal is to preserve the existing social order,
and the existing pattern of distributions within it’.
The stability-biased policy style behind the welfare state spills over
into how these countries regulate their production system. For example,
ensuring ‘[j]ob security for the core male breadwinner […] is a welfare
value that also spills over into the work realm’ (Schmidt 2000: 267f.).
Indeed, many of these countries allowed organized labour and business
only to the degree that they sustained political and social stability (cf.
Crouch 1993: 301). Social partnership was ‘to address the issue of the
social and political integration of industrial workers […] and a guaran-
tee of social peace’ (Palier 2010a: 604). In this way, the goal behind a
conservative welfare state is mirrored by conservative economic coor-
dination. This means that the social groups, which these countries pre-
serve through their welfare state, also coordinate the economy, through
occupational collective bargaining, to cite one example (cf. Palier
2010a: 607; Goodin et al. 1999: 51). Thelen (2010: 195f.) calls this form
of economic coordination ‘group based’, which she distinguishes from
the ‘solidaristic coordination’ of the Scandinavian countries. However,
there is considerable variety within this regime type, even more than
in the two others. The following sections show how conservatism and
coordination evolved in tandem in these countries, reflecting an under-

lying policy style that aims towards social stability. I start to show this
with the prime example of a conservative welfare state and a coordi-
nated market economy: Germany.
5.2.1 Germany
Germany was unified late (in 1871) under Prussian leadership. Its cleav-
ages were numerous and overlapping. They ran between regions (North
and South), religions (Catholics and Protestants) and, cross-cutting other
cleavages, between bourgeois liberals and lower classes (Ebbinghaus
2006: 60; Mann 1993: 321f.). Conservative German Chancellor Otto
von Bismarck pioneered social insurance to protect industrial work-
ers against sickness (1883), industrial injury (1884), disability and old
age (1889). Lutheranism and Catholicism inspired him (cf. Kahl 2005:
115). Different from Scandinavian countries, Bismarck’s policy did not
reflect or aim at widespread solidarity. Instead, he wanted to prevent a
working-class revolution and to stabilize the young state by building
‘worker loyalty towards the imperial regime’ (Kuhnle and Sander 2010:
65; Baldwin 1989: 6, 23; Manow 2001a: 110). Different from social
democratic policy, social benefits were tightly coupled to prior earnings
in the German welfare state. Different from liberal policy, the system
was dictated from above and compulsory, so that it was actually ‘a radi-
cal break with liberalism’ (Kuhnle and Sander 2010: 65f.). The basic aim
of this system was ‘to maintain an individual’s relative position in the
social hierarchy’, thereby conserving the social order (Bode 2003: 158;
also cf. Leisering 2001: 172; Mann 1993: 675). Fittingly, to the degree
that industrialists feared social uprisings, they supported the new social
insurance (Bueck 1905: 791ff.; Baldwin 1989: 5).
The conservative welfare state that emerged from this made work-
ers dependent on a stable working career, as only this could guarantee
the steady contributions, on which benefits were calculated. This gave
employees an incentive for ‘cooperation and coordination in the work-
place, which in turn supported the evolution of a specifically German
production regime’ (Streeck 2001a: 31f.; Ebbinghaus 2006: 70; Manow
1997: 33; 2001a: 114). Because the welfare state promoted stable
employment in this way, Germany’s ‘[s]ocial policy and industrial rela-
tions are interwoven and act in tandem’ (Bode 2003: 158).
Interestingly, Germany’s economic coordination is segmented along
similar social cleavages as its welfare state. It stems from the same
‘authoritarian alliance to restrain socialism’ (Martin and Swank 2012:
126). Because economic coordination in Germany was segmented along
the social groups that composed German society, it does not reflect
national, but group-based solidarity. As Martin and Swank (2012: 111)
argue: ‘Germany’s failure to rise above sectoral coordination, rather than
the fact of coordination itself, may well reflect the deep divisions in the
economy’. German coordination is segmented into occupations, regions
and industries (Goodin et al. 1999: 72). Different from the Scandinavian
countries, with their nationwide solidarity, redistribution and national
coordination in Germany are rare, to the degree that ‘no institutional-
ised tripartite or bipartite economic and social council exists at national
level’ (EIRO 2009f: 8). While the welfare state documents little national
solidarity in the form of redistribution, associations tend to be similarly
unable to agree nationally (cf. Bispinck and Schulte 2000; Trampusch
2009a: 132ff.; Hassel and Schiller 2010).
While Germany did not adopt a neoliberal course in the 1980s, its con-
servative welfare state and coordinated production system came under
pressure afterwards. As long as they could, ‘German leaders expressed
a normative preference for the status quo’ (Cox 2001: 491). But under
increasing pressure from low-cost countries and with a service sector
that was unable to shoulder Germany’s high labour costs, the Hartz
IV legislation finally diverged from the status stabilization of previous
policy (cf. Palier 2011: 45). While this happened, the industrial relations
system was fragmented as well (Baccaro and Howell 2011: 21ff.; Menz
2005a; Streeck 2009). The tricky part about these changes is that they
have often taken place under an institutional structure which – from
the outside – appeared relatively stable, but from which ever more
firms defected (Hall and Thelen 2009: 20). However, segmentalization
of industrial relations is actually what should be expected for a country
that has always had a group-based segmented production system, owed
to solidarity contained in social groups. Indeed, current reforms in wel-
fare and production, which stabilize the inner core of manufacturing
workers, while the fringe around them expands, are exactly what we
would expect from a conservatively coordinated country, which does
not aspire to egalitarianism, but tries to protect social inequality as it is.
As Martin and Swank (2012: 209) argue, such ‘dualist tendencies [are]
embedded in the logic of sector coordination’.
At the same time, liberalization in the welfare and production sys-
tem does not go as far as in liberal countries. German companies still
have a more stable stockholder structure than Anglo-American ones.
Associations still negotiate wage agreements for their fragmented mem-
bers, guaranteeing wage-stability for some years. Most companies still
aim for long term, if somewhat lower and less volatile profits than
Anglo-American ones (cf. Glassmann 2009). It also seems as if organized

industrial relations stopped declining, with collective trade agreements
covering about 50 per cent of all employees (cf. Ellguth and Kohaut
2011: 245). While the Anglo-American growth model has fallen into
disgrace in 2008, many German reforms actually reinforced the segmen-
talist makeup of the country, especially strengthened short-term work
(Kurzarbeit) and cooperation at industrial and company level (Soskice
2009: 134f.).
The basic goal of German social and economic policy was and is
to ensure a stable social structure and stable economic transactions
(cf. Goodin et al. 1999: 73f.). Germany was and is therefore an inher-
ently conservative country, both in its welfare and in its economic
coordination. Streeck (1997b: 53) calls it an ‘all-too-prudent’ model. The
idea is that everyone should stay where they are in society. Both the
state and the market have the potential to change this. So historically,
German policy not only showed distrust of the state, but also of the mar-
ket (Manow 2001a: 112; cf. Mann 1993: 322). This conservative stability
bias sets Germany apart from liberal and social democratic countries.
Mann (1993: 353) claims that the German route to modernity was
exemplary for other countries in continental Europe. They applied
similar conservative measures to stabilize their societies and economies
under unfavourable circumstances (Kuhnle and Sander 2010: 64). The
following sections show how conservative policy influenced production
and distribution in other countries, starting with the case most close to
Germany, and possibly even closer to a typical conservatively coordi-
nated country: Austria.
5.2.2 Austria
What is true for Germany is truer for Austria. Predominantly Catholic
and segmented into regions and languages, Australia was historically
a multi-ethnic state, especially before the First World War (cf. Mann
1993: 330ff.). Faced with deep social divisions, it was hard to envision
universalistic welfare policy and nationwide economic coordination.
This does not mean that social policy was late or weak. But it was
socially segmented, just as the coordination of the economy.
In 1887 and following Bismarck, the Austro-Hungarian Empire intro-
duced compulsory industrial accident insurance. It introduced compul-
sory sickness insurance for industrial workers in 1889, and compulsory
pensions insurance for white-collar workers in 1906. In 1920, it was
the third European country to introduce compulsory unemployment
insurance and extended compulsory pension insurance to blue-collar
workers in 1927 (Alber 1981: 156; Flora and Alber 1981: 59). But the
segmentations of these social programs mirrored social cleavages, which
ran even deeper than in Germany. In 1934, they even erupted into
a civil war. Eastern workers fought mainly (Western) capital-owners
(Hemerijck, Unger and Visser 2000: 187f.). On the one side, the country
desperately needed a policy to bind worker loyalty to the state. On the
other side, the solidarity necessary for nationwide redistribution did
not exist between a people fighting a civil war. Thus, Austria’s welfare
arrangements were and are divided between occupations and strictly
oriented towards replacing prior wages. They stabilize an individual’s
position in the social hierarchy and in consequence stabilize the social
order itself. They also tie worker loyalty to the state, from which work-
ers receive their pension and status-conserving welfare benefits (cf.
Hemerijck, Unger and Visser 2000: 186ff.; Katzenstein 1984: 64f.; Mann
1993: 341).
Austria allowed the labour movement to the degree that it stabilized
state and society. Redistribution, which could change the social order,
is and was not an aim of the Austrian labour movement (Katzenstein
1984: 36; EIRO 2009a: 5). This laid the groundwork for an associational
structure that could coordinate and stabilize economic transactions
(Crouch 1993: 149). In this, Austria was again very similar to Germany.
Both countries historically aimed to ‘integrate most people into one
corporation or another’ (Sorge 2005: 129f.). In these social groups,
redistribution and economic coordination take place. This contributes
to and relies on another form of solidarity than the Scandinavian
countries. Instead of promoting national solidarity, it conserves people
within their social group and their economic association, not allowing
upward, but also preventing downward mobility. What it does allow for
is social and economic security and stability (Busch 1993: 36; also cf.
Katzenstein 1984: 30, 44). This conservative policy style, keeping people
where they are, shows not only in German but also in Austrian welfare
and production arrangements (cf. Katzenstein 1984: 38). Austria may
even be more conservative than Germany (cf. Crouch 1993: 14). Even
though it has a central union confederation, its collective bargaining
reflects social divisions, as negotiations about working conditions are
mostly sectoral (Hemerijck, Unger and Visser 2000: 194; cf. Katzenstein
1984: 61f.).
Recently, Austria has – again much like Germany – liberalized its
welfare and production system. It reduced social insurance benefits,
encouraged private pension savings, tightened eligibility criteria for
unemployment benefits and promoted private health-care payments
(cf. Palier 2011: 45f.). Parallel liberalization took place in industrial rela-
tions. However, the country is still a far cry from liberal ones in both its
welfare and production arrangements (EIRO 2009a: 3; Kelly and Hamann
2008: 144f.). It is also different from the Scandinavian countries in spite
of its high degree of economic coordination however, because coordina-
tion is – similar to the welfare state – not nationally universal. The social
cleavages that are underlying this not only prevent national economic
coordination, but also universal social policy, which could lead to national
redistribution (Katzenstein 1984: 78). Overall, Austria has a similar social
cleavage structure as Germany, resulting in a conservative welfare state
and conservative economic coordination, stabilizing individuals within
social groups, but not redistributing or coordinating between them. With
different shades, this conservative coordination – strong within, but weak
between social groups – also characterizes other countries that had strong
cleavages between, and strong solidarity within, social groups.
5.2.3 Switzerland
Switzerland is unusual. On the one hand, it is similar to other con-
servatively coordinated countries, as it is historically marked by cleav-
ages between languages, religions and cantons. This would lead one to
expect a conservative policy, keeping everyone in their well-entrenched
group, while preventing nationwide solidarity (cf. Schmidt 2000: 268f.).
On the other hand, Switzerland shares traits with liberal countries.
Calvinism is strong, while labour and social democracy are weakly
organized (Trampusch 2010b: 59). Following these two influences, the
country’s production and welfare system is marked by a strange mix of
liberalism and conservatism (Lipset and Rokkan 1967: 42; Katzenstein
1984: 28; also cf. Obinger 2009). This shows in both the welfare and the
production system.
Around the turn of the nineteenth century, the Swiss state supported
employer-run training schemes and helped establish a peak association
for workers (Crouch 1993: 84; Trampusch 2010a: 212f.). It also helped
establish an employer confederation, so that Switzerland developed
an early capacity for economic coordination, which was hindered by
social segmentation however (Crouch 1993: 89ff.). Notably, employers
were split between large and small companies and labour was divided
by ‘cross-cutting linguistic, religious, rural-urban and citizen-foreigner
cleavages’ (Trampusch 2010a: 213; also cf. Katzenstein 1984: 112ff.).
Encompassing coordination was possible during moments of national
unity, such as in 1937, when the country had to stand united against
fascist Germany, so that Swiss employers and unions reached a peace
agreement. However, typical for a conservative country, social segmen-

tation prevented national bargaining in the style of the Scandinavian
countries (Oesch 2007: 338; Trampusch 2010a: 214).
A similar trend characterized Switzerland’s welfare state. While
the Swiss welfare system was not developed on the national level,
Switzerland’s cantons experimented with compulsory unemployment
insurance as early as 1895. Cantons helped to finance schemes, but
these were mostly run by unions (Trampusch 2010a: 213). After 1917,
the federal state subsidized the union-run schemes (cf. Alber 1981: 152;
Trampusch 2010a: 202). Until the 1970s, still trailing behind other
countries on the national level, Switzerland was more similar to liberal
than to conservative countries (Trampusch 2010b: 59; also cf. 2010a:
202; Alber 1981: 153, 156; Katzenstein 1984: 110).
Switzerland’s schemes are still an odd mix of liberalism and conserva-
tism. Even though it is mostly German-speaking, it has a less gener-
ous welfare state and a less regulated labour market than Germany or
Austria (Bonoli 2003: 179). However, Switzerland cannot be described
as liberal either. It is a strange mix. The pension system provides a basic
pension, but also uses payroll contributions and tax deductions for pri-
vate savings. A similar mix of logics exists in unemployment insurance,
which is very conservative in compensating up to 80 per cent of former
wages, compulsory, and financed by employee and employer contribu-
tions. On the other side, social assistance is highly stigmatizing, provid-
ing meagre benefits and even being regarded as a loan in some cantons
(Bonoli 2003: 183f.). An equally unusual mixture can be seen in health
insurance. Switzerland ‘still constitutes an exception in Europe for not
having a public health insurance scheme or a national health service’; it
instead relies on regional schemes (Bonoli 2003: 184). Thus, Trampusch
(2010b: 58) describes Switzerland’s modern welfare policy as post-liberal
or, in other words, as unclassifiable. Some scholars see it as conservative,
in that it ‘uses private provision along with the conservative aims of
status maintenance and the protection of the breadwinners’ earnings’
(Bonoli 2003: 179f.; also cf. Katzenstein 1984: 109).
The state’s minor role in the provision of welfare ‘matches Switzerland’s
traditional preference that the government should play a minor role in
the economy’ (Katzenstein 1984: 107). Indeed, Switzerland’s production
system shows the same mix of liberal and conservative elements as its wel-
fare system. Some see Switzerland as a typical coordinated market econ-
omy (Hancké, Rhodes and Thatcher 2007b: 29; Hall and Soskice 2001a:
18). Others acknowledge its liberal traits, mainly weak labour and decen-
tralized industrial relations, while they also acknowledge cooperation
between capital and labour (Katzenstein 1984: 29; Trampusch 2010b:

62). Also, Switzerland mixes dominant sectoral collective bargain-
ing with low levels of employment protection for low-skilled workers
(Bonoli 2003: 194). The Swiss corporate governance system similarly
has elements of both liberal and market economies (Börsch 2007:
174). But then again, Switzerland is similar to typically conservative-
coordinated market economies in having ‘decision mechanisms that
overarch and connect the different levels [of labour relations and]
facilitate strategies of long-term cooperation and conflict resolution’
(Busch 1993: 36).
Summing up Switzerland’s ambiguous arrangements, Katzenstein
(1984: 124) coined the term ‘liberal corporatism’. A mixture between
conservative and liberal elements is not only typical for Switzerland’s
welfare, but also for its production system. While the production sys-
tem seems to become more liberal, the welfare state seems to become
more conservative. Possibly, both systems will converge on a mid-
dle ground between liberalism and conservatism (Trampusch 2010b:
69; also cf. Oesch 2007; Bonoli 2003: 194). What is important in the
context of the argument presented here is that a similar conservative-
liberal policy style can be observed in the area of welfare and produc-
tion regulation. However, that being said, Switzerland is the country
that is the least compatible with any scheme.
5.2.4 Belgium
When Belgium was founded in 1830, hopes were high that it would be a
unified country. However, Belgium was and is marked by cleavages that
are typical for a conservatively coordinated country. It is split in two. In
the northern part live the Dutch-speaking Flemings, who are Catholic
and were agricultural. The South is the Walloon part of the country,
French-speaking, secular and historically industrial (Lipset and Rokkan
1967: 42). Political parties, trade union confederations and health
insurance funds were segmented along these mutually reinforcing
socio-economic, regional and linguistic cleavages. The central state had
limited power and had to adapt economic coordination and the welfare
system to these existing cleavages (Pasture 1993: 708; Hemerijck, Unger
and Visser 2000: 189f.; Schmidt 2000: 268).
Belgium therefore developed a segmentalized welfare state, with a
strong Catholic influence. It pioneered the famous ‘Ghent system’ of
unemployment insurance, where unemployment payments and ben-
efits are redistributed within trade unions, which in turn are organized
along social cleavages (cf. Pasture 1993: 698). In 1920, the Belgian state
subsidized these schemes, leaving control and management to the respec-

tive trade union however (Alber 1981: 153). Sickness and invalidity were
insured similarly: mutualities from the different unions collect fees and
distribute benefits for their members, preventing nationwide solidarity,
instead limiting it to social groups, so that these are preserved, again
resulting in conservative social policy (Pasture 1993: 699).
Central for Belgium’s subsequent development was a moment of
national unity. In 1944, under German occupation, trade unions,
employers and the state concluded the ‘Draft Accord for Social Unity’.
Even though trade unions and employer associations never officially
ratified it, it influenced post-war social policy by making unemploy-
ment, health, and invalidity insurance compulsory and based on earn-
ings (cf. Flora and Alber 1981: 59). The Belgian government in exile
found no support for a more universal, tax-financed Beveridge-style or
social democratic system. Contributions and benefits thus continued to
be channelled through trade unions, which were split along social cleav-
ages, requiring no nationwide solidarity and in turn preventing it in the
future (cf. Pasture 1993: 696f.). This exceptional moment of cooperation
in welfare legislation was mirrored by an instance of strong economic
coordination. With the ‘Draft Accord for Social Unity’, employers and
unions recognized each other’s legitimate authority in economic matters
and started to coordinate their actions (cf. Pasture 1993: 701; Arcq and
Pochet 2000: 115; Hemerijck and Marx 2010: 139). While cost contain-
ment since the 1970s oriented social insurance towards needs, instead
of prior contributions, trade unions still administer the unemployment
insurance (cf. Hemerijck and Marx 2010: 140ff.; OECD 2009: 76; Van
Rie, Marx and Horemans 2011: 128). As benefits are still calculated on
previous wages, the system remains conservative (cf. Hemerijck, Unger
and Visser 2000: 185ff.; Böckerman and Uusitalo 2006: 301). Most
importantly, social transfers behind this conservatism still reflect social
cleavages, because unemployment and early retirement benefits are ‘dis-
tributed by the payment bodies of the three trade union confederations,
organized along the traditional ideological pillars of Belgian society’
(Van Rie, Marx and Horemans 2011: 130).
Notably, trade unions are split between a socialist, a liberal and a
Christian branch, overlapped by differences between a Walloon and a
Flemish branch (EIRO 2009b: 3f.; Hemerijck and Marx 2010: 154). In
this sense, social cleavages that mark social insurance also mark the
production system. While Belgian trade unions are among the strong-
est outside Scandinavia, they are ‘integrated with (and divided among)
separate political parties, and they must constantly perform a delicate
balancing act between Belgium’s two linguistic communities and

regions’ (Hemerijck, Unger and Visser 2000: 182). The same cleavages
that lead to a conservative welfare state, thus also prevent nationwide
economic coordination (Crouch 1993: 89, 118; Hemerijck and Marx
2010: 129; Hemerijck, Unger and Visser 2000: 182). However, in spite of
Belgium’s social divisions, which affect welfare and production arrange-
ments, the general argument presented here also has serious limitations
for Belgium. Despite the country’s social divisions, it manages to extend
intersectoral agreements to virtually all employers and employees, lead-
ing to almost complete coverage (cf. EIRO 2009b: 3f.). This shows that
the social divisions of a country influence its welfare and production
system along similar lines, but do not determine it. However, few would
dispute that the social and cultural cleavages that marked the Belgian
welfare state in a conservative way also fragmented economic coordina-
tion. In this sense, a common policy style again extends its influence
on both subsystems.
5.2.5 France
France is not a clear-cut case. It is conservative because its pension and
unemployment schemes establish proportionality between payments
and benefits, keeping people where they are socially. However, France
deviates from other conservatively coordinated countries as its state,
instead of associations – or, broadly speaking, social groups – coordinates
welfare schemes and production. What made the French state so power-
ful? Similar to other European states, France fought out a fierce conflict
with the Catholic Church at the end of the nineteenth century about
state-funding of churches and the control of education. This resulted in
the 1905 law on the Separation of the Churches and the State. Other
South European states were unable to gain such a decisive victory against
the authority of the church (cf. Crouch 1993: 302f.). Resulting from
this, the state did not have to share authority with the Catholic Church,
nor with other any other groups that could have intervened in welfare
schemes or economic coordination. The French state thus instituted
social insurance schemes (unemployment in 1905 and 1914, old age
pensions in 1910 and 1930) that were not run by associations, but cen-
trally administered by the Ministry of the Interior. This led to a ‘preemi-
nence of public action over any private forms of welfare provision’
(Manow and Palier 2009: 151f.; Alber 1981: 153; Kudrle and Marmor
1981: 83; Flora and Alber 1981: 59; Schmidt 2000: 269). However, the
insurance-based system still reflects a Bismarckian tradition in that it
keeps everyone in their place socially through proportionality between
payments and benefits (cf. Levy 2000: 318f.). In this regard, France is no
less conservative than other countries. The aim of social legislation was
to stabilize the social order. What distinguishes France from other con-
servatively coordinated countries is the degree of state involvement in
welfare schemes (Palier 2010b: 73f.). So while aims were similar, means
were different. The state not only administered welfare schemes, but
also coordinated the economy, reflecting deeply held beliefs that only
Paris could bind the nation together:
The French polity located sovereignty in the central state, as the

only force that could orchestrate political order and hold the nation
together. […] French policy aimed to guide major manufacturing
and infrastructural sectors from above, on the principle that only the
state can prevent self-interested entrepreneurs and market irration-
alities from disrupting progress.
(Dobbin 1994: 2, 4; also cf. Boyer 1997: 97)
The early Le Chapelier laws banned ‘all combinations of economic

groups’. Strong associations that could coordinate the economy were
consequently never formed and industrial relations remained con-
flictual (Crouch 1993: 303, 77, 141). Even though the French state
basically built a national employers’ association after the First World
War, ‘France’s corporatism […] has always been more apparent than
real, with state authorities making all critical decisions’ (Levy 2000:
312; Baccaro and Howell 2011: 15). The state controlled the produc-
tion system through nationalized banks, conglomerates and public
subsidies (Ebbinghaus 2006: 66). It also coordinated wages through a
minimum wage, to which 40 per cent of all wages were linked and by
imposing wage agreements with unions that extended to entire sec-
tors (Hall 2007: 49; Levy 2000: 320; EIRO 2009e; also cf. the classical
studies by Shonfield 1965; Hall 1986). Basically, the economic coordi-
nation that associations undertook in other countries, orchestrating
finance, technology transfers and skill-creation, was taken up by the
state in France (Hall 2007: 50; also cf. Boyer 1997: 78; Busemeyer and
Trampusch 2011: 21).
The important point is that it is not in some institutional subsystem
of France that a policy style of strong state involvement can be found;
instead, state involvement influenced all institutional subsystems of the
production and welfare system. Some scholars even maintain that this
uniformly strong state-involvement sets France apart as a distinct vari-
ety of ‘étatist’ capitalism (Hancké, Rhodes and Thatcher 2007b: 24f.).
However, the French state propagates economic coordination along

sectoral lines and administers conservative welfare schemes, so France
can be reasonably regarded as conservatively coordinated. The coor-
dinating role of the state has come under increasing pressure through
European integration and the European Commission (cf. Levy 2000).
But the state still uses loopholes to coordinate the economy. It promotes
industrial champions internationally, while protecting them at home.
It increasingly uses regulatory instruments rather than direct interven-
tion however (cf. Thatcher 2007: 163; Maclean and Harvey 2008: 219).
Thus, while French welfare schemes and actors that could potentially
coordinate the economy are fragmented (mostly along occupational
lines), the state makes up for much of this. It administers conservative
welfare schemes and coordinates the economy. It thereby aligns both
subsystems to similar principles of conservatism, as state-coordination
in welfare and production stabilize economic transactions and the
social order.
5.2.6 The Netherlands

Using empirical indicators and established typologies, the Netherlands
is hard to categorize (cf. Chapter 2). This is due to the ambivalent policy
style of this country, which has its roots in Dutch society. On the one
side, the country is famously ‘pillarized’. Catholics, Protestants and
secular workers each had their own union, media and political party (cf.
Ebbinghaus 2006: 60; Kersbergen 2009: 124f.; Lipset and Rokkan 1967).
Social life was structured through these social pillars, because everyone
was ‘to be a member of one and only one of those pillars; all of one’s
associations were supposed to be within one and the same pillar; all
of one’s needs were supposed to be catered for within the institutional
structure of that pillar’ (Goodin et al. 1999: 64).
Among those cleavages, the Catholic-Calvinist one was ‘the single
most important feature of what constitutes the nation’ (Kersbergen
2009: 121). Due to its Calvinism, the Netherlands had some liberal
traits, such as workhouses (cf. Kahl 2005: 107). While Catholics and
Protestants, but also socialists and liberals, asked for sovereignty to decide
about social policy within their group, they had no interest to extend soli-
darity beyond it (cf. Trampusch 2009b: 293; Lijphart 1968). Different
from social democratic countries, it is difficult to organize national
solidarity in the face of such cleavages. Different from liberal countries,
solidarity exists within social groups however, and social policy has to
do justice to this. The logical result is conservative social policy, which
stabilizes social groups and people’s positions in them. Indeed, the
Netherlands used such conservative social policy to stabilize its social

order:
Social policy functioned as a central means to mediate social and

political conflicts. The politics of social welfare in the Netherlands has
been an indispensable constituent of consociationalism and therefore
of the development of a stable democracy in a segmented society.
(Kersbergen 2009: 133)
In 1901, the country introduced compulsory industrial accidents

insurance; compulsory pension and sickness insurance were only imple-
mented in 1930 (Van Leeuwen 1997: 782f.; Flora and Alber 1981: 59). In
1916, the Netherlands was the fourth country in Europe to provide a vol-
untary but subsidized national unemployment insurance (Alber 1981:
153; Trampusch 2009b: 303). When unemployment insurance became
mandatory in 1952, sectoral associations decided on contribution levels
and benefits, so that these differed between economic sectors, creating
social policy that could hardly be more conservative (Trampusch 2009b:
312). The pillarization of Dutch society not only affected its welfare
state in a conservative way. Economic coordination was segmented
along the same social cleavages. Liberals founded their trade union in
1871, socialists in 1892, Catholics in 1893 and Protestants in 1909. Each
of these unions claimed to represent exclusively the interests of ‘their’
constituency. The same was true for employers, where a nationwide asso-
ciation could not have represented members with a liberal background,
alongside those with different religious backgrounds (Trampusch 2009b:
306f.). Since Dutch labour unions were split along religious and ideologi-
cal lines, Dutch labour at the beginning of the twentieth century had
‘produced organizations, but there is little evidence that these engaged
seriously in whatever fragmentary industrial relations activity was in
progress’ (Crouch 1993: 118, 89). To cite a practical example, when the
country tried to build a nationally coordinated training system, the
Catholic union demanded Catholic vocational training schools, while
the socialist union demanded state-run vocational schools (Trampusch
2010a: 208). Bargaining remained mainly at the firm level. The state
tried to pacify social segmentations and passed a law in 1937 that
empowered the Ministry of Labour to extend industrial agreements
to all workers of a particular industry (Trampusch 2010a: 208; EIRO
2009h). From then on, collective bargaining slowly reached the sectoral
level (cf. Hemerijck, Meer and Visser 2000: 259). Yet, compared to social
democratically coordinated countries, ‘Dutch industrial relations were
organized slowly and remained relatively uncoordinated and decentral-

ized’ (Trampusch 2010a: 208). This is because social cleavages not only
influenced the welfare state, but also the scope of economic coordina-
tion in the production system.
By using payroll taxes, income replacement remained ‘targeted at the
(male) breadwinner in order to safeguard traditional family patterns’,
so that the Netherlands upheld a conservative welfare state (Hemerijck,
Meer and Visser 2000: 259). But when social pillarization decreased after
the Second World War, this made more universalist social policy possible:
‘the sort of solidarity implicit in the older consociationalism [solidarity
contained in social groups] tended to generalize into the sort of solidar-
ity across society as a whole which is the ideological hallmark of social
democracy’ (cf. Goodin et al. 1999: 66; Trampusch 2009b: 313). When the
welfare state moved in a social democratic direction, economic coordina-
tion followed. After the Second World War, the Netherlands established
the Social-Economic Council, where employers, trade unions and the
government met and decided on nationwide coordination. When society
became more unitary, so did the trade union movement, the socialist
and Catholic union confederations merged (EIRO 2009h: 4). Especially
well-known and influential was the 1982 Wassenaar accord, where the
leading representatives of employees and employers agreed to allow
wages to increase more slowly than productivity. Also in the 1980s, the
Netherlands developed a school-based and therefore state-led vocational
training system, the management of which was taken over by the social
partners when their divisions decreased (Busemeyer and Trampusch
2011: 31). The leading employer representatives agreed to reduce weekly
working hours and thereby to share work among more employees, to
renew collective agreements and to abstain from widespread layoffs
(Hemerijck, Meer and Visser 2000: 262f.). This national approach is the
sort of concertation that we would expect of Scandinavian countries.
It came along with decreasing social divisions. At the same time, and
this is due mainly to the unitary political system of the Netherlands,
the government was able to unilaterally impose liberalizing measures
without having to take the social partners on board (cf. Schmidt 2000:
284ff.).
The Netherlands is thus a case, where – similar to other conservatively
coordinated countries – social segmentations produced a conservative
welfare state and sectoral economic coordination. Thus, structural
characteristics that influenced the welfare state also influenced the
production system, aligning both to a common principle of con-
servative coordination – coordination within, but not between social
groups. However, when social cleavages – for example, the structural

preconditions for conservative social policy and sectoral coordination –
decreased, the country became more social democratic. This again was
evident in its welfare and its production system. Thus, the Netherlands
is not a clearly conservatively coordinated country, but has some ele-
ments of social democratic coordination. It is striking however, that the
welfare and production system were regulated in tandem, following an
overarching policy style, which slowly changed from conservatism to
social democracy, reflecting underlying structural and cultural charac-
teristics, which influenced both subsystems.
5.2.7 Italy
In a fundamental sense, Italy was and is similar to other conserva-
tively coordinated countries. It is divided by ethnicities, ideologies and
regional differences. It also had to endure a strong Catholic influence
(cf. Crouch 1993: 304). The Catholic Church sees the family as the basic
unit of society. In Italy’s south, this led to an ‘amoral familialism’, as
Edward Banfield (1958) famously called it or, in Robert Putnam’s (1993)
words, to a lack of social capital. People were unwilling to extend soli-
darity beyond their most limited social group: their own family. Yet in
spite of these cleavages, Italy somehow had to establish a functioning
state and social stability (cf. Lipset and Rokkan 1967: 42f.; Schmidt
2000: 269).
This led to conservative social policy when Italy adopted ‘Bismarck’s
programs in Germany as a direct model’ to construct a welfare state
that conserves social positions, instead of redistributing (Lynch 2009:
97; Fargion 2001: 183ff.; Kudrle and Marmor 1981: 83; Flora and Alber
1981: 59). This basic architecture still shows in that Italy protects its
core labour force and the elderly, while social investments, such as
‘family benefits, social services, and housing benefits are underdevel-
oped’ (Maino 2003: 200; Fargion 2001: 184). For example, Italy has
a generous state-pension scheme, while it still has no national social
assistance (Fargion 2001: 189; Kahl 2005: 113). In this sense, and typi-
cal for Catholic social policy, the Italian welfare state preserves social
status, but – different from social democratic countries – does not help
people to improve their social situation. Instead, it stabilizes social
groups, mainly families, through social security schemes (Jessoula and
Alti 2010: 157; Fargion 2001: 185f.; Goodin et al. 1999: 52). At the
same time, the welfare state relies on the family for the provision of
social services (Karamessini 2008: 46). An exception to this is the health
system, which is universal since 1978 (Fargion 2001: 189f.).
If such a strong conservative character affected Italy’s welfare system

(by keeping everyone where they are socially), it might also affect the
production system. Indeed, mirroring cleavages that prevented a more
universalist welfare system, Italy’s industrial relations system was histor-
ically marked ‘by a division between socialist and Catholic unionism’ so
that Italy’s unions remained fragmented (Crouch 1993: 118). Industrial
relations remain segmented along similar social cleavages as social
policy, namely through occupational, but also ideological and regional
divisions (Karamessini 2008). Scholars claim that social divisions cre-
ated through ‘the existing labour market and welfare state regimes pre-
vent both employers and unions from creating a cross-class, pro-reform
coalition’ (Molina and Rhodes 2007a: 825). So what can a state do with
a society that is so segmented, with solidarity hardly extending beyond
families? As far as the welfare state is concerned, it can rely on families;
as far as the economy is concerned, it can support family enterprises.
This is what Italy did. The state protected small and medium-sized
family enterprises by tolerating tax evasion and irregular employment
alongside a protected core workforce (cf. Piore and Sabel 1984; Regini
1997: 116; Trigilia and Burroni 2009: 633ff.). There are some excep-
tions, where Italy was capable of larger economic coordination how-
ever. While Italy proved unable to muster a concerted response to the
inflationary pressures of the 1970s (cf. Negrelli 2000), it was during a
moment of exceptional national unity, namely when all of Italy agreed
to join the European monetary union, that ‘[i]ndustrial relations coop-
eration and concerted welfare reform’ was possible (Molina and Rhodes
2007a: 805 – italics added; Schmidt 2000: 298ff.). Yet, while national
concertation seems to remain possible under special circumstances, such
as joining the European Monetary Union or during the financial crisis of
2008, Italy’s industrial relations system remains fragmented (EIRO 2009g:
3; Michelotti and Nyland 2008: 188f.).
Italy’s welfare state and its production system reinforce each other by
following a coherent logic. Italy’s industrial relations system and labour
legislation provide strong protection for insiders, while excluding
outsiders (Michelotti and Nyland 2008). The resulting long job tenure
allows for long-term commitments, which supports long-term coordi-
nated production. On the other side, the same regulations keep outsid-
ers from regular employment, deepening the division into social groups,
which the welfare state promotes (Maino 2003: 209; Fargion 2001: 192;
Kahl 2005: 113; Palier 2011: 47). The policy style that spans the welfare
state and the production system seems to be that in both fields, the
state promotes families and is only rarely able to bring cooperation to a
higher level. But mainly, what marks Italy is a policy style where pro-
duction and distribution are based on small-scale social groups, above
which coordination and solidarity do not extend, so that both social
policy and coordination are not nationally universalistic, but conserva-
tive, limited to supporting family structures.
5.2.8 Spain
Similar to other countries in the conservative group, but different from
most social democratic or liberal countries, Spain’s history is marked by
social unrest. This left Spain in need of social stability, so that ‘every
political shift has been characterized by the preservation of the previ-
ously existing institutional design and by the expressed will of policy-
makers to enlarge it and make it more efficient’ (Guillén 1992: 120).
However, even if such a conservative desire existed, it was still difficult
to indeed achieve social stability, as Spain was – and is – marked by
cleavages along regional and linguistic identities (Catalan, Basque and
Castilian), so that identification with one’s social group was high, while
national identification was weak. This obviously makes it difficult to
construct a stable state (cf. Lipset and Rokkan 1967: 42; Linz 1967: 198).
These cleavages, plus a desire to stabilize the social order in spite of
them, plus Catholicism, influenced the welfare and production system
along similar lines.
The Spanish welfare state started with the introduction of compul-
sory old age insurance in 1919. The next steps were maternity benefits
in 1926 and compulsory labour accident insurance in 1932, which
left many risks uncovered however (Guillén 1992: 121). During this
early welfare expansion, the production system also became more
coordinated. In 1925, the dictator Miguel Primo de Rivera formed an
alliance with socialists and their trade union, the Unión General de
Trabajadores (UGT), after he had made enemies among traditional con-
servatives, the Church, the liberal bourgeoisie, and the anarchist and
communist movements. In exchange for their support, he instituted
comités paritarios, where workers and employers were represented and
consulted on collective bargaining and employment regulation (Crouch
1993: 134f.). After Spain’s social cleavages erupted in a civil war, the
victorious General Franco formed a government in Spain in 1939 and
united ‘a heterogeneous alliance of conservative and fascist groups, whose
paramount concern was law and order. Their legitimacy was founded on
the support of both the army and the Catholic Church’ (Moreno 2001:
101). Strongly influenced by Catholicism and conservatism, Franco
aimed to uphold social stability, not least to stabilize its own regime.
He understood social policy as conservative Catholic policy, entitling

workers ‘to receive social protection against unexpected risk. In turn, he
or she should be obedient and diligent’ (Moreno 2001: 101; Arriba and
Moreno 2005: 144). The state not only took social policy in its hands,
but also governed the production system with a similar authoritarian-
ism, so there were ‘no independent employer or worker organizations’
(Meer 1996: 311; Guillén 1992: 122).
Spain introduced a new retirement policy in 1939, complemented
by mutual societies that protected against social risks and whose
organization reflected cleavages between regions and occupations
(Guillén 1992: 123; Arriba and Moreno 2005: 144). It then introduced
compulsory health insurance in 1942 and unemployment insurance
in 1947. Even though this system mainly targeted the poor, it was ‘a
contributive one, based on the fees of employers and workers’ and using
social policy to achieve ‘a situation of security and stability’ (Guillén
1992: 124f.). Franco emulated Bismarck’s model of social insurance.
Similar to Germany, he tried to achieve social stability through ‘job
and income security for male workers’, a rigid labour market and sup-
port for traditional family roles (Guillén 2010: 185f.). Just as he tried to
stabilize the social order through the welfare state, he also submitted
the production system to state regulation, so that there was little room
for bargaining between independent associations (Molina and Rhodes
2007b: 237). He accompanied conservative social policy by moves
towards a conservatively coordinated production system, when in 1958
the Collective Agreements Act allowed collective bargaining at ‘local,
provincial or regional levels’ and in companies and plants (Lucio 2000:
428; Moreno 2001: 102). In 1963, Spain somewhat unified its different
social insurance programs. But the country’s system remained distinctly
conservative, in that ‘the organization of the system was kept along
occupational lines’ and in that each occupation had its own insurance
scheme (Guillén 1992: 125).
Similar to other conservatively coordinated countries, social policy
became more universalistic during a moment of national unity, soon
after Franco’s death. When the entire country could agree on a peaceful
transition to democracy, the 1978 constitution ‘introduced the princi-
ple of universal extension of coverage of social security to all Spaniards’
(Guillén 1992: 127). Parallel to this universalist moment in the welfare
state, socialist and communist trade unions and labour associations
engaged in national coordination, with the 1977 ‘Moncloa Pact’ (Meer
1996: 312). They linked wage increases to the money supply to combat
inflation and ensure a smooth transition to democracy. Basically, this
meant that trade unions restrained wage demands, wherefore the gov-
ernment committed to a progressive tax system and universal social
coverage. However, this was conservative in that it aimed towards social
stability to avoid a coup d’état after social unrest (Moreno 2001: 103).
While during this moment of national unity concertation could be
reached to consolidate democracy, the country’s social cleavages erupted
again after the transition to democracy was successful (Lucio 2000: 440;
Royo 2007: 49). Again, the numerous social groups into which society
at large, but also associations were split, hindered not only national eco-
nomic coordination, but also universal welfare legislation (Guillén 1992:
133; Meer 1996: 325; Pérez 2000: 343).
Namely, ‘health care, education and social services, were left largely
in the hands of the Comunidades Autonomas’ (Gallego, Coma and
Subirats 2005: 105; Moreno 2001: 105; Arriba and Moreno 2005: 147).
Instead of universal social policy, ‘special schemes continued to be cre-
ated for specific categories of citizens’ (Guillén 1992: 127). The resulting
difference between well-protected insiders and outsiders is all but social
democratic (Glatzer 2005: 115). Instead, it is typical for a conservative
country, aiming to maintain a stable social structure by keeping insiders
and outsiders where they are. Spanish welfare is and was thus highly
fragmented. Indeed, scholars claim that ‘[p]erhaps the most relevant
factor conditioning welfare development in Spain is the importance of
decentralization both at the level of planning and policy implementa-
tion’ (Arriba and Moreno 2005: 142). Similarly, Spain now has regional
and sectoral economic cooperation between labour and governments
(Royo 2007). However, similar to other conservatively coordinated coun-
tries, divisions along regional, political and class lines kept on hindering
nationwide coordination (Pérez 2000: 345ff.). At the end of the 1990s,
there was again some national concertation in welfare and production
regulation (McVeigh 2005: 100f.). However, this is not a divergence from
a conservative model, in that it ‘aimed principally at reinforcing the
contributory nature of the Spanish system of social security’ (Moreno
2001: 107; Pérez 2000: 343ff.). Spain can thus not only be considered
conservative because its reforms were driven by a desire to achieve social
stability (Royo 2007: 51). It is also conservative because ‘segmentation
between insiders and outsiders has been a constant feature’ (McVeigh
2005: 100).
There again appears a similarity between, on the one hand, con-
servative welfare legislation, centred on contributions and on uphold-
ing existing social structures (Gallego, Coma and Subirats 2005: 104)
and, on the other hand, a production system where coordination is
segmented along similar cleavages as the welfare state, so that ‘unions

have rarely been able to co-ordinate their bargaining strategies at a
national level’ (Lucio 2000: 443; Meer 1996: 317). In this sense, Spain is
another example of a country where reasons that hindered a universal
welfare state also hindered encompassing coordination, so that a policy
style of conservative coordination marks both the welfare state and
production system.
5.2.9 Portugal
When Portugal’s monarchy was overthrown and a republic established
in 1910, Portugal was, similar to other countries that are now in the
conservative group, ‘beset by political factionalism, economic strife
and social unrest’ (Chuliá and Asensio 2007: 607). In 1933, António de
Oliveira Salazar, a devout Catholic, established a de facto dictatorship.
He ‘perceived his mission as restoring the social order’ – a phrase famil-
iar from other countries under Catholic leadership (Hampson 1997:
154). By abolishing all autonomous labour and employer organizations,
he did restore order, for better or worse (Crouch 1993: 136; Chuliá and
Asensio 2007: 623). His declared aim was to keep Portugal ‘traditional,
rural, and Catholic’, basically: keeping it as it was (Crouch 1993: 177). In
1935, he introduced a mandatory rudimentary social insurance scheme
for sickness, invalidity and old age. In accordance with ‘the tenets of
exclusionary corporatism, each professional group was responsible for
the protection of its members against social risks, and consequently
each profession had to establish its own scheme’ (Chuliá and Asensio
2007: 624; Capucha et al. 2005: 207ff.). The schemes introduced were
in that sense strictly conservative, solidarity was limited to in-group
solidarity, as the state refused to support the schemes (Hampson 1997:
154). This remained so until the revolution of 1974, making Portugal
an extreme latecomer in terms of welfare development (cf. Chuliá and
Asensio 2007: 625). Just as the state wanted to keep the social structure
as it was, it also militated ‘against the modernization of economic and
social structures’ and rejected organized industrial relations (Capucha
et al. 2005: 211).
Similar to Spain, the most universal legislation came during a
moment of national unity. When the Carnation Revolution established
democracy in 1974, Portugal introduced a pension system, which was
the country’s ‘first non-contributory element of state income support’
(Hampson 1997: 155). The minimum statutory pension now had to
amount to at least half the minimum wage, establishing some redis-
tribution to the poor. Portugal also introduced social pensions for the
needy elderly, as well as non-contributory municipal unemployment

assistance. This was a move towards universalism and redistribution
(Hampson 1997: 155f.; Chuliá and Asensio 2007: 625f.; Capucha et al.
2005: 213). Similar to Spain, Portugal drew up a constitution in 1976
that stated that a right to social security was deriving from citizenship.
Consequently, the self-employed were soon included in the social secu-
rity schemes (Hampson 1997: 155f.). In 1979, Portugal even introduced
universal health care. This was clearly an exception against the norm of
occupationally stratified schemes (cf. Bentes et al. 2004). In 1984, the
‘Basic Law of Social Security’ tried to unite the hitherto 14 contribu-
tory schemes, each with different contribution rates, into a universalist
scheme with a common contribution rate.
Interestingly, the welfare state moved towards universalism during a
period of national unity, when everyone agreed that it was important
to move towards democracy. It was also during this time that economic
coordination increased (Barreto and Naumann 2000: 397ff.). Portugal
established the Permanent Council for Social Concertation, which marks
the beginning of Portuguese neo-corporatism (Lima and Naumann 2000:
322; Chuliá and Asensio 2007: 622). At the same time, the country also
started to develop organized vocational training schemes (Capucha et al.
2005: 205).
However, the government was unable to unify social security rates,
so that different professions still paid different rates, and solidarity
remained limited to social groups (Hampson 1997: 155f.). Likewise in
the production system: national bargaining and representation were
impossible, as ‘unions have defended traditional demarcations and have
resisted even the most obvious mergers’ (Barreto and Naumann 2000:
410). Again, broader social cleavages influenced the welfare and the pro-
duction system, aligning both to a common principle. At the end of the
1990s, employers and unions remained divided (Barreto and Naumann
2000: 395; Chuliá and Asensio 2007: 623; Glatzer 2005: 113).
Overall, neither a picture of universal solidarity in the welfare state, nor
of encompassing economic coordination emerges. Instead, Portugal is
conservative insofar as occupational status groups are stabilized through
occupational bargaining and the welfare state is structured along similar
occupational lines (Lima and Naumann 2000: 323, 332; Barreto and
Naumann 2000: 416f.). This conservative policy style, keeping people
where they are in the social hierarchy, is epitomized by Portugal’s rigid
protection of the core labour market, while the peripheral or irregular
sector is marked by job instability and differentiated wages (Ferrera
2010: 619; EIRO 2009i). While strongly protecting insiders, Portugal
scarcely helps outsiders, with only discretionary municipal poor relief

(Kahl 2005: 113). This keeps both outsiders and insiders where they are
in the social hierarchy, and, similar to Italy’s and Spain’s systems, is in
this sense maximally conservative. At the same time, strict dismissal
regulations allow for a coordinated economic sector that can count
on a long-term core workforce. During moments of exceptional unity,
Portugal knew universal welfare legislation and – during these same
times – national coordination. However, differently from the social
democratically coordinated countries, it does not protect all workers.
Differently from the liberal countries, it protects core workers. This
keeps everyone in their place in society, for better and for worse. This
conservative policy style is especially pronounced in Portugal’s welfare
and production system, but it is similar in Italy and Spain.
5.2.10 Japan
Why should Japan, a country that is not Catholic, not European and
not socially fragmented, be similar to other conservatively coordinated
countries? The answer is that Japan took Germany and other conserva-
tive countries as an example when modernizing its welfare state and
production system. In 1867, Japan experienced the Meiji Restoration.
Running the risk of being colonized, it wanted to modernize its welfare
and production system to catch up industrially and militarily (Steinmo
2010: 89ff.). Itō Hirobumi, future prime minister of Japan, visited
Germany to receive week-long instructions from German chancellor
Otto von Bismarck and Lorenz von Stein. As the two countries pursued
similar aims of catching up with early industrializers, Japan used ‘inte-
gral parts of an early Modell Deutschland […] to reconcile two main
objectives of the Meiji elite: a policy of accelerated modernization and
the preservation of stability and social harmony’ (Lehmbruch 2001:
60). Similar to Germany, it granted ‘privileges to groups whose coop-
eration in economic modernization and nation building was deemed
indispensable’ (Streeck 2001a: 12). Both countries therefore instituted
status-oriented welfare states and integrated ‘labour and capital into the
welfare state at either the enterprise or sectoral level’ (Manow 2001a:
118; Esping-Andersen 1997: 183).
When Japan started a health insurance system in the 1920s, unem-
ployment insurance in the 1930s, and when it extended its social
security schemes for sickness, unemployment, old-age and accidents
to the entire population in the 1940s, it not only followed Germany in
the substantive construction of these schemes; it also adopted reforms
for the same reason as Germany: to foster social stability (Aspalter and
Lai 2003: 245; Uzuhashi 2001: 106; Peng and Wong 2010: 660; Streeck
2001a: 11). Redistribution took place within social groups, mainly
within occupations or companies, to preserve the social order (Uzuhashi
2001: 109f.). Also similar to Germany, welfare schemes were not residual,
but targeted to skilled workers, helping to ‘establish a production model
based on shop-floor cooperation, stable career paths, and high skill
levels’ (Manow 2001a: 98). Japan’s insurance system relied on long job
tenure and stable employment. This in turn promoted peaceful relations
between labour and employers, so that Japan’s welfare system ‘supported
the evolution of a range of powerful mechanisms of economic coordi-
nation that are today regarded as defining elements of the German and
Japanese models of embedded capitalism’ (Streeck 2001a: 11). Aiming to
achieve social stability in this sense not only influenced the construction
of the welfare state, but also of the production system.
The conservative coordination of the production system, within
social groups, but hardly beyond them, took place in ‘keiretsus’. These
vertically integrated conglomerates upheld stable ties to each other and
to their workforce. Workers became a ‘member’ of a company for life-
time and were socially insured through it (Aspalter and Lai 2003: 252;
Steinmo 2010: 133). It is also within conglomerates that vocational
training takes place (Busemeyer and Trampusch 2011: 21; Lauder, Brown
and Ashton 2008). These vertically integrated companies are functional
equivalents to German social and economic legislation, because they
‘reward long working careers, […] buffer the employment relationship
from cyclical economic swings, and […] provide economic agents with
a stable institutional framework’ (Manow 2001a: 119). However, while
unquestionably helping insiders, this system exposes outsiders to an
inflexible labour market, keeping both groups where they are in society,
again making this country deeply conservative (cf. Dore 1997: 20ff.).
The same conservative approach, focused on stabilizing the social order,
in this sense characterizes the welfare and production system (cf. Tricker
2009: 187).
The main difference from other conservative-coordinated countries
is that in economic matters, Japan established stability on the level of
the company, not through industry-wide associations (cf. Ebbinghaus
2006: 67f.). In terms of its welfare state, Japan ‘never developed a
modern “state centered” welfare system, but instead subsidized families
and encouraged companies to provide many of the services that in
other countries are provided by the state’ (Steinmo 2010: 91; Aspalter
and Lai 2003: 248). Organizing solidarity within social groups, while
not extending it beyond them, is typical conservative policy however.
In Japan, these groups happened to be companies organized as social

groups, because policy makers were influenced by a company paternal-
ism originating in Germany, which preceded German coordination
through associations (Lehmbruch 2001: 66). Japan was not exempt
from the liberalization that affected countries since the 1980s. But its
companies still provide welfare, vocational training and security to an
outstanding degree (Busemeyer 2009: 391).
So even though Japan uses different institutional set-ups compared
to other conservatively coordinated countries, it is similar in that it
uses society’s social groups, in this case community-like companies,
to institute social solidarity and economic coordination (cf. Manow
2001a: 119). Therefore, ‘the emerging vertical enterprise communities
of modern Japan […] played the role that in Germany was taken by
corporatist associations’ (Streeck 2001a: 26; also cf. Dore 1997). Japan
is thus similar to other conservatively coordinated countries in that it
constrains the market through group-based solidarity and coordina-
tion, to achieve social and industrial peace (Lehmbruch 2001: 92). This
coordination and group-based solidarity distinguishes it from liberal
countries. Japan is also different from social democratically coordinated
countries however, in that its welfare and production arrangements
were never nationally encompassing, but focused on social groups,
whose stability it tried to uphold. As Steinmo (2010: 98) sums up: ‘The
individual is not seen as the key stone of this system, instead it is the
group’. In this sense, there is clearly a conservative policy style, centred
on upholding society’s social groups in companies, which marks both
how welfare is delivered and how production is organized.
5.2.11 Summing up conservatively coordinated capitalism

Summing up, what unites the conservatively coordinated countries and
what differentiates them? All conservatively coordinated countries have
been heavily afflicted by wars, the Catholic Church, and a heterogene-
ous population that made it exceptionally difficult to establish a stable
democratic state (with the exception of Switzerland). The overarching
aim of these states, with Germany as the most extreme example, was
therefore to create and maintain a stable political and social order
(Goodin et al. 1999: 55; Sorge 2005: 139).
Not only was the welfare state built along the cleavages of existing
social groups, and often directly managed by them. The borders of the
very same social groups influenced the scope of employer and labour
coordination in the economy (cf. Martin and Swank 2008: 183; Palier
2010a: 603; Crouch 1993: 107). Thus, these countries not only have
socially segmented welfare states, but also segmented economic coor-

dination, mostly aligned to the same social cleavages as the welfare
state. This brings economic coordination along occupational lines in
Germany and Austria, along social communities in Italy, Spain and
Portugal, within encompassing ideological and regional pillars in
Belgium and the Netherlands, and within large companies in Japan
and partially France. Because these societies use, and thereby preserve,
established social groups, they keep society stable through their welfare
and production system and can therefore be designated conservatively
coordinated, combining the labels of welfare state research and varieties
of capitalism, while highlighting the policy style according to which
these countries manage their welfare states and production systems.
These countries constitute the most heterogeneous group. Yet they
are different from social democratic ones as they reject nationwide
redistribution or coordination. They are different from liberal countries
as they largely rejected market arrangements. It is logical that they have
rejected nationwide redistribution and coordination, as well as market
mechanisms, because all of this could upset the social order, which
is what these countries try to preserve. In this sense, social stability,
as opposed to freedom or solidarity, is their overarching principle
behind economic and welfare regulation (also cf. Bernard and Saint-
Arnaud 2003). In the course of their history, conservatively coordinated
countries changed. But their reforms often ‘reinforce the segmentalist
tendencies’ that they already exhibit (Palier 2011: 50). Namely, these
countries demonstrate the strongest trends toward dualization. They
also gain competitive advantage through their high regard for stability
however, because their welfare and production system supports long-
term oriented production. Yet, the very same stability-oriented arrange-
ments impede a dynamic service sector, radical innovation and social
mobility for outsiders (cf. OECD 2004, 2005b). France, Switzerland,
Japan and the Netherlands only partially fit this schema, which should
hardly come as a surprise; it is exactly these countries that are outsiders
in many typologies as well (cf. section 2.3).
5.3 Social democratically coordinated capitalism
In a classical study, Kuhnle (1981: 125) argues that Sweden, Finland,

Norway and Denmark ‘share a basic cultural unity and are often, and
not completely without foundation, looked upon as one single entity.
The Nordic countries have traditionally had an eye on each other,
learned from each other’s experiences, and used one or several of the
other countries as justifying references for the implementation of spe-

cific policies or for the introduction and refusal of new ideas. After World
War II, closer formal cooperation was instituted at the top political and
administrative levels. Ties were established that not only furthered the
coordination of public policies on a number of topics, but which con-
tributed, and still contribute, to the development of similar policies
within each of the national territories’.
This section traces whether the four countries indeed share a simi-
lar history and culture, which aligns their welfare and production to
a similar principle. The literature tends to argue that this is the case.
Goodin et al. (1999: 45f.) describe the welfare states of these countries
as ‘characterized by (1) class politics, (2) socialist economics and (3)
redistributive social policies. […] The fundamental value underlying
the social democratic welfare regime is social equality’. But why would
these four small countries have a fundamental value of social equal-
ity, which sets them apart from other countries? One reason might
be their lack of social cleavages. Whereas social divisions marked
continental European countries, Scandinavian ones have ‘homogene-
ously Protestant societies with Lutheran state churches’ (Kersbergen
and Manow 2010: 267; cf. a summary of similarities and differences
Thorkildsen 1997). What is more, it seems probable that Lutheranism
influenced welfare and production arrangements. Lutheranism ‘rejected
the idea that generous donations could prevent sinners from eternal
damnation and agony in fire and brimstone, or that the poor would
be justiﬁed by living in poverty’ (Kahl 2005: 103). Thus, the culture of
these countries rejected the Calvinist idea that everyone is responsible
for himself. It also rejected Catholic almsgiving, which perpetuates
poverty by supporting it materially. Instead, the Lutheran doctrine
of these countries proposed that the state has ‘a responsibility to pro-
vide generous benefits and work opportunities’ (Kahl 2005: 116; also
cf. 2009: 289). In this philosophy, society provides for the individual
(different from Calvinism), but the individual has a duty to work and
to ‘get back on his feet’ (different from Catholicism). In addition,
Lutheranism despised profit seeking beyond individual needs. Instead,
its ‘”priesthood of all believers” promoted a culture of equality, where
obvious wealth and large social differences were not acceptable because
fundamentally all individuals are equal and have the same worth’
(Thorkildsen 1997: 159).
Therefore, and opposed to liberal countries, Sweden, Denmark, Finland
and Norway share an understanding of citizenship where everyone is
to participate fully in social life. Different from countries influenced
by Catholicism, they do not understand community as a subgroup of

society, but as society itself (Goodin et al. 1999: 46; also cf. Lipset and
Rokkan 1967: 15). When religion and social homogeneity had a strong
effect on welfare legislation, it would be conceivable that it also influ-
enced the production system of these countries. Indeed, this seems to
be the case, as the following sections will show. Of course, each of the
four countries is unique, to which the following sections will try to do
justice. Yet, in each one, not only production, but also welfare arrange-
ments followed a similar policy style of nationwide solidarity. The fore-
runner in this group was clearly Sweden.
5.3.1 Sweden
Sweden was less torn apart by war and social conflicts than continental
European nations (Steinmo 2010: 46). The country overcame social
cleavages more easily and the strong feeling of national identity, which
this entailed, led to social democracy and an egalitarian conception
of social justice (Ahn 1996; Mann 1993: 711; Steinmo 2010: 11). This
egalitarian conception of social justice characterizes Sweden’s welfare
and production system.
As early as the late seventeenth century, the Swedish monarchy took
land rights away from the nobility and distributed them to small-scale
farmers (Valocchi 1992: 193). With their small and irregular earnings,
these peasants never developed an interest in a Bismarck-style insurance
system, whose benefits depended on stable employment and contribu-
tions. As ‘Sweden was the only unified European state in which peasants
were represented in parliament’ (Knudsen and Rothstein 1994: 207),
their interests had to be taken into account, leading to universalistic
tendencies in welfare legislation even before social democrats rose
to power in the 1930s (cf. Manow 2009: 107). A second aspect that
promoted Sweden’s national unity was Lutheranism, whose guiding
philosophy was to improve the situation of the poor, instead of sustain-
ing it through transfers, which contrasted starkly with the almsgiving
doctrine of the Catholic Church. Lutheran legitimation of state aid also
differentiated Sweden from the Calvinism of liberal countries (Anderson
2009: 213). In short, Lutheranism supported solidarity in the form of
an empowerment of the poor, which both Catholicism and Calvinism
discouraged.
The nationwide solidarity that marks Sweden was inscribed in its
welfare institutions. When Sweden introduced accident insurance
in 1901 (made compulsory in 1916), sickness insurance in 1910 and
compulsory pension insurance in 1913, these programs ‘went further
than any existing national law in the world: insurance was made
universal, so that, with minor exceptions, the entire population was
covered […], which was a far more radical idea than proposed in the
German law’ (Kuhnle 1981: 129; Valocchi 1992: 193f.; Anderson 2009:
225; Kudrle and Marmor 1981: 83; Flora and Alber 1981: 59). When
social democrats came to power in 1932, they introduced the concept of
the ‘Folkhemmet’, the ‘people’s home’. This meant that solidarity was
not confined to society’s different groups, but comprised the nation as
a whole. Not only did it mean that all individuals were to be treated
equally – instead of treating them according to the social group that
they were in. It also meant that society should cooperate on a nation-
wide level (Salonen 2001: 146). There are some exceptions to strong
national cooperation. Sweden adopted national subsidized voluntary
unemployment insurance as late as 1934. However, it had extensive
public employment schemes to make up for this, foreshadowing the
idea of the modern Swedish welfare state, where public employment
instead of transfers reduces unemployment (Alber 1981: 153, 156;
Edling 2006: 120).
The same national unity that was necessary to construct a nation-
ally universal welfare state also led to national economic coordination
(cf. Swenson 2002: 15; Steinmo 2010: 11, 47). The Swedish production
system has been ‘the Mecca of north European corporatist industrial
relations’ (Teague 2009: 502; Baccaro and Howell 2011: 27). Its bargain-
ing was more centralized and more redistributive than in continental
European countries (Crouch 1993: 141, also cf. Pontusson 1997: 56;
Steinmo 2010: 23). Just as Sweden’s welfare state differs from con-
servative ones in that it does not segregate benefits into occupations,
Sweden’s traditional bargaining system was similarly inclusive, capable
of nationwide coordination (cf. Crouch 1993: 89; Pontusson 1997: 64).
Most relevant in this respect is the Saltsjöbaden compromise between
the Swedish social democrats, the Swedish Trade Union Confederation
and the Swedish Employers Association in 1938 (cf. Edling 2006: 118;
Valocchi 1992: 146; Ervasti et al. 2008: 9ff.; Steinmo 2010: 50ff.).
Sweden reached another compromise in 1951, with the Rehn-Meidner
model. This was a system of solidaristic nationwide wage bargaining,
which followed the same logic as the welfare state in equalizing incomes
across the nation, as it tied all wages to increases in national productiv-
ity and raised low wages disproportionately (Hall 2007: 48; Ebbinghaus
2006: 66; Steinmo 2010: 54ff.; Dølvik and Martin 2000: 296ff.). Similar
to welfare arrangements, which redistributed not within occupations,
but across the nation, economic cooperation was not restricted to
industries or regions, but took place nationally (Steinmo 2005: 151; for
vocational training, cf. Busemeyer and Trampusch 2011: 13).
Until the 1970s, Sweden thus not only coordinated its economy in a
way that reflected nationwide solidarity; the welfare state reflected the
same policy style of national solidarity by providing public services to
all and redistributing secondary incomes (Valocchi 1992: 148; Pontusson
1997: 62; Steinmo 2010: 33). Ervasti et al. (2008: 9f.) claim that national
wage solidarity is not only complemented by the welfare state, but also
by similarly cooperative arrangements in the workplace. They argue that
status differences between workers and management are less important
than in liberal or coordinated conservative countries. This similar princi-
ple behind regulation of the welfare and production system can be called
social democratically coordinated, since both the production and the
welfare system attest to national solidarity. Such a system is obviously
different from liberalism, which eschews interference in the market –
either through the welfare state or through the production system. It is
also different from the coordination that exists in countries with a con-
servative welfare state, where redistribution and coordination take place
within social groups below the national level.
In 1983 however, this system changed. The Swedish Engineering
Employers Federation reached an individual agreement with the
Swedish Metalworkers Union, sidestepping a national agreement
between the Swedish Employer Confederation and the Trade Union
Confederation (Dølvik and Martin 2000: 293f.). National bargaining
eroded, as across the board wage increases became harder to implement
for ever more heterogeneous workers (Valocchi 1992: 148). The main
difference was between public- and private-sector employees (Steinmo
2010: 61ff.). During the 1990s, first with the Industry Agreement, then
with the Mediation Authority, industry-wide bargaining gained accept-
ance. While decentralizing its industrial relations, Sweden also deregu-
lated its financial and corporate governance market and cut welfare
programs (Valocchi 1992: 149). However, the Swedish industrial sector
still sets the pace for bargaining in the rest of the economy and the
government mediates the different industry agreements, so that these
‘avoid the free-rider danger and proceed on the basis of a common view
of the wage growth consistent with macroeconomic stability’ (Dølvik
and Martin 2000: 313; EIRO 2009j: 5f; also cf. Baccaro and Howell 2011:
26ff.; EIRO 2009k: 3; Martin and Thelen 2007: 7).
Most interesting, the direction of change was similar for the pro-
duction system and the welfare state. While economic coordination
decreased, the welfare state was retrenched. However, Sweden liberalized
neither of the two nearly as much as Anglo-American countries, which

already had the most liberal welfare state and production system to
begin with. This is partly because of the still-widespread willingness
of Swedes to support high taxes and think collectively, which influ-
ences the welfare state and the production system (cf. Steinmo 2005:
162; Teague 2009: 513). For the same reason, liberalization in both
the welfare state and the production system meant something differ-
ent in Sweden, than in liberal or conservatively coordinated countries.
Liberalization in Sweden was accompanied by measures to help people
survive in a more market-oriented environment, through active labour
market policy and continuing education. Thus, liberalization in Sweden
did not lead to dualization, as it did in conservatively coordinated coun-
tries, nor was it a frontal assault on non-market institutions, as in liberal
countries (Thelen 2012). Again, welfare and production arrangements
seem to have developed in tandem and according to a similar policy
style, which stresses national solidarity. The same encompassing solidar-
ity that sets Sweden’s welfare arrangements apart also distinguishes its
production system. Indeed, if Lutheranism and social democracy influ-
enced welfare arrangements, why would they not have an analogous
influence on the production system? A similar explanation accounts for
Denmark, Finland and Norway.
5.3.2 Denmark
After Denmark lost the regions of Schleswig and Holstein to Prussia and
Austria in 1864, it had a fairly homogeneous population (cf. Mann 1993:
711). Its religious denomination further supported its social homogene-
ity. The dominant Lutheran church understood the state as the uncon-
tested provider of welfare. As a result of this, by ‘the beginning of the
19th century, the Danish government saw itself as ultimately responsi-
ble for the poor, and [in 1849] the duty of the state to provide relief was
even written into the constitution’ (Kahl 2005: 106; 2009: 273).
Similar to Sweden, Denmark took land rights away from the nobility
and gave them to small farmers, who received a substantive voice in
parliament. As in other Scandinavian countries, these independent peas-
ants shifted public policy to the left (Manow 2009: 107). Exceptionally
in Europe, Denmark’s early capacity of national concertation meant
that there was no revolutionary unrest in the nineteenth century, but a
peaceful settlement on a democratic constitution, which took the peas-
ants’ interests into account (Knudsen and Rothstein 1994: 214). Their
modest landholding size inclined them to sympathize with low-income
groups and blue-collar workers (Knudsen and Rothstein 1994: 215; also cf.
Nørgaard 2000). But, similar to Sweden, their interests were not served by
welfare measures that relied on prior contributions from industrial work,
as conservative occupationally segmented schemes do, so the peasant’s
political influence leaned towards universalism (Sørensen 1998: 367ff.).
Due to Denmark’s homogeneous society and its policy of social inclu-
sion and Lutheranism, scholars speak of ‘national values, or ideological
foundations, or a fundamental mind-set which lay behind Danish social
legislation [in that] society’s different groups are bound together by
mutual responsibility’ (Levine 1978: 54, 68; Sørensen 1998: 367). And
indeed, Denmark legislated early and inclusive social policy. In 1862,
even before Germany, it appointed commissions that recommended
sickness insurance. In 1891, it passed a non-contributory pensions law
for the destitute, which, exceptional for the rest of the world, while typi-
cal for what was later to be called a social democratic welfare regime,
was financed by taxes. Denmark set up a sickness fund in 1892, only
preceded by Sweden and Germany (Flora and Alber 1981: 59). Equally
early, it passed an accident insurance scheme in 1898, which became
compulsory in 1916 (Kuhnle 1981: 129). It was also one of the first coun-
tries to provide national subsidized voluntary unemployment insurance,
in 1907 (Alber 1981: 153; Edling 2006: 100). The administration of the
unemployment insurance encouraged trade unions to centralize. This
in turn gave rise to an institutional structure that could later coordinate
the production system (Trampusch 2010a: 205; Edling 2006: 107). Since
all could pay into the system, yet skilled workers had a lesser chance
of becoming unemployed and paid higher contributions, they sup-
ported unskilled workers through this scheme. In this sense, nationwide
unemployment insurance redistributed income beyond the confines of
social groups. This is contrary to conservative unemployment schemes,
which are occupationally segmented. Social provision moved away from
means-testing to providing security for all. In 1922, the country revised
its old age pensions. It started to cover disability in 1933. In the same
year, it also made health-insurance semi-compulsory (cf. Flora and Alber
1981: 59). After Denmark passed this law, it had ‘the world’s most exten-
sive coverage of social risks’ (Kuhnle and Sander 2010: 69).
After this, protection increased, as social benefits became statutory
rights. At the same time, recipients of social assistance were ‘exposed to
strict control from the authorities and they still lost franchise, [but] fewer
were regarded as not deserving poor’ (Abrahamson 2003: 115). Denmark’s
modern unemployment assistance still replaces up to 90 per cent of prior
earnings and about 70 per cent of an average production worker’s wage.
But the unemployed are still strongly pressured to seek and accept work
(cf. Madsen 2002: 249ff.). It is striking how much this conforms to

Lutheran thought, trying to help not through almsgiving, but by per-
suading and enabling the destitute to leave their situation.
If Lutheranism and social homogeneity promoted nationwide solidar-
ity in the welfare state, did they also promote nationwide coordination?
As early as 1879, Danish artisans formed a nationally unified association
(Trampusch 2010a: 203). Already in 1899, trade unions and employer
associations had developed confederations that conducted national
wage bargaining (Crouch 1993: 80; Martin and Swank 2012: 63). Thus,
‘Danish industrial relations were organized relatively early and with
a high degree of coordination and centralization’ (Trampusch 2010a:
204; Lind 2000: 136; Edling 2006: 119). Following this, skill formation
became more collectivist and corporate governance more stakeholder-
oriented (Trampusch 2010a: 214; Mallin 2010: 220ff.). Again, it seems as
if social policy and production regulation moved in tandem.
Differently from Sweden however, Danish farmers needed no protec-
tion from foreign competition (Ebbinghaus 2006: 66). Also, the Danish
Lutheran Church was more liberal than the Swedish one (Thorkildsen
1997: 156). Accordingly, there was a ‘stronger position of liberalism in
Danish politics, which has continued all the way from the mid nine-
teenth century to the present’. This influence of liberalism again not only
influenced the welfare state, but also the production system (Knudsen
and Rothstein 1994: 216; also cf. Mann 1993: 711; Kuhnle 1981: 135).
While some national bargaining eroded in the 1970s, Denmark kept
more national coordination than most countries (cf. EIRO 2004; 2009c).
Martin and Thelen (2007: 34) claim that: ‘Even though Denmark had
decentralized collective bargaining to (sic!) the industrial level, it cre-
ated national framework agreements to coordinate bargaining and pre-
served strong political macro coordination in policy-making channels’.
Lind (2000: 135; also cf. Kesting and Nielsen 2008: 46) claims that since
‘the relationship between the state, capital and labour in Denmark is
co-operative, it can be argued that a social pact is present in the shape
of a continuous, tacit and implicit pattern of co-operation embedded in
well-defined institutions in society’. Welfare and production arrange-
ments in Denmark are structurally similar in the sense that ‘state,
capital and labour are engaged in the same macroeconomic and social
discourse on working life and welfare state policies’ (Lind 2000: 136).
Similarly, Trampusch (2010a: 206) maintains that ‘cross-class coalitions
which evolved in skill provision and unemployment insurance were
further strengthened by consensual labour relations’. Martin and Swank
(2008: 183; also cf. Levine 1978: 56) mention that the high degree of
solidarity behind Denmark’s universalist welfare state also promoted a

unified associational structure and thereby economic coordination.
They also claim that peak associations of capital and labour still nego-
tiate national policy in a consensual way (Martin and Swank 2012: 7).
There seems to be wide agreement that Denmark has an enduring policy
style of national cooperation, which not only influences its welfare, but
also its production system, aligning both to a similar principle of social
democratically coordinated capitalism. Denmark stands out in regard to
other social democratically coordinated countries however, because in
both subsystems, it has elements of liberalism, which other social dem-
ocratically coordinated countries lack. As the following sections show, a
similar situation of an overarching policy style, connecting production
and welfare arrangements, exists in Finland and Norway.
5.3.3 Finland
Finland is not as clear-cut as Denmark and Sweden. It was not as homo-
geneous as either of the two countries and has travelled a bumpier
road to social democratic coordination. Until 1809, Finland was part
of Sweden; 600 years of shared history ‘had a deep impact on Finnish
society and left durable traces on the governance of modern Finland’
(Pesonen and Riihinen 2002: 20). Not only was Finland influenced by
Swedish Lutheranism; its peasants were also freed much like the Swedish
(and Danish) ones (Pesonen and Riihinen 2002: 24ff.). From 1809 until
1917, Finland became part of the Russian Empire. Before the beginning
of the twentieth century, it was therefore not autonomous (Crouch
1993: 310). It nonetheless got an accident insurance law in 1895, which
was only semi-compulsory however, and, untypical for Nordic countries,
not supported by state subsidies (Kuhnle 1981: 130f.). Being trapped
between Russia and Sweden, the country experienced a strong cleavage
between Finnish natives and a Swedish minority. A second cleavage
formed when a nationalistic movement (Fennomania) united against
Russian rule. This movement wanted to unify Finland along the lines
of the Nordic countries. It ‘resisted the prevailing class society’ and
‘agreed on the necessity of preserving the social system of Nordic ori-
gin’ (Pesonen and Riihinen 2002: 27). The Sweden-friendly movement
stressed national solidarity and equality; it represented a strong labour
movement and encompassed social democrats and socialists (Kautto
2010: 589). In this environment, Finnish employers ‘were developing, in
direct imitation of the Swedes, centralized power, but had few dealings
with a government that was mainly concerned with issues of national
independence and agrarian policy’ (Crouch 1993: 114). Around the same
time, Finland followed Denmark and Norway as a forerunner in pro-

viding subsidized, voluntary and union-run unemployment insurance
(Alber 1981: 153; Edling 2006: 100; Böckerman and Uusitalo 2006: 286).
The state subsidized about 40 per cent of the scheme, but, untypical for
a socially-democratically coordinated country, every union could run its
own fund. At the same time, Finland made its industrial accident insur-
ance compulsory (cf. Flora and Alber 1981: 59).
Things changed when the Russian Empire broke down in 1917 and
Finland declared independence. It was but a matter of weeks before the
young nation experienced a civil war. The government tried to expel
Russian forces and fend off a coup d’état by socialists (Allardt and
Pesonen 1967: 326). Finland was torn between market-oriented, social
democratic nationalists on the one hand and international socialists
that upheld ties to Russia, on the other. Following these social con-
flicts, labour relations were conflictual. In 1930, Finland outlawed the
Confederation of Trade Unions. Employers also strongly opposed col-
lective bargaining (Edling 2006: 118). The country only overcame its
divisions in the Winter War of 1939, when it was attacked by Russia,
ending the nation’s central cleavage between communists and adher-
ents of Scandinavian social democracy in favour of the Scandinavian
model. Therefore, after the Second World War, Finland took Sweden
as a model (Manning and Shaw 1998: 573, 588; Pesonen and Riihinen
2002: 43). It is seen as ‘a receiver country’ of Swedish policy in welfare
and production arrangements (Knudsen and Rothstein 1994: 204;
Kautto 2010: 590; Crouch 1993: 331; Flora and Alber 1981: 59). The
war against Russia not only brought a unity and orientation towards
Sweden that permitted nationwide welfare transfers (Bergholm 2009:
41). It also allowed economic coordination in the form of national
social pacts:
Social pacts have a long history in the Finnish labour market, begin-
ning from the Winter War in 1939–1940, when the employees’ and
employers’ central organisations agreed to solve their disputes through
negotiation, […] employers acknowledged employees’ right to unionise
and employees acknowledged employers’ right to direct work (Haataja,
1995). After the Second World War, the Finnish social partners started
to negotiate social pacts on the level of the main organizations.
(Kauppinen 2000: 162)
Starting in 1968, the social partners pursued centralized wage bar-

gaining, which reflected nationwide solidarity. But they also included
nationwide ‘changes in the social security system’ (Böckerman and

Uusitalo 2006: 286; Kauppinen 2000: 163). Thus, when coordination
started in the production system, it also strengthened the welfare state.
Still in 2007, Finland had nationally centralized wage bargaining, which
it further differentiated at the industry level. It seems that the industry
level is now becoming dominant (Jackson 2005: 12; Venn 2009: 16;
EIRO 2009k). So far however, the same widespread cooperation that we
find in the modern Finnish welfare state can also be found in its pro-
duction system (cf. Fulton 2011; EIRO 2009d: 2). Again, it seems as if
an overarching policy style influenced both subsystems, aligning them
to a common principle.
5.3.4 Norway
Norway was part of the Danish Kingdom between 1375 and 1814;
afterwards it was under the Swedish crown until 1905, although it
had its own constitution and independence in domestic matters
(Knudsen and Rothstein 1994: 204). Similar to other Scandinavian
countries, Norway’s population shared a strong feeling of national
unity and a ‘historical-cultural story of the state as furthering integra-
tion’ (Christensen 2005: 726; Mann 1993: 711). Also similar to the
other Scandinavian countries, Norway’s peasants were independent
and politically represented. They blocked non-universalistic policy (cf.
Manow 2009: 107):
The farmers maintained that other groups lived under equally as bad
conditions as industrial workers and that there should be equality
before the law in the sense that everyone with the same social needs
and with an income under a certain level should be insured, and
thereby be part of the “state care” originally intended for factory
workers.
(Bjørnson 2001: 202f.)
Given these structural characteristics, Norwegian social policy was

early and encompassing. In 1894, Norway passed compulsory accident
insurance, which differed from the logic of Bismarckian policy in that it
organized solidarity on a nationwide basis (Kuhnle 1981: 130). In 1906,
Norway became the second European country (after France) to legislate
national, state-subsidized, if voluntary unemployment insurance (Alber
1981: 153). In 1909, equally early by international standards, it passed
compulsory sickness insurance. It created an old-age pension scheme in
1936, modelled on the Danish law of 1891 (Kuhnle 1981: 130). In 1938,
it introduced compulsory unemployment insurance, which was earlier

and more universal than in any other Scandinavian country (Edling
2006: 99; generally, cf. Flora and Alber 1981).
Norway thus laid the foundation for its universal welfare state early.
It consolidated it after the Second World War. Its social policy, espe-
cially in the area of poor relief, was inspired ‘by a more inclusive idea
of solidarity that embraced farmers and fishermen as well as industrial
workers [which] came partly from other Scandinavian countries’
(Bjørnson 2001: 207). The literature names nationwide solidarity as
the reason behind Norway’s early and extensive welfare schemes. Stein
Kuhnle (1980: 102) mentions that a dominant motive in Norwegian
social policy is ‘the value of national individual equality’. Again, we are
not talking about equality within one’s social group, but about national
equality. Bjørnson (2001: 209) similarly claims that ‘ideas of solidar-
ity and […] national community’ influenced Norway. Christensen
(2005: 722) argues that the welfare state was and is ‘widely accepted
as appropriate by the citizens and cultural acceptance or trust in state
institutions has been relatively high. Collective goals and values were
very important. And when interest groups started to make an impact
on governmental decision-making processes, mainly after World War
II, they were also seen as participating in a collective state-led effort: to
build the Norwegian welfare state’.
Such principles of nationwide solidarity brought about a universalis-
tic and solidaristic policy style in welfare legislation. Christensen (2005:
733) claims that the Norwegian production system was influenced by
the same principles. Regulation in both areas reflects a belief that mar-
kets should be contained to limit social inequality. The same nation-
wide solidarity that marked the welfare state thus also left its imprint
on the production system:
Historically, Norway and Sweden shared the Nordic tradition

of labour market regulation through national level negotiations
between highly organised employers and unions, aimed initially at
managing conflict so as to maintain social peace and more recently
at moderating wage growth so as to maintain macroeconomic stabil-
ity. Moreover, there has also been an implicit understanding that
co-operation in the labour market would be coupled with expan-
sion of a universalistic welfare state and the maintenance of full
employment. In both countries, then, there was thus something like
sustained, institutionalised social pacts.
(Dølvik and Martin 2000: 279)
Thus, from early on, Norway not only had passed early welfare leg-
islation. It was also one of the countries with the highest capacity for
national economic coordination (Crouch 1993: 89). Since then, ‘Norway
has always been ranked as one of the most corporatist and centrally co-
ordinated economies in the world’ (Dølvik and Martin 2000: 280). Thus,
not only the welfare state but also the Norwegian production system was
influenced by a notion of nationwide solidarity. In fact, when Sweden’s
national pacts broke down in the 1980s, Norway largely managed to
uphold its own nationally negotiated social pacts. Its welfare state is also
largely intact. In this sense, it is now probably the most typical social
democratically coordinated country, both in terms of its welfare, as in its
production system, which are both marked by a social democratic policy
style that mirrors a commitment to nationwide egalitarianism.
5.3.5 Summing up social democratically coordinated capitalism

As the preceding chapters showed, all four Scandinavian countries
achieved remarkable nationwide solidarity. This is because they were
all influenced by Lutheranism, which extended the idea of solidarity to
the entire nation, instead of keeping it contained within social groups.
Lutheranism argued that the state should minimize social inequalities,
as it promoted the view that each ‘individual had to be awakened to
fulfil himself and the purpose of his own life’ (Thorkildsen 1997: 157f.).
This is far from the Catholic view that poverty should be accepted or the
Calvinist view that it is God’s punishment. These countries thus firmly
believed that the state should help the worse-off.
What accounted for a strong welfare state also accounted for strong
coordination (cf. Goodin et al. 1999: 249; Kelly and Hamann 2008:
138; Crouch 1993: 14). National unity supported nationally central-
ized employers and trade unions. In the language of quantitative social
science, independent variables behind welfare states become independent
variables behind production systems, exercising the same influence on
both, thereby aligning them to a common principle. The ensuing policy
style of social democratic coordination, which influences welfare and pro-
duction, seems to be relatively stable, persisting over decades and through
times of crises (cf. Nordlund 2000; Nygård 2006; Kautto 2010: 595f.).
However, there are also important differences between the four social
democratically coordinated countries. Whereas Sweden was the fore-
runner, Denmark and especially Norway and Finland oriented them-
selves to Sweden, once they devised their own policy at the beginning
of the twentieth century (Knudsen and Rothstein 1994: 204, Crouch
1993: 337). Later, Sweden’s centrally coordinated industrial relations
broke down in the 1980s. Denmark was and remains more liberal than
the other three countries. Norway is by now probably the most typical
representative of a social democratically coordinated country. But it can
also buy its way out of fiscal problems with its oil and gas revenues.
Common to all these countries is a general mindset however, that
social solidarity and economic success can mutually entail, rather than
prevent each other. Each of these countries shares a strong believe that
‘everyone pulls on the same string’; so each of them tries to gain competi-
tive advantage by empowering individuals (cf. Rothstein 1998: 192ff.).
Accordingly, the high cost of the welfare state of these countries is paid by
income and consumption taxes, rather than charges on capital (cf. OECD
2005c). Production is built on incremental and some radical innovation
(cf. Crouch 2005a: 444). The huge public sector employs – and pro-
vides social services for – a large share of the population. High quality
production thrives because the state delivers an excellent infrastructure
for research and to conciliate work and life. For these reasons, scholars
conclude that ‘[d]espite globalization and industrial restructuring, the
Scandinavian countries have largely retained “macrocorporatism” [while]
in Germany and other continental European countries, economy-wide
negotiations are on the decline, although economic sector-based coordi-
nation is resilient’ (Martin and Swank 2012: 28).
A common social identity and an ensuing national solidarity allowed
for social transfers and empowerment of weak members through the
welfare state. It also allowed for national coordination in the produc-
tion system. This explains why all social democratic welfare states are
coupled with coordinated market economies, which were traditionally
coordinated nationally and remain so to a higher degree than in other
countries. Since cooperation is nationwide and concerned with level-
ling inequalities, both in the welfare state and in the production system,
these countries deserve the label social democratically coordinated.
The difference of these countries from the liberal ones is obvious.
But their policy style is also different from conservative countries, even
though both are coordinated. Indeed, the Scandinavian countries ‘are
often described as defining the end of the spectrum of possible devel-
opments farthest from Bismarck. Where his intentions were to preserve
the existing order, co-opt the working class and restrict redistribution,
socialist social policy was universal and egalitarian, concerned with
solidarity’ (Baldwin 1990: 43). What made the conservative countries
different is that their overarching goal was not nationwide solidarity,
but stability. The following section summarizes the three policy styles
connected to welfare and production in each of the three regime types.
It also shows the limits of a unified typology that looks at welfare states
and production systems through one lens.
5.4 Summary and limitations of the approach
How can the aforementioned country differences and similarities be

summed up? The broadest difference runs between liberal and non-
liberal countries. Non-liberal countries can be subdivided into those
where solidarity and coordination are nationwide and those where
solidarity and coordination exist in social groups, but not beyond them.
In one sentence, coordination and solidarity failed to thrive in Anglo-
American countries, while both extended to social groups in continental
Europe (and Japan) and to entire nations in Scandinavia (Martin and
Swank 2012: 28). Solidarity and economic coordination have a similar
reach because the same influences either extend or limit both.
All English-speaking countries (with the exception of Ireland) to a large
degree adhere to Calvinism or closely related forms of Protestantism.
The individualism that Calvinism propagates makes both the welfare
state and the production system liberal. Scandinavian countries adhere
to Lutheranism. Lutheranism promotes the idea of a nationwide com-
munity of believers and it promotes state-interference in economic
and social life. This allows nationwide welfare solidarity and economic
coordination. The conservatively coordinated countries share (with the
exception of Japan) an adherence to Catholicism. Catholicism advocates
that the social order should not be upset, neither through the market,
nor through the state. Instead, the Catholic principle of subsidiarity
promotes cooperation within one’s social group, but not beyond it. It
therefore promotes a conservative welfare state and a group-coordinated
production system. Of course, this is only the rough picture, leaving
much room for country-level details. For example, Ireland was only indi-
rectly influenced through Calvinism, by Britain; Japan was not directly
influenced by Catholicism, but imitated Germany. In any case, this
scheme does not imply that countries uniformly adhere to one value or
even regime type, as Goodin et al. (1999: 36) aptly mention:
Liberal welfare regimes attach relatively more importance to eco-

nomic efficiency than do other welfare regimes. Social democratic
welfare regimes value economic efficiency too: they just attach rela-
tively more importance to equality than do other welfare regimes.
Corporatists value economic efficiency too; they just attach relatively
more importance to social integration and social stability than do
other welfare regimes. Those differing emphases are what is distinc-

tive about each welfare regime.
It is indeed crucial to talk of ‘differing emphases’, instead of thinking

that countries follow exclusively one goal to the detriment of all others.
Nonetheless, the scheme proposed here argues that countries clearly
belong to certain poles more than to others. Liberal capitalism limited
cooperation in the welfare and in the production system; social demo-
cratic capitalism implies nationwide cooperation in the welfare and the
production system and conservatively coordinated capitalism exhibits
group-based cooperation in both systems. Table 5.1 sketches out the
typology that results from this.
This typology is implicit in existing scholarship. It is the same classifica-
tion we arrive at when we take Pontusson’s (2005a: 27) distinction between
Table 5.1 Three types of capitalism
Liberal Conservatively Social

coordinated democratically
coordinated
… advantage Efficiency Stability for Extensive

of market incrementally public-sector
Economic distribution innovating infrastructure
industry
… disadvantage Little Inflexibility High costs
possibility
for strategic
behaviour
… advantage Low taxes, Social peace, Resources
high stable social available for
individual structure what is seen as
Social freedom morally right
policy
… disadvantage High Low social High taxes
inequality mobility
Underlying value Liberalism Conservatism Egalitarianism
Prerequisites Acceptance Strong social Acceptance of
of social groups redistribution
inequalities
Reach of institutionalized Individual Social group Nation
solidarity or family
Dominating institution Market Social groups State
liberal and non-liberal countries, coupled with his observation that the dis-
persion of welfare benefits distinguishes ‘Nordic (social democratic) welfare
states from their continental (conservative) counterparts’. One only has to
take up all the disparate threads that already exist in the literature, to arrive
at a two-by-three typology of liberal, conservatively and social democrati-
cally coordinated capitalism. The same distinction is alluded to in recent
publications that look at varieties of coordination. These split non-liberal
countries into a sectorally and a nationally coordinated variety (Martin
and Thelen 2007; Thelen 2010, 2012; Martin and Swank 2012).
Summing up crudely, the liberal regime type links the welfare of its cit-
izens as well as its production system to the market; the social democrati-
cally coordinated regime links both to the state and the conservatively
coordinated regime to those social groups, into which society is divided.
Production and distribution is therefore governed through markets,
states or social groups. The essence of an overarching policy style is thus
that the state, specific social groups or markets are seen as legitimate and
this shows in the regulation of production and distribution.
This argument squares well with studies that show how capitalist
regimes go along with different value orientations (Svallfors 1997: 295;
cf. Blekesaune and Quadagno 2003; Arts and Gelissen 2001). It also has
an affinity to Castles’ ‘Families of Nations’ approach, which stresses cul-
tural proximity in terms of language and religion to explain similar insti-
tutions between countries (Castles 1993b; Castles and Obinger 2008). My
argument also extends a power resources approach (Korpi 1985; 2006).
Simply put, when organized labour was successful in creating a strong
welfare state, it also influenced the production system. While welfare
regimes have long been acknowledged to rely on such political power
distributions, the varieties of capitalism approach was so far vulnerable
to the criticism that its functionalism cannot explain capitalist origins
(cf. Streeck 2009). In response, proponents of a new varieties approach
use power coalitions to explain where capitalist diversity comes from
(Hall and Thelen 2009: 13; Thelen 2012). Such arguments can build on
the scheme developed here, to explain why power coalitions in some
countries were able to build the production system and the welfare state
along certain policy styles, while this was impossible in other countries.
At the same time, while the varieties of capitalism approach got us think-
ing in terms of a dichotomous approach of liberal/inegalitarian versus
coordinated/egalitarian capitalism, a combination with welfare state
research promotes a view in which coordinated capitalism comes with
different sorts of equality, contrasting within-group equality to nation-
wide equality.
I do not argue however that the concept of a policy style is to be

understood as a culture that functions as an ultimate independent
variable. Policy styles are tied to the social structure of countries, not
only influencing it, but also resulting from it, as the preceding chapter
showed. In this sense, the argument presented here does not contradict
materialistic views, which claim that ‘for historically tractable reasons,
the institutions through which employers coordinated their activities
were often empirically deeply entangled with the genesis of the institu-
tions now associated with the more egalitarian models of capitalism’
(Thelen 2012). Instead, it substantiates these views by making it clear
how this co-occurrence comes about and what stabilizes it.
That welfare and production arrangements are similar for structural
reasons does not mean that there was a conscious master plan behind
the construction of either. Instead, institutions evolved through chance
and underlying social structures (cf. Streeck and Thelen 2005a; Steinmo
2010). The explanation advanced here cannot equally well account for
all countries. But no country has to fully belong to one of the three
regime types for a unified typology to make sense, as the poles of this
typology are only ideal-typical conceptions that countries embody to
varying degrees, even though the poles themselves have been distilled
from the history of countries. What is striking however is that countries
that fit uneasily into their regime type have ambiguous production-
and welfare arrangements (Switzerland, Japan, Ireland, France and the
Netherlands). This is a good indicator that similarities in the histories
of nations group these into similar regime types, while countries that
had an unusual history compared to others in their regime – resulting
in different social structures and culture, also have unusual welfare and
production regulation, as compared to their regime type. It also explains
why
CMEs featuring national coordination [what Kitschelt et al. (1999a)

called ‘national CMEs’] typically scored highest on most measures of
equality, whereas liberal countries scored the lowest. Cases of ‘sector-
coordinated CMEs’ like Germany (coordinated but at an industry
level) came out in between, both on measures of coordination and
on various measures of equality – not as egalitarian as Scandinavia
but still more solidaristic than the Anglo-Saxon countries.
(Thelen 2012: 144)
While it might have been correct to conflate egalitarian welfare

regimes and coordinated capitalism after the Second World War, as
both coordinated varieties of capitalism delivered egalitarian outcomes,

this is less correct now, because the two types of coordinated countries
liberalized in different ways, as the following chapter will show.
Some reservations are due before we proceed. It has to be acknowl-
edged that the historical account given above is stylized. But it is stylized
in a necessary way, to answer a specific question. While it cannot do full
justice to the richness of countries’ history, it presents one important
way of summarizing their historical development. It is certainly to be
contested. But it highlights factually existing historical uniformity in
the experience of those countries, which now constitute a regime type.
The possibility that similar historical experiences and social structures
gave rise to a certain culture, which was then instilled in different insti-
tutions, should be considered as aligning welfare states to production
system. While we have innumerable studies rightly highlighting how
each country is unique, we hardly have studies that show their struc-
tural similarities in terms of production and distribution. However, if
one feels compelled to argue that Switzerland constitutes its own regime
type, or that the southern European countries do, there is no reason to
object from the perspective adopted here.
6
What Can a Unified Typology
Explain?
This book was in many ways structured like a play. We have been intro-
duced to the main protagonists: the two typologies. We have seen what
keeps them apart, and how they can nonetheless be brought together.
But while it is usually self-evident that we want a play’s protagonists to
come together, one question must be answered here, which Paul Pierson
(2000: 808f.; also cf. Korpi 2000: 28) summarizes pointedly: ‘Much
discussion has focused on whether there were more than three regimes
and whether particular countries were correctly categorized. Less atten-
tion has been paid to the question of why it makes sense to talk about
regimes or worlds of welfare at all’. Similarly, Arts and Gelissen (2010:
572) emphasize that typologizing is useless unless it helps to explain
what we are interested in. So even if we can integrate typologies of pro-
duction and welfare into a unified two-by-three typology of capitalist
diversity, the question remains why we should do so. What is the added
explanatory value of a unified typology?
There is one obvious reason why it makes sense to combine the two
typologies. If reality is such that production and distribution regimes
form one coherent and unified system in each country, then so too
should the typologies that describe this system, giving rise to a unified
typology of welfare and production regimes. Neither varieties of capital-
ism, nor welfare regime typology can provide this so far. Specifically, a
typology named ‘varieties of capitalism’ (and not ‘varieties of produc-
tion systems’) must include the welfare state in its analysis because
the term ‘capitalism’ is commonly understood not only as a mode of
production, but also of distribution. Second, being able to focus on
‘vertical’ connections of Figure 3.1 (which linked the different typolo-
gies) allows us to ask new questions: How do non-liberal capitalisms
relate to each other? Why are they always a stable subgroup? How are
162
What Can a Unified Typology Explain? 163
the social democratic and the conservative welfare states different in the
economic coordination that they entail?
But this chapter goes further than these arguments. It indicates that
the proposed unified typology helps to answer contentious questions:
First, it helps to understand why some countries are hard to classify. If
a certain culture – that in turn stems from a particular historical experi-
ence and social structure – stands behind the welfare and production
systems of each regime type, then we can understand better why some
countries (especially Switzerland and Japan) combine welfare and pro-
duction arrangements that do not seem to fit: their institutions reflect
less coherent influences than those of other countries. But the explana-
tory value of a unified typology goes beyond this. It can explain produc-
tion regulation through the welfare state, and it can explain the welfare
state through production regulation. The following section illustrates
this. It shows how coupling welfare state research with research on pro-
duction systems increases the explanatory power of the former and of
the latter, bringing order into hitherto disorderly empirical facts.
6.1 How a unified typology can explain more than

existing typologies
Proponents and critics of the varieties of capitalism approach often talk

at cross-purposes when discussing the demise or resilience of coordi-
nated market economies. While proponents of the varieties approach
argue that employer coordination remains stable, critics counter that
social solidarity in coordinated market economies declines. So while
one group talks about declining solidarity, the other talks about stable
coordination. Kathleen Thelen (2010: 202) therefore urges that we
should not ‘fixate on a one-dimensional continuum-based concep-
tion of liberal versus coordinated – we should look instead for varieties
of coordination based on different coalitional underpinnings (some
clearly more consistent with solidarism than others)’. This is what I
did. I disentangled coordination and social solidarity – even though I
also highlighted that empirically, they can go together. But what is won
through this?
Attempts to link economic coordination and economic success have
not been very fruitful so far (cf. the criticism of Kenworthy 2006 against
Hall and Gingerich 2004). But such attempts have not yet taken into
account that coordination is embedded in different welfare regimes (cf.
Rhodes 2005: 367). Not only might welfare state differences explain why
countries with similar degrees of coordination perform differently. The
opposite is also true: Incorporating economic coordination into studies

of the welfare state helps to answer challenging questions. For example,
while coordination often promotes a highly productive manufacturing
sector, it goes along with decommodification, as preceding sections
of this book showed. So the overall level of work that is available in
coordinated economies might be lower, because labour in coordinated
countries is not as readily available as a commodity, it is ‘decommodi-
fied’. Accordingly, even proponents of the varieties typology argue that
coordination in the long run ‘militates against job growth’ (Hall 2001:
60). To check for this, Figure 6.1 plots the labour participation rate
(employed and self-employed workers employment rate, adjusted for
average working time and normalized to vary between 0 and 1) against
the degree of coordination for different economies.
As one can see, all liberal countries deliver a higher employment vol-
ume than all coordinated ones, except Japan and Ireland. This might
not be directly due to coordination however, but owed to the fact that
economic coordination goes along with welfare conservatism, which
keeps women away from the workplace. Indeed, Figure 6.2 on the
following page does not seem to suggest that female employment fol-
lows economic coordination, but welfare regimes.
vary between 0 and 1 – Data for 2003/04 (OECD 2005b: 238 and 255)
Gross working volume – labor force participation (age group 15–64)
times average annual hours worked per person, normalized to
New Zealand
1.00 United States
Canada
0.80 Japan
United Kingdom
Switzerland
Spain Sweden
0.60 Portugal
Finland
Denmark
0.40
Ireland
Austria
0.20 Norway
Germany
Netherlands
R Sq Linear = 0.481
Belgium
Italy
0.00 France
0.00 0.20 0.40 0.60 0.80 1.00

Figure 6.1 Coordination and employment volume

48
47
46
45
44
43
42
41
40
39
38
Sw nd
No n
De ay
rk
Ze d
Po nd
Un Can l
a
ng s
m
Au ce
Ge alia
y
er tria
Be ds
um
n
ain
ly
ga
an
e
lan
ad
lan
pa
Ita
Sw nma
do
rw
an
n
nla
ed
ala
rtu
Sp
s
lgi
str
rm
ala
Ja
Ne zer
ite St
Ne Au
Ire
Fr
Fi
Ki
Un ited
it
th
w
Figure 6.2 Female employment as percentage of total employment

Source: World Development Indicators Database 2005.
As can be seen, women in the four Scandinavian countries have the

highest employment rates compared to men. Then the liberal countries
follow, together with Switzerland and Portugal. Conservative countries
have the lowest female employment, together with Ireland. This is con-
sistent with welfare regime theory: The social democratic welfare regime
forces women to work, by not insuring them through a male breadwin-
ner. But it also enables them to work, through a large public sector.
The liberal countries also force women into work, by denying them
generous transfers. But they also offer work in their dynamic private
sector. The conservative countries, true to their calling, conservatively
keep women at home by insuring them via the main wage earner and
by making entry into the labour force difficult through their insider-
outsider segregated labour market.
But the picture is more complicated when we look at what positions
working women actually occupy. Figure 6.3 on the following page shows
the percentage of female elites (lawyers, senior officials and managers as
share of the entire labour force). The data refers to the latest available
year during the period 1991–2000. Those countries that have imple-
mented the recent International Standard Classification of Occupations
(ISCO-88) are not strictly comparable with countries using the previous
classification (ISCO-68), but the differences are not large.
50
45
40
35
30
25
20
15
10
5
0
Ze es
Ca d
da
Ki nd
Po n
Sw al
en
er tria
Fi s
er d
y
No ia
y
Be d
m
ly
n
d
an
r
an
ai
pa
Ita
g
l
do
Sw ma
iu
ra
at
an
rw
na
ed
la
Sp
rtu
lg
al
ite Irel
nl
Ja
Ne Au
Ne St
st
er
ng
al
n
Au
De
itz
d
G
te
th
w
d
i
Un
Un
Figure 6.3 Female elite decision-makers as percentage of total

Source: ILO Laborsta Database, data for 1991–2000, accessed in 2012. Classification based on
International Standard Classification of Occupations (ISCO-88 and ISCO-68).
Now the picture is different than when looking at overall employ-

ment. All liberal countries, except Australia, now stand against all other
countries. The image is not one of welfare regimes, but of varieties
of capitalism. Proponents of the varieties approach have argued that
coordinated production systems disadvantage women in reaching elite
positions. This is because in coordinated countries, vocational train-
ing usually takes place in companies, not in schools and universities.
Companies have a higher incentive to train men, as these have a higher
likelihood of remaining with a company than women, who are more
likely to interrupt their career for childrearing. While schools and uni-
versities can be indifferent to whether those whom they train will have
an uninterrupted working career, companies, which teach company-
specific skills cannot; they must be sure that trained employees will stay
within the company (Estevez-Abe, Iversen and Soskice 2001; Estevez-Abe
2005, 2006, 2009; Iversen and Rosenbluth 2010). Thus, in coordinated
countries, where qualifications are predominantly acquired through
company training, women are especially disadvantaged. Companies
refuse to teach them the specific qualifications that allow them to
reach the top, while in liberal countries, such qualifications are not
acquired in companies in the first place, but in schools and universities.
Accordingly, we should see that it is especially hard for women to reach
top-management positions in companies in coordinated countries.
Figure 6.4 below therefore shows the percentage of female manager in

companies.1
Few countries classify their labour force according to the ISCO-88
code, which provides the basis of this graph. But the only two liberal
countries that do so display the second and third highest share of
female managers in companies. So it indeed seems as if women in non-
liberal countries have a harder time reaching company-management
positions, as the rigid and firm-centred skill systems of coordinated
economies make it hard for them to get to top positions (Estevez-Abe,
Iversen and Soskice 2001; Estevez-Abe 2005, 2006, 2009; Soskice 2005;
Iversen and Rosenbluth 2010).
The field of employment was one example to show how economic
and welfare regimes interact to produce outcomes. In this case, pro-
duction and welfare analysis illustrate two sides of the same coin:
gender discrimination in accessing the labour force and in accessing
top positions of the labour force. While welfare state analysis can
explain the first, production systems can explain the latter. But it is
only by combining the two approaches that we get a complete pic-
ture. In this sense, combining the two approaches lets us understand
empirical fields that are hard to understand when using one approach
in isolation. As the following section will show, a combination of
40
38
36
34
32
30
28
26
24
22
20
ce
nd
ria
en
ds
k
ga
an
ar
ai
an
do
iu
an
an
la
st
ed
rtu
Sp
nm
lg
nl
Ire
Au
ng
Fr
al
Sw
Be
er
Fi
Po
De
er
Ki
th
d
Ne
ite
Un
Figure 6.4 Women in leading management positions as percentage of total

Source: (BA) Labour force survey. Coverage: Total employment. Classification: ISCO-88.
http://laborsta.ilo.org/ Data for 2000+.
the two typologies also makes predictions that can guide further
research.
6.2 How a unified typology can explain trajectories of

liberalization
As mentioned above, a recurring discussion between critics and propo-

nents of varieties of capitalism goes something like this: Critics of the
varieties typology argue that coordinated market economies have liber-
alized, to the point that they are now similar to liberal market econo-
mies. Defenders of the typology retort by downplaying the extent of
liberalization (cf. Martin and Swank 2012: 10). One wonders how intel-
ligent people can reach such different conclusions when faced with the
same historical facts. The reason is that critics of the varieties approach
look at a different dimension of liberalization than proponents, so
the two groups are talking past each other when discussing whether
there was ‘more’ or ‘less’ liberalization. The question is not how much
liberalization there was. The question is: What kind of liberalization?
Distinguishing between liberalism, conservative coordination and
social democratic coordination helps to answer this.2
Proponents and opponents of varieties of capitalism talk past each
other because critics point out that solidarity, which historically came
with coordination, has declined in most countries. From this viewpoint,
it makes indeed little sense to differentiate liberal from coordinated
market economies (Streeck 2009). Proponents of the varieties approach
argue that such a distinction still makes sense however, because
employers in coordinated market economies still coordinate (Hall and
Thelen 2009). Thus, critics and proponents of the approach look at two
different aspects of liberalization. Critics look at social solidarity; pro-
ponents look at economic coordination, which does not imply solidar-
ity, but means that companies conduct strategic long-term cooperation
in game-theoretic dilemmas (cf. Höpner 2007). Because coordinated
market economies were historically more egalitarian, the dimensions of
egalitarianism and coordination are often amalgamated. It is important
to keep solidarity and coordination analytically apart however, because
we can distinguish three varieties of liberalization, in which solidarity
and coordination play very different roles. First, in liberal countries,
liberalization meant full-blown deregulation; market institutions dis-
placed non-market institutions. Examples are the dismantling of union-
led bargaining in the United States and the United Kingdom, or the
destruction of arbitration courts in New Zealand. Such a ‘direct frontal
assault on institutions supporting the collective regulation of labor rela-

tions is most closely associated with what Hall and Soskice call liberal
market economies (LMEs), an association that is not wholly surprising,
since history suggests that where employers do not themselves possess
stable coordinating capacities, they will press vigorously to weaken
unions as well’ (Thelen 2012: 145).
Liberalization in conservatively coordinated countries was different.
In this second variety of liberalization, collective regulations were not
directly attacked but neglected. Instead of expanding the collective
regulation of manufacturing wages into the service sector, ‘[t]raditional
arrangements for labor-market insiders are maintained even as an
unorganized and unregulated periphery is allowed to grow that is char-
acterized by inferior status and protections for labor-market outsiders’
(Thelen 2012: 145). Liberalization in conservative countries thus meant
dualization, providing stability for insiders, while instituting precari-
ous employment for those that are already disadvantaged (Martin and
Swank 2012: 129). Thereby, liberalization in these countries reinforces
the conservatism that they are already known for (cf. the essays in Palier
2010c). Such dualization is not a displacement of old institutions with
new ones. It is not, as in liberal countries, a direct attack on negotiated
labour arrangements. Instead, this type of liberalization takes the form
of an institutional drift, meaning that manufacturing institutions are
not adapted for workers outside the core economic sectors (cf. Streeck
and Thelen 2005b; Thelen 2012: 145). Letting such dualization hap-
pen and thereby segmenting society into insiders and outsiders is the
by now familiar policy style of conservatively coordinated countries.
However, it is not the same as a direct assault on coordinated arrange-
ments, as these (non-)reforms aim at preserving coordination (and even
the associated group-based egalitarianism), albeit for a shrinking core of
manufacturing workers.
The third variety of liberalization takes place in social democratically
coordinated countries. Thelen calls it ‘embedded flexibilization’, mean-
ing that countries introduce market mechanisms, but ‘combined with
social programs designed to ease the adaptation of society, especially its
weaker segments, to changes in the market (flexicurity). While social
protections remain strong, the thrust of policy is organized less than
before around protecting individuals from the market and more around
facilitating their successful (re)integration into it’ (Thelen 2012: 145).
This does not have to rely on outdated forms of coordination, as in the
conservatively coordinated countries; for example, employment does
not have to be stable for workers and companies to cooperate. Instead,
worker mobility is not only allowed but even stimulated. Workers are
not supposed to stay within one sector or social group. State-sponsored
training and costly active labour market policy encourage workers to
move between sectors and social groups. These countries therefore
clearly introduced market mechanisms for the entire population –
different from the conservatively coordinated ones – but they avoided
the social segmentation of the conservatively coordinated countries,
because they also introduced skills training for everyone, instead of
passive support to retain people where they already are socially. Instead,
these countries’ labour market policy and comprehensive skill training
preserved an egalitarian society that is upwardly (and downwardly)
mobile (Martin and Swank 2012: 128).
In this sense, conservatively coordinated countries moved towards dual-
ization, while upholding strong economic coordination – for a shrink-
ing core of industrial workers. This ended the historical co-occurrence
between coordination and egalitarianism, which marked these countries
after the Second World War. But it fits the conservatively coordinated
policy style of these countries, which had always separated society into
social groups. The social democratically coordinated countries sustained
egalitarianism, while coordination declined. Here, workers are encour-
aged to re-learn skills through continuous state-sponsored education
and re-education. That social democratically coordinated countries let
go off some coordinated arrangements cannot be accounted for by a
straightforward variety of capitalism model, which conflates coordina-
tion with egalitarianism. But it is fully fitting with the social demo-
cratically coordinated policy style of Scandinavian countries, which
has always stressed national solidarity and a nationally coordinated
approach to social problems, over the dualization into social groups.
Only in the liberal countries did liberalization mean a simultaneous
decline of egalitarianism (in the sense of attacking the welfare state)
and coordination (in the sense of attacking economic coordination).
This in turn is fully consonant with the historical policy style of these
countries.
Thus, enduring policy styles of conservatism, social solidarity and lib-
eralism not only explain contemporary liberalization trajectories better
than a continuous scale of more or less. Also, while a simple continuum
cannot explain different forms of liberalization, a combination with
welfare state research shows that coordination and egalitarianism must
be analytically decoupled from each other, even if they tended to coin-
cide historically. Social democratically coordinated countries have been
and remain more egalitarian than conservatively coordinated ones,
which rely on group-based solidarity and coordination (cf. Thelen 2012:

146). That social democratically coordinated countries, which share the
same understanding of national unity and solidarity, organize skill crea-
tion for the most disadvantaged parts of society is unsurprising in light
of their historical policy style and their capacity for national solidarity.
That the conservatively coordinated countries, which are segmented
into social groups, answer to de-industrialization by protecting a core
group of workers, while neglecting others, is also what we would expect
based on their historical policy style. However, this is a different pic-
ture than what the original varieties approach proposes, which simply
argues that coordinated countries resist liberalization.
A view that focuses on different policy styles can also yield
hypotheses on what production and distribution systems developing
countries will adopt. Countries that liberate themselves from below
and against oppressive rulers should have a population that values
individual liberty. This should find its way into liberal production
and welfare regulation. Countries that built their state structures in
the face of strong social cleavages should tend towards a conservative
welfare state and group-based economic coordination. Both the wel-
fare state and economic coordination should be split along the social
cleavages that mark these societies. Countries that share a strong
feeling of national solidarity and have weak cleavages should instead
develop a social democratic welfare state and be capable of national
coordination.
Summary and Conclusion
In a widely read review of the comparative capitalism literature, Jackson

and Deeg (2006: 31) claim that a ‘major issue dividing opinion in the
literature is whether the role of the state and the impact of the welfare
state should be included’. This book spilt a lot of ink to give a simple
answer: yes, it should be. It suggested that two different approaches,
welfare state research and varieties of capitalism, can be integrated into
one unified typology. A principal factor and hierarchical cluster analysis
validated that countries fit into regimes if the organization of their wel-
fare state and economy is analysed simultaneously. The ensuing clus-
ters consisted of liberal market economies with a liberal welfare state,
coordinated market economies with a conservative welfare state and
coordinated market economies with a social democratic welfare state.
Switzerland and Japan were examples of rather unclear cases.
Existing typologies confirmed this view (Crouch and Streeck 1997a;
Boyer 2004b; Amable 2003; Ebbinghaus 2006). They depict the same
image of capitalist diversity as the typologies that this book combined –
albeit in different resolutions. This posed the question why all
typologies distinguish broadly similar regime types. A partial answer
is complementarities. Liberal production and welfare regimes support
each other, while coordinated production regimes fit with social demo-
cratic and conservative welfare states. But the unification of welfare and
production typologies also revealed hitherto hidden patterns of how
countries regulate both. Continental European countries, which are
conservatively coordinated, share a policy style of valuing social stabil-
ity. This affects the long-term orientation of their production regime
and the status stabilization of their welfare regime. The Scandinavian
countries share a policy style of being concerned with the nationwide
172
common good. This shows in their production regime, through econ-

omy-wide coordination, and in their welfare regime, which promotes
national redistribution. Anglo-American countries share a policy style
of liberalism, which stresses the value of personal freedom in the wel-
fare and production system. In the production system, this shows in
free markets; in the welfare state, it shows through individuals that are
free to ensure their well-being through private contracts.
By taking up the many open threads of the existing literature, I sub-
stantiated claims that ‘the more basic causal factor [behind capitalist
differences] seems to be differences in the attitudes or orientations
of the relevant political actors, with roots deep in national history.
The great advantage of such analysis is that they are able to capture
a range of attitudinal variables, often cultural in character, that have
especially broad effects over the political economy’ (Hall 1999: 137).
In this sense, what different societies see as rational regulation of
their production and welfare system differs from one regime-type to
another. Continental Europe’s bias towards privileging stability results
from its chaotic and often bloody history of upheaval and war, its
segmentation of countries into social groups and the influence of
the Catholic Church. The demand for individual freedom that liberal
countries share stems from a history of early democratization and a
modernization process that was not managed by an overarching state.
Scandinavian solidarity shows in the welfare state and the production
system and is rooted in a long-lasting view of national unity and few
intra-national cleavages.
In this sense, systematic similarities and differences between countries
influence welfare and production systems. While the stable historical
roots of these welfare/production combinations lead us to suppose that
liberalism will endure in Anglo-American countries, social solidarity in
Scandinavia and conservative social stability in continental Europe, these
historical roots are no guarantee that the association of liberal capitalism
with liberal welfare states, as well as of conservative and social democratic
welfare states with economic coordination will persist. Still, while some
form of liberalization is taking place in all countries, it seems hard to
envision that Anglo-American liberalism turns into egalitarianism. Also,
Scandinavian solidarity would have a long way to go until it would sup-
port widespread inequalities. The conservative social stability bias of the
continental European countries supports a social structure that separates
underprivileged outsiders from privileged insiders, and it is again hard to
imagine how this can turn into egalitarianism or liberalism.
Before closing, a word of caution is due. As scholars of capital-

ist variety mention: ‘The point of generalization is economy of
explanation – to be able to see the forest rather than myriad unique
trees’ (Esping-Andersen 1997: 179). I indeed wanted to show that there
is a ‘forest’ and that its fauna comes in a limited variety. I argued that
by exclusively discerning individual trees, we lose sight of the forest’s
regularities. In this sense, the approach adopted here can rightly be
accused of (over-)simplification. However, it is to be viewed against a
literature that often analyses each country as a specific case. As I high-
lighted throughout, I do not deny that every country is unique. Much
as I could stress that there is a forest somewhere, any skilful observer
could rightly disagree, claiming that it makes no sense to speak of a
forest, but that one should speak of its individual trees. This love of
detail is not wrong. Quite the contrary, it is entirely necessary. But it
needs a counterpoint, which is the analysis of commonalities, against
which uniqueness can be established. For this reason, the approach
adopted here is not an alternative but a complement to country-stud-
ies, which rightly stress that every country is unique.
The immensity of the field shows how limited the scope of this
book was. A literature on the difference between common law and
civil law countries suggests a similar classification of nations as sug-
gested here (cf. La Porta et al. 1997, 1998; La Porta, López-de-Silanes
and Shleifer 2008). Another literature argues that political systems go
along with welfare states and production systems, recommending a
similar grouping of countries (cf. Estevez-Abe, Iversen and Soskice 2001;
Soskice and Iversen 2001; Iversen and Soskice 2006; Cusack, Iversen
and Soskice 2007; Martin and Swank 2012). Both literatures largely had
to be neglected, even though it is perfectly possible that political or
judicial systems bind varieties of capitalism and welfare states together
into a market-prone variety, on the one side and a group-based, as well
as a nationally solidaristic variety, on the other. Incorporating this into
a typology that unifies production, welfare, legal and political systems
would be perfectly compatible with what has been done here. It could
be the next step in typologies of capitalist diversity.
In addition, if different cultures – owing to different structural
conditions – indeed led to country clusters in which the desire for
solidarity, individual liberty and social stability permeate not only the
welfare, but also the production system, then we need to understand
how exactly these different values became important in the differ-
ent country clusters – through different voting systems, as Iversen
et al. (Cusack, Iversen and Soskice 2007) argue, directly as a result of
religious legacies, as Kahl (2005; 2009) argues, through political power

of the labour movement, as Korpi (1985; 2006) argues, or through
as yet unchartered pathways of influence. However, we can only
ask these questions when we know that the welfare and production
regimes of some countries systematically fit, which is the basic point
this book tried to convey.
Notes
Introduction
1. ‘Regime’ is defined here as ‘a set of norms, rules, procedures and institutions
that constrains the behavior of [its] members’ (Esping-Andersen 1990: 2). My
additional definition is that a regime stretches over a number of countries
and extends to families of countries. To give an illustrative example: Sweden
possesses a production system and a welfare state, whereas the group of coor-
dinated market economies constitute a production regime and the group of
social democratic countries a welfare regime.
My definition of a ‘production system’ is borrowed from Soskice (1999:
101f.): ‘By a production system is meant the organization of production
through markets and market-related institutions. It analyzes the ways in
which the microagents of capitalist systems – companies, customers, employ-
ees, owners of capital – organize and structure their relationships, within a
framework of incentives and constraints of “rules of the game” [reference
to North 1990 and Aoki 1994] set by a range of market-related institutions
within which the microagents are embedded. […] The most important of the
institutions contributing to the institutional frameworks are the financial
system, the industrial relations system, the education and training system, and
the intercompany system (the latter governing relations between companies –
competition policy, technology transfer, standard setting and so on)’.
I follow Esping-Andersen (1999: 73) for my definition of welfare arrange-
ments: ‘the ways in which welfare production is allocated in between state,
markets and households’. This includes decommodification, financing
modes, rights and entitlements.
2. I refer to Hall and Soskice’s (2001b) introduction, when I refer to the ‘varieties
of capitalism’ typology. I do not mean other typologies that are sometimes
grouped under this label, for example Amable (2003) or Boyer (2004b, 2005),
but I will address these in later chapters.
3. The grouping of countries in welfare regimes that I will use is based on
Esping-Andersen’s (1999: 20, 77) refined typology in Social Foundations of
Postindustrial Economies. Japan and the Netherlands are classified somewhat
differently in his first book (cf. 1990). In this book, both countries are seen as
conservative welfare states. The grouping of countries for production systems
follows Hall and Soskice (2001b: 19f.).
4. For a further explanation of the terms used, see sections 1.1 and 1.2.
5. This obviously means that many countries will have to be left out of the
picture. This is mainly because they have hardly been categorized (but com-
pare Witt 2010; Redding and Witt 2009 for China; Adam, Kristan and Tomsic
2009; Cook 2010; Lane and Myant 2007; Nölke and Vliegenthart 2009 for
Eastern Europe; Schneider 2009 for Latin America; Tipton 2009 for Asia).
6. My definition of an ‘ideal type’ is borrowed from Max Weber (1949: 90, 43):
‘An ideal type is formed by the one-sided accentuation of one or more points
176
Notes 177
of view and by the synthesis of a great many diffuse, discrete, more or less
present and occasionally absent concrete individual phenomena, which are
arranged according to those one-sidedly emphasized viewpoints into a uni-
fied analytical construct. […] Its function is the comparison with empirical
reality in order to establish its divergences or similarities, to describe them
with the most unambiguously intelligible concepts, and to understand and
explain them causally’.
1 Welfare State Research and Varieties of Capitalism

1. Admittedly, this is a functionalist view. However, it is adequate at the present
stage and will be refined later on.
2. The model of three different welfare states in turn strongly relies on that
developed by Titmuss (1974).
3. Decommodification ‘occurs when a service is rendered as a matter of right, and
when a person can maintain a livelihood without reliance on the market. […]
[T]he concept refers to the degree to which individuals, or families, can uphold
a socially acceptable standard of living independently of market participation’
(Esping-Andersen 1990: 21f., 37).
4. Whenever there are contradictions between Esping-Andersen’s initial classi-
fication (1990) and later works (1999), I will use his most recent publication.
Other important works for the classification of countries include Pierson
(2001a) and Scharpf and Schmidt (2000a).
5. There is some confusion concerning Japan, which was in the ‘liberal’ cluster
in Esping-Andersen’s first typology (1990: 74) and is defended by Esping-
Andersen (1999: 82) as being ‘conservative’ in later refinements of his typol-
ogy. However, Esping-Andersen does not explicitly comment on this change
of Japan to another regime. As mentioned earlier, I will use Esping-Andersen’s
later refinements if there are contradictions. Japan is thus seen as part of the
conservative cluster.
6. There is also some confusion whether the Netherlands should be part of
the ‘social democratic’ or the ‘conservative’ regime (cf. Arts and Gelissen
2002: 151). Esping-Andersen (1990: 74, 1999: 78ff.) sometimes saw the coun-
try as conservative, sometimes as social democratic. Even though Esping-
Andersen’s social democratic group is often seen as a Scandinavian cluster,
some works place the Netherlands in the social democratic group as well
(cf. Goodin et al. 1999), others state that it belongs to the group of conserva-
tive countries (cf. van Kersbergen 1995: 56). I see the Netherlands as pertain-
ing to the conservative cluster, as its level of transfer-payments (as opposed
to direct social security benefits measured as total social expenditure) is at
97.7 per cent, which is one of the highest rates in the OECD. It is therefore
distinctly different from the service-intensive Scandinavian countries, whose
average transfer-spending rate is very high (cf. the Table 4.1 in Huber and
Stephens 2001b: 88). The Netherlands therefore does not belong to the group
of countries that focus on providing social services, but rather to the group
that relies on income-replacement (also cf. Pontusson 2005a: 147f.). As the
difference between service- and transfer-intensive welfare states is seen as a
distinguishing feature between conservative and social democratic welfare
178 Notes
states (cf. Huber and Stephens 1996; also cf. Bernard and Saint-Arnaud 2003),
the Netherlands can confidently be put into the conservative group.
7. My definition of ‘institution’ is borrowed from Peter Hall (1986: 19): ‘The
concept of institutions is used here to refer to the formal rules, compliance
procedures, and standard operating practices that structure the relationship
between individuals in various units of the polity and economy. As such, they
have a more formal status than cultural norms but one that does not neces-
sarily derive from legal, as opposed to conventional standing’.
This has to be understood in the sense of Douglass North (1990: 3): ‘Institutions
are the rules of the game in a society or, more formally, are the human devised
constraints that shape human interaction. In consequence, they structure
incentives in human exchange, whether political, social, or economic’.
8. One prominent exception, where the market does not govern even liberal coun-
tries, is coordination within hierarchies, which structures companies. Such hier-
archies complement markets when they save transaction costs (cf. Williamson
1975; critically: Granovetter 1985).
9. For an in-depth discussion on the subject of coherence and complementarity,
cf. Amable et al. (2005) and Höpner (2005a, b).
2 Empirical Indicators and Existing Typologies

1. All correlations in this book are two-tailed.
2. I thank Vivien Schmidt for helpful insights on this point.
3. This however, is paid for by applying rather rigid variants of path depend-
ency, which in turn are criticized (Crouch 2001; Crouch and Farrell 2002).
4. In so doing, he sees Austria as pertaining to the social democratic and Italy as
pertaining to the state-led cluster (cf. Boyer 1997: 90). This is quite unusual,
since it is inconsistent with Boyer’s later typologies (cf. 2004b). In addition,
Italy is well-known for the weak (formal) regulatory capacity of its state, as
Regini (see especially 1997: 106f.) argues.
5. In contestational systems, unions are sufficiently strong to exert veto power,
for example, by mobilizing workers, but are not encompassing enough to
participate in corporatist bargaining (cf. Crouch 1993).
6. Measured as public social expenditure (cash benefits, provision of goods and
services and tax breaks with social purposes) as share of GDP – data for 2001
(OECD data from 2006, URL: http://thesius.sourceoecd.org/vl=30620998/
cl=14/nw=1/rpsv/factbook/data/09-02-01-t01.xls).
7. An interesting approach for this has been taken by Kenworthy (2006). He
develops a coherency-index for the production system of countries. In terms
of their production system, Switzerland and the Netherlands have a very
incoherent institutional framework.
8. Note that Pontusson subsumes Esping-Andersen’s approach under varieties of
capitalism, whereas I regard it as a distinct one.
3 A Unified Typology of Capitalisms

1. However, like Amable, Boyer (2004b: 20) also sees Norway as sometimes
belonging to the ‘liberal’ cluster.
Notes 179
2. For a call to enlarge the dual varieties of capitalism distinction for much the
same regime types, also compare Hay (2005: 114f.) and Pontusson (2005a:
167).
3. Yet, as Schmidt only provides case studies of three countries (Germany, France
and the United Kingdom), we cannot be sure to what extent her scheme fits
the one drawn out above.
4 How Complementarities Stabilize Three Capitalisms

1. One could argue that in the Mediterranean countries it instead leads to a
highly protected core workforce that might possess firm-internal skills, next
to a precarious outsider group.
2. Source: OECD (data for 2006), URL: http://thesius.sourceoecd.org.
3. Source: Eurostat (data for 2003) and Gough et al. (1997) for non-European
countries.
4. Once labour and capital have reached the point where coordination is real-
ized to be a beneficial constraint to employers, these can even develop a
genuine interest to preserve a strong labour movement in order to have a
bargaining partner that possesses an encompassing power of implementation
for collective bargaining agreements, in which employers have an interest
(Wood 2001b: 378; Thelen 2000).
5. Admittedly, as in many other cases of this stylized description, such a func-
tionalist account simplifies reality.
6. However, causality points to ambiguous directions here. It has been argued
that decommodification is typical of coordinated market economies. It
could thus be said that in coordinated market economies, workers can be
more mobile, as they are less constrained by market forces. Yet it is exactly
because workers are less mobile, as they have longer and less opportunistic
relationships with their employers, that coordinated market economies can
exist (cf. Manow 2001c: 155f.). For coordinated market economies, decom-
modification thus has to be designed in a manner that workers are not
structurally disadvantaged over capital, but at the same time have no inter-
est to use their high degree of decommodification in a way that endangers
long-term work relationships. This contradiction might be the reason why
in many coordinated market economies, prior stable employment records
determine (relatively generous) unemployment-, sickness- and pension-
payments, as it is this complex interplay, which can reconcile long-term
employment relationships and decommodification.
7. As much as this seems to be a win-win situation on first sight, costs are
simply shifted to those that have to finance transfer-incomes: taxpayers
and social-security contributors. The problem of increasing tax- and social-
security burdens is also the reason why programs of early retirement have
been retrenched.
8. Source: OECD (data for 2006), URL: http://thesius.sourceoecd.org.
9. Source: Eurostat (data for 2003) and Gough et al. (1997) for non-European
countries.
10. Years in which chief executive and largest party in congress have left or cen-
tre political orientation from 1928–95 (Porta et al. 2004: 1362f.).
180 Notes
11. Welfare liberalism of 18 welfare states was measured with Esping-Andersen’s

(1990: 74) indicator. Centralized union power was measured by the indicator
in Martin and Swank (2001: 899).
12. Labour decommodification has been taken from Esping-Andersen (1990:
50). The degree of coordination for different economies has been taken from
Hall and Gingerich (2004: 14).
13. The degree of employment protection is taken from the OECD (2004: 117).
Economic coordination is again measured by Hall and Gingerich (2004: 14).
14. The pension replacement rate of countries has been measured by net manda-
tory pension programme replacement rates as percentage of pre-retirement
net earnings for men earning an average wage (OECD 2005a). The index
used to measure coordination is again from Hall and Gingerich (2004: 14).
15. Public social expenditure is defined as cash benefits, provision of goods and
services and tax breaks with social purposes as listed by the OECD (2006).
For coordination, cf. Hall and Gingerich (2004: 14) again.
5 Diversity’s Source: Three Policy Styles, Three Capitalisms

1. This started with the 1907 Workmen’s Compensation Act for workers’ inju-
ries; the 1908 Labour Exchanges Act provided help to find employment; the
1908 Old-Age Pensions Act provided a non-contributory pension for men and
women over the age of 70; the 1911 National Insurance Act provided health
and unemployment insurance for some workers, in 1920 it was extended to
‘all nonpermanent manual and nonmanual workers earning less than £250
per year’ (Lavalette and Penketh 2003: 64; Flora and Alber 1981: 59).
2. Probably no country is more famous for its inability to forge a nationally
organized response of wage restraint to preserve business profitability in the
face of rising oil prices. Instead, typical for a liberal country, the fragmented
union movement of the United Kingdom was incapable of coordinating for
necessary wage restraint (Rhodes 2000: 33ff.; King and Wood 1999).
6 What can a Unified Typology Explain?

1. This comprises the following categories: directors and chief executives (ISIC
category 121), production and operations managers (122), other specialist
managers (123) and managers of small enterprises (131).
2. This section owes much to the most recent work of Kathleen Thelen (2012: 143f.)
and her distinction between ‘sector coordinated’ CMEs, like Germany, ‘national
CMEs’, like the Scandinavian countries and liberal market economies (for an
essentially similar description, cf. Martin and Swank 2012: 7). This distinc-
tion mirrors my separation into liberal, conservatively coordinated and social
democratically coordinated countries.
References
Abrahamson, Peter (1999): ‘The Welfare Modeling Business’. Social Policy and
Administration, (33)4, 394–415.
Abrahamson, Peter (2003): ‘The Welfare State in Denmark’. In Christian Aspalter
(ed.) Welfare Capitalism around the World. Taichung: Casa Verde Publishing, pp.
113–37.
Adam, Frane, Kristan, Primoz and Tomsic, Matevz (2009): ‘Varieties of Capitalism
in Eastern Europe (with special emphasis on Estonia and Slovenia)’. Communist
and Post-Communist Studies, 42, 65–81.
Aglietta, Michel (1976): Régulation et Crises du Capitalisme. Paris: Calman-Lévy.
Ahn, Jae-Hung (1996): ‘Ideology and Interest: The Case of Swedish Social
Democracy, 1886–1911’. Politics & Society, 24, 153–87.
Alber, Jens (1981): ‘Government Responses to the Challenge of Unemployment:
The Development of Unemployment Insurance in Western Europe’. In Peter
Flora and Arnold J. Heidenheimer (eds) The Development of Welfare States in
Europe and America. New Brunswick, NJ: Transaction Books, pp. 151–83.
Albert, Michel (1992): Kapitalismus contra Kapitalismus. Frankfurt and New York:
Campus.
Alcock, Pete and Craig, Gary (2001): ‘The United Kingdom: Rolling Back
the Welfare State?’ In Pete Alcock and Gary Craig (eds) International Social
Policy. Welfare Regimes in the Developed World. New York: Palgrave Macmillan,
pp. 124–42.
Alford, Robert R. (1967): ‘Class Voting in the Anglo-American Political Systems’.
In Seymour Martin Lipset and Stein Rokkan (eds) Party Systems and Voter
Alignments: Cross-National Perspectives. New York: Free Press, pp. 67–94.
Allan, James and Lyle, Scruggs (2002): ‘Politics As Usual? Political Parties and
Welfare Regime Change in Advanced Industrial Societies.’ Presented to the
Thirteenth International Conference of Europeanists, Chicago.
Allan, James and Scruggs, Lyle (2004): ‘Political Partisanship and Welfare State
Reform in Advanced Industrial Societies’. American Journal of Political Science,
48(3), 496–512.
Allardt, Erik and Pesonen, Pertti (1967): ‘Cleavages in Finnish Politics’. In Seymour
Martin Lipset and Stein Rokkan (eds) Party Systems and Voter Alignments: Cross-
National Perspectives. New York: Free Press, pp. 325–66.
Allen, Kieran (2000): The Celtic Tiger: The Myth of Social Partnership in Ireland.
Manchester, New York: Manchester University Press.
Amable, Bruno (2003): The Diversity of Modern Capitalism. Oxford: Oxford
University Press.
Amable, Bruno, Boyer, Robert, Crouch, Colin, Hall, Peter, Jackson, Gregory and
Streeck, Wolfgang (2005): ‘Dialogue on “Institutional Complementarity and
Political Economy”’. Socio-Economic Review, 3(2), 359–82.
Anderson, Karen (2001): ‘The Politics of Retrenchment in a Social Democratic
Welfare State’. Reform of Swedish Pensions and Unemployment Insurance.
Comparative Political Studies, 34(9), 1063–91.
181
182 References
Anderson, Karen M. (2009): ‘The Church as Nation? The Role of Religion in the
Development of the Swedish Welfare State’. In Kees van Kersbergen and Philip
Manow (eds) Religion, Class Coalitions, and Welfare States, 1st edn. Cambridge:
Cambridge University Press, pp. 210–35.
Aoki, Masahiko (1994): ‘The Japanese Firms as a System of Attributes: A Survey
and Research Agenda’. In Masahiko Aoki and Ronald Dore (eds) The Japanese
Firm: Sources of Competitive Strength. Oxford: Clarendon Press, pp. 11–40.
Aoki, Masahiko (1998): ‘Unintended Fit: Organizational Evolution and
Government Design of Institutions in Japan’. In Masahiko Aoki, Hyung-Ki
Kim and Masahiro Okuno-Fujiwara (eds) The Role of Government in East Asian
Economic Development: Comparative Institutional Analysis. Oxford: Clarendon
Press, pp. 233–53.
Aoki, Masahiko (2000): Information, Corporate Governance and Institutional
Diversity. Oxford: Oxford University Press.
Aoki, Masahiko and Dore, Ronald (1994): The Japanese Firm: Sources of Competitive
Strength. Oxford: Clarendon Press.
Aoki, Masahiko, Kim, Hyung-Ki and Masahiro Okuno-Fujiwara (1998): The Role
of Government in East Asian Economic Development: Comparative Institutional
Analysis. Oxford: Clarendon Press.
Arcq, Étienne and Pochet, Philippe (2000): ‘Toward a New Social Pact in
Belgium?’ In Giuseppe Fajertag and Philippe Pochet (eds) Social Pacts in Europe.
Brussels: European Trade Union Institute, pp. 113–34.
Arriba, Ana and Moreno, Luis (2005): ‘Spain – Poverty, Social Exclusion and
“Safety Nets’’. In Maurizio Ferrera (ed.) Welfare State Reform in Southern Europe.
Fighting Poverty and Social Exclusion in Italy, Spain, Portugal and Greece. New York:
Routledge, pp. 141–95.
Arts, Wil and Gelissen, John (2001): ‘Welfare States, Solidarity and Justice
Principles: Does the Type Really Matter?’ Acta Sociologica, 44, 283–99.
Arts, Wil and Gelissen, John (2002): ‘Three Worlds of Welfare Capitalism or More?
A State-of-the-Art Report’. Journal of European Social Policy, 12(2), 137–58.
Arts, Wil and Gelissen, John (2010): ‘Models of the Welfare State’. In Francis
G. Castles, Stephan Leibfried, Jane Lewis, Herbert Obinger and Christopher
Pierson (eds) The Oxford Handbook of the Welfare State. Oxford: Oxford
University Press, pp. 569–83.
Aspalter, Christian and Lai, On-kwok (2003): ‘Welfare Capitalism in Japan:
Past, Recent, and Future Developments’. In Christian Aspalter (ed.) Welfare
Capitalism around the World. Taichung: Casa Verde Publishing, pp. 241–58.
Baccaro, Lucio and Howell, Chris (2011): Reformulating the Argument for Neoliberal
Convergence: Institutional Change in European Industrial Relations. Unpublished
manuscript.
Baldwin, Peter (1989): ‘The Scandinavian Origins of the Social Interpretation of
the Welfare State’. Comparative Studies in Society and History, 31, 3–24.
Baldwin, Peter (1990): The Politics of Social Solidarity. Class Bases of the European
Welfare State 1875–1975. Cambridge: Cambridge University Press.
Banfield, Edward (1958): The Moral Basis of a Backward Society. New York: The
Free Press.
Barreto, José and Naumann, Reinhard (2000): ‘Portugal: Industrial Relations
under Democracy’. In Anthony Ferner and Richard Hyman (eds) Changing
Industrial Relations in Europe, 2nd edn. Oxford: Blackwell, pp. 395–425.
References 183
Becker, Uwe (2007): ‘Open Systemness and Contested Reference Frames and
Change. A Reformulation of the Varieties of Capitalism Theory’. Socio-Economic
Review, 5, 261–86.
Beckert, Jens (1996): ‘What is Sociological about Economic Sociology? Uncertainty
and the Embeddedness of Economic Action’. Theory and Society, 25(6), 803–40.
Bentes, Margarida, Dias, Carlos Matias, Sakellarides, Constantino and Bankauskaite,
Vaida (2004): Health Care Systems in Transition: Portugal. Copenhagen: WHO
Regional Office for Europe on behalf of the European Observatory on Health
Systems and Policies.
Beramendi, Pablo and Rueda, David (2008): Inequality and Labor Market
Coordination in the Early 20th Century. Paper presented at the Annual Meeting
of the American Political Science Association. Boston.
Bergholm, Tapio (2009): ‘The Making of the Swedish Model’. Scandinavian Journal
of History 34, 29–48. http://dx.doi.org/10.1080/03468750902770931.
Bermeo, Nancy (2001): Unemployment in the New Europe. New York: Cambridge
University Press.
Bernard, Paul and Saint-Arnaud, Sébastien (2003): ‘More of the Same?
A Hierarchical Cluster Analysis of the Welfare Regimes in Advanced Countries’.
Current Sociology, 51, 499–527.
Beyer, Jürgen (2002): Deutschland AG a.D.: Deutsche Bank, Allianz und das
Verflechtungszentrum großer deutscher Unternehmen. MPIfG Working Paper 02/04.
Cologne: Max Planck Institute for the Study of Societies. Available at: http://
www.mpi-fg koeln.mpg.de/pu/workpap/wp02-4/wp02-4.html – downloaded
26 April 2006.
Birch, Anthony Harold (1955): Federalism, Finance, and Social Legislation in
Canada, Australia, and the United States. Oxford: Clarendon Press.
Bispinck, Reinhard and Schulte, Thorsten (2000): ‘Alliance for Jobs – Is Germany
following the Path of “Competitive” Corporatism?’ In Giuseppe Fajertag and
Philippe Pochet (eds) Social Pacts in Europe. Brussels: European Trade Union
Institute, pp. 187–217.
Bjørnson, Øyvind (2001): ‘The Social Democrats and the Norwegian Welfare
State: Some Perspectives’. Scandinavian Journal of History, 26, 197–223.
Blekesaune, Morten and Quadagno, Jill (2003): ‘Public Attitudes toward Welfare
State Policies’. European Sociological Review, 19, 415–27.
Block, Richard N. (2006): ‘Industrial Relations in the United States and Canada’.
In Michael Morley, Patrick Gunnigle and David G. Collings (eds) Global
Industrial Relations. Routledge Global Human Resource Management Series.
London: Routledge, pp. 25–52.
Böckerman, Petri and Uusitalo, Roope (2006): ‘Erosion of the Ghent System and
Union Membership Decline: Lessons from Finland’. British Journal of Industrial
Relations, 44, 283–303.
Bode, Ingo (2003): ‘The Welfare State in Germany: Corporatism and the German
Welfare Model’. In Christian Aspalter (ed.) Welfare Capitalism around the World.
Taichung: Casa Verde Publishing, pp. 157–78.
Bonoli, Giuliano (1997): ‘Classifying Welfare States: A Two Dimensional
Approach’. Journal of Social Policy, 26(3), 351–72.
Bonoli, Giuliano (2003): ‘The Welfare State in Switzerland’. In Christian Aspalter
(ed.) Welfare Capitalism around the World. Taichung: Casa Verde Publishing,
pp. 179–97.
184 References
Börsch, Alexander (2007): ‘Institutional Variation and Coordination Patterns

in CMEs. Swiss and German Corporate Governance in Comparison’. In Bob
Hancké, Martin Rhodes and Mark Thatcher (eds) Beyond Varieties of Capitalism.
Conflict, Contradictions and Complementarities in the European Economy, 1st edn.
Oxford: Oxford University Press, pp. 173–94.
Boston, Jonathan (1999): ‘New Zealand’s Welfare State in Transition’. In
Jonathan Boston, Paul Dalziel and Susan St. John (eds) Redesigning the Welfare
State in New Zealand. Problems, Policies, Prospects. Oxford: Oxford University
Press, pp. 3–19.
Boyer, Robert (1997): ‘French Statism at the Crossroads’. In Colin Crouch
and Wolfgang Streeck (eds) Political Economy of Modern Capitalism: Mapping
Convergence and Diversity. London: Sage, pp. 71–101.
Boyer, Robert (2004a): Théorie de la Régulation. Les Fondamentaux. Paris: La
Découverte.
Boyer, Robert (2004b): How and Why Capitalisms Differ. MPIfG Discussion Paper
04/5. Cologne: Max Planck Institute for the Study of Societies. Available at:
http://www.mpifg.de/pu/mpifg_dp/dp05-4.pdf – downloaded 22 October
2005.
Boyer, Robert (2004c): Une Théorie du Capitalisme est-elle Possible? Paris: Odile
Jacob.
Boyer, Robert (2005): ‘How and Why Capitalisms Differ’. Economy and Society
34, 509–57.
Boyer, Robert and Hollingsworth, J. Rogers (1997): Contemporary Capitalism. The
Embeddedness of Institutions. Cambridge: Cambridge University Press.
Bradshaw, Jonathan, Ditch, John, Eardley, Tony and Gough, Ian (1997): ‘Social
Assistance in OECD Countries’. Journal of European Social Policy, 7(1), 17–43.
Brodie, Janine (2003): ‘The Welfare State in Canada’. In Christian Aspalter
(ed.) Welfare Capitalism around the World. Taichung: Casa Verde Publishing,
pp. 9–26.
Brooks, Clem and Manza, Jeff (2006): ‘Social Policy Responsiveness in Developed
Democracies’. American Sociological Review, 71, 474–94.
Brooks, Clem and Manza, Jeff (2007): Why Welfare States Persist: The Importance of
Public Opinion in Democracies. Chicago: University of Chicago Press.
Bryson, Lois (2001): ‘Australia: The Transformation of the Wage Earners’ Welfare
State’. In Pete Alcock and Gary Craig (eds) International Social Policy. Welfare
Regimes in the Developed World. New York: Palgrave Macmillan, pp. 64–84.
Bueck, Henry Axel (1905): Der Zentralverband Deutscher Industrieller, 1876–1901.
Berlin, Deutscher Verlag.
Bührer, Werner and Grande, Edgar (2000): Unternehmerverbände und Staat in
Deutschland. Baden-Baden: Nomos.
Burke, Helen (1999): ‘Foundation Stones of Irish Social Policy, 1831–1951’. In
Gabriel Kiely, Anne O’Donnell, Patricia Kennedy and Suzanne Quin (eds) Irish
Social Policy in Context. Dublin: Dublin University Press, pp. 11–32.
Busch, Andreas (1993): ‘The Politics of Price Stability: Why the German-Speaking
Nations are Different’. In Francis Castles (ed.) Families of Nations: Public Policy in
Western Democracies. Brookfield, VT: Dartmouth University Press, pp. 35–91.
Busemeyer, Marius (2009): ‘Asset Specificity, Institutional Complementarities
and the Variety of Skill Regimes in Coordinated Market Economies’. Socio-
Economic Review, 7, 375–406.
References 185
Busemeyer, Marius and Trampusch, Christine (2011): ‘Introduction: The

Comparative Political Economy of Collective Skill Formation’. In Marius
Busemeyer and Christine Trampusch (eds) The Political Economy of Collective
Skill Formation. Oxford: Oxford University Press, pp. 3–40.
Calmfors, Lars and John Driffill (1988): ‘Bargaining Structure, Corporatism and
Macroeconomic Performance’. Economic Policy, 3, 13–61.
Campbell, John, Hollingsworth, Roger and Lindberg, Leon (1991): Governance of
the American Economy. Cambridge: Cambridge University Press.
Capucha, Luis, Bomba, Teresa, Fernandes, Rita and Matos, Gisela (2005):
‘Portugal – A Virtuous Path Towards Minimum Income?’ In Maurizio Ferrera
(ed.) Welfare State Reform in Southern Europe. Fighting Poverty and Social Exclusion
in Italy, Spain, Portugal and Greece. New York: Routledge, pp. 204–59.
Casey, Terence (2006): Mapping European Capitalisms and the Challenge of Social
Context. Paper presented at the Fifteenth International Conference of the
Council for European Studies, Chicago, IL, 29 March–2 April, 2006.
Casper, Steven (2002): ‘National Institutional Frameworks and High-
Technology in Germany: The Case of Biotechnology’. In Karl Müller, Ellen
Hollingsworth and Roger Hollingsworth (eds) Advancing Socio-Economics: An
Institutionalist Perspective. Lanham, MD: Rowman & Littlefield Publishers,
pp. 277–305.
Castles, Francis (1993a): Families of Nations: Public Policy in Western Democracies.
Brookfield, VT: Dartmouth University Press.
Castles, Francis (1993b): ‘Changing Course in Economic Policy: The English-
Speaking Nations in the 1980s’. In Francis Castles (ed.) Families of Nations:
Public Policy in Western Democracies. Brookfield, VT: Dartmouth University
Press, pp. 3–34.
Castles, Francis (1993c): ‘Introduction’. In Francis Castles (ed.) Families of Nations:
Public Policy in Western Democracies. Brookfield, VT: Dartmouth University Press,
pp. xiii–xxiii.
Castles, Francis (1996): ‘Needs-Based Strategies of Social Protection in Australia
and New Zealand’. In Gøsta Esping-Anderson (ed.) Welfare States in Transition.
National Adaptations in Global Economies. London: Sage, pp. 88–115.
Castles, Francis (2010): ‘The English-Speaking Countries’. In Francis G. Castles,
Stephan Leibfried, Jane Lewis, Herbert Obinger and Christopher Pierson (eds)
The Oxford Handbook of the Welfare State. Oxford: Oxford University Press,
pp. 630–42.
Castles, Francis, Gerritsen, Rolf and Vowles, Jack (eds) (1996): The Great
Experiment: Labour Parties and Public Policy Transformation in Australia and New
Zealand. Auckland: Auckland University Press.
Castles, Francis and Mitchell, Deborah (1993): ‘Worlds of Welfare and Families
of Nations’. In Francis Castles (ed.) Families of Nations: Public Policy in Western
Democracies. Brookfield, VT: Dartmouth University Press, pp. 93–128.
Castles, Francis G. and Obinger, Herbert (2008): ‘Worlds, Families, Regimes:
Country Clusters in European and OECD Area Public Policy’. West European
Politics, 31, 321–44.
Cerny, Philip (1997): ‘International Finance and the Erosion of Capitalist
Diversity’. In Colin Crouch and Wolfgang Streeck (eds) Political Economy
of Modern Capitalism: Mapping Convergence and Diversity. London: Sage,
pp. 173–81.
186 References
Cerny, Philip, Menz, Georg and Soederberg, Susanne (2005a): Internalizing

Globalization: The Rise of Neoliberalism and the Decline of National Varieties of
Capitalism. New York: Palgrave Macmillan.
Cerny, Philip, Menz, Georg and Soederberg, Susanne (2005b): ‘Different Roads
to Globalization: Neoliberalism, the Competition State, and Politics in a More
Open World’. In Susanne Soederberg, Georg Menz and Philip Cerny (eds)
Internalizing Globalization: The Rise of Neoliberalism and the Decline of National
Varieties of Capitalism. New York: Palgrave Macmillan, pp. 1–30.
Cheyne, Christine, O’Brien, Mike and Belgrave, Michael (2008): Social Policy in
Aotearoa/New Zealand. South Melbourne: Oxford University Press.
Christensen, Tom (2005): ‘The Norwegian State Transformed?’ West European
Politics, 28, 721–39.
Chuliá, Elisa and Asensio, María (2007): ‘Portugal: In Search of a Stable
Framework’. In Ellen M. Immergut, Karen M. Anderson and Isabelle Schulze
(eds) The Handbook of West European Pension Politics. Oxford: Oxford University
Press, pp. 605–59.
Clarke, John and Piven, Frances Fox (2001): ‘United States: An American Welfare
State?’ In Pete Alcock and Gary Craig (eds) International Social Policy. Welfare
Coates, David (2005): Varieties of Capitalism, Varieties of Approaches. New York:
Palgrave Macmillan.
Coleman, James (1990): Foundations of Social Theory. Cambridge, MA: Harvard
University Press.
Conroy, Pauline (1999): ‘From the Fifties to the Nineties: Social Policy Comes
Out of the Shadows’. In Gabriel Kiely, Anne O’Donnell, Patricia Kennedy and
Suzanne Quin (eds) Irish Social Policy in Context. Dublin: Dublin University
Press, pp. 33–50.
Cook, Linda J. (2010): ‘Eastern Europe and Russia’. In Francis G. Castles, Stephan
Leibfried, Jane Lewis, Herbert Obinger and Christopher Pierson (eds) The Oxford
Handbook of the Welfare State. Oxford: Oxford University Press, pp. 671–86.
Cousins, Mel (1997): ‘Ireland’s Place in the Worlds of Welfare Capitalism’. Journal
of European Social Policy, 7, 223–35.
Cox, Robert Henry (2001): ‘The Social Construction of an Imperative – Why
Welfare Reform Happened in Denmark and the Netherlands But Not in
Germany’. World Politics, 53, 463–98.
Crouch, Colin (1993): Industrial Relations and European State Traditions. Oxford:
Oxford University Press.
Crouch, Colin (2000): ‘Die Vielfalt des europäischen Verbandswesens’. In Werner
Bührer and Edgar Grande (eds) Unternehmerverbände und Staat in Deutschland.
Baden-Baden: Nomos, pp. S23–35.
Crouch, Colin (2001): ‘Welfare State Regimes and Industrial Relations Systems.
The Questionable Role of Path Dependency Theory’. In Bernhard Ebbinghaus
and Philip Manow (eds) Varieties of Welfare Capitalism. London: Routledge,
pp. 105–24.
Crouch, Colin (2005a): ‘Models of Capitalism’. New Political Economy, 10(4),
439–56.
Crouch, Colin (2005b): Capitalist Diversity and Change: Recombinant Governance
and Institutional Entrepreneurs. Oxford and New York: Oxford University
Press.
References 187
Crouch, Colin and Farrell, Henry (2002): Breaking the Path of Institutional Development?
Alternatives to the New Determinism. MPIfG Discussion Paper 02/5. Cologne: Max
Planck Institute for the Study of Societies. Available at: http://www.mpi-fg-koeln.
mpg.de/pu/abstracts/dp02-5.html – downloaded 14 September 2004.
Crouch, Colin, LeGalès, Patrick, Trigilia, Carlo and Voelzkow, Helmut (2001):
Local Production Systems in Europe: Rise or Demise? Oxford: Oxford University
Press.
Crouch, Colin, LeGalès, Patrick, Trigilia, Carlo and Voelzkow, Helmut (2004):
Changing Governance of Local Economies: Responses of European Local Production
Systems. Oxford: Oxford University Press.
Crouch, Colin, Schröder, Martin and Voelzkow, Helmut (2009): ‘Regional and
Sectoral Varieties of Capitalism’. Economy and Society, 38, 654–78.
Crouch, Colin and Streeck, Wolfgang (1997a): Political Economy of Modern
Capitalism: Mapping Convergence and Diversity. London: Sage.
Crouch, Colin and Streeck, Wolfgang (1997b): ‘Introduction. The Future of Capitalist
Diversity’. In Colin Crouch and Wolfgang Streeck (eds) Political Economy of Modern
Capitalism: Mapping Convergence and Diversity. London: Sage, pp. 1–18.
Crouch, Colin and Voelzkow, Helmut (eds) (2009): Innovation in Local Economies.
Germany in Comparative Context. Oxford: Oxford University Press.
Cusack, Thomas, Iversen, Torben and Soskice, David (2007): ‘Economic Interests
and the Origins of Electoral Systems’. American Political Science Review, 101,
373–91.
Czada, Roland and Lütz, Susanne (2000): Die politische Konstitution von Märkten.
Opladen: Westdeutscher Verlag.
D’Iribarne, Philippe (1997): ‘A Check to Enlightened Capitalism’. In Colin
Crouch and Wolfgang Streeck (eds) Political Economy of Modern Capitalism:
Mapping Convergence and Diversity. London: Sage, pp. 161–72.
Davey, Judith (1999): ‘New Zealand: The Myth of Egalitarianism’. In Jonathan
Boston, Paul Dalziel and Susan St. John (eds) Redesigning the Welfare State in
New Zealand. Problems, Policies, Prospects. Oxford: Oxford University Press,
pp. 85–103.
De Beer, Paul, Vrooman, Cok and Wildeboer, Jean Marie (2001): On Worlds of
Welfare: Institutions and their Effects in Eleven Welfare States. The Hague: Social
and Cultural Planning Office.
Deeg, Richard (2001): Institutional Change and the Uses and Limits of Path Dependency:
The Case of German Finance. MPIfG Discussion Paper 01/6. Cologne: Max Planck
Institute for the Study of Societies. Available at: http://www.mpi-fg-koeln.mpg.
de/pu/mpifg_dp/dp01-6.pdf – downloaded 14 March 2006.
Deeg, Richard and Jackson, Gregory (2006): How Many Varieties of Capitalism?
Comparing the Comparative Institutional Analyses of Capitalist Diversity. MPIfG
Discussion Paper 06/2. Cologne: Max Planck Institute for the Study of Societies.
Available at: http://www.mpifg.de/pu/mpifg_dp/dp06-2.pdf – downloaded 14
March 2006.
Deeg, Richard and Jackson, Gregory (2007): ‘Towards a More Dynamic Theory of
Capitalist Variety’. Socio-Economic Review, 5, 149–79.
Dobbin, Frank (1994): Forging Industrial Policy. The United States, Britain, and
France in the Railway Age. Cambridge: Cambridge University Press.
Dobbin, Frank and Jung, Jiwook (2010): ‘The Mis-application of Mr. Michael
Jensen: How Agency Theory Brought Down The Economy and Why It Might
188 References
Again’. In Michael Lounsbury and Paul M. Hirsch (eds) Markets on Trial: The
Economic Sociology of the U.S. Financial Crisis. Bingley: Emerald, pp. 29–64.
Doherty, Eileen M. (2001): ‘Globalization, Social Partnership, and Industrial
Relations in Ireland’. In Christopher Candland and Rudra Sil (eds) The Politics
of Labor in a Global Age. Continuity and Change in Late-Industrializing and Post-
Socialist Economies. Oxford: Oxford University Press, pp. 29–64.
Dølvik, Jon Erik and Martin, Andrew (2000): ‘A Spanner in the Works and Oil on
Troubled Waters: The Divergent Fates of Social Pacts in Sweden and Norway’.
In Giuseppe Fajertag and Philippe Pochet (eds) Social Pacts in Europe. Brussels:
European Trade Union Institute, pp. 279–319.
Dore, Ronald (1997): ‘The Distinctiveness of Japan’. In Colin Crouch and
Wolfgang Streeck (eds) Political Economy of Modern Capitalism: Mapping
Convergence and Diversity. London: Sage, pp. 19–32.
Douglas, Mary (1982): In the Active Voice. London: Routledge.
Ebbinghaus, Bernhard (1999): ‘Does a European Social Model Exist and Can It
Survive?’ In Gerhard Huemer, Michael Mesch and Franz Traxler (eds) The Role
of Employer Associations and Labour Unions in the EMU: Institutional Requirements
for European Economic Policies. Aldershot: Ashgate, pp. 1–26.
Ebbinghaus, Bernhard (2001): ‘When Labour and Capital Collude: The Political
Economy of Early Retirement in Europe, Japan and the USA’. In Bernhard
Ebbinghaus and Philip Manow (eds) Varieties of Welfare Capitalism. London:
Ebbinghaus, Bernhard (2006): Reforming Early Retirement in Europe, Japan and the
USA. Oxford: Oxford University Press.
Ebbinghaus, Bernhard and Manow, Philip (2001a): Varieties of Welfare Capitalism.
London: Routledge.
Ebbinghaus, Bernhard and Manow, Philip (2001b): ‘Varieties of Welfare
Capitalism. An Outlook on Future Directions of Research’. In Bernhard
Ebbinghaus and Philip Manow (eds) Varieties of Welfare Capitalism. London:
Edling, Nils (2006): ‘Limited Universalism: Unemployment Insurance in
Northern Europe 1900–2000’. In Niels Finn Christiansen, Klaus Petersen, Nils
Edling, and Per Haave (eds) The Nordic Model of Welfare: A Historical Reappraisal.
Copenhagen: Museum Tusculanum Press, pp. 99–143.
EIRO (2004): Industrial Relations in the EU, Japan and USA, 2002. Dublin:
Eurofound. Available at: http://www.eurofound.europa.eu/eiro/2005/02/
feature/tn0502102f.htm – downloaded 21 January 2013.
EIRO (2009a): Austria: Industrial Relations Profile. Dublin: Eurofound. Available
at: http://www.eurofound.europa.eu/eiro/country/austria.pdf – downloaded
21 January 2013.
EIRO (2009b): Belgium: Industrial Relations Profile. Dublin: Eurofound. Available
at: http://www.eurofound.europa.eu/eiro/country/belgium.pdf – downloaded
21 January 2013.
EIRO (2009c): Denmark: Industrial Relations Profile. Dublin: Eurofound. Available
at: http://www.eurofound.europa.eu/eiro/country/denmark.pdf – downloaded
21 January 2013.
EIRO (2009d): Finland: Industrial Relations Profile. Dublin: Eurofound. Available
at: http://www.eurofound.europa.eu/eiro/country/finland.pdf – downloaded
21 January 2013.
References 189
EIRO (2009e): France: Industrial Relations Profile. Dublin: Eurofound. Available at:
http://www.eurofound.europa.eu/eiro/country/france.pdf – downloaded 21
January 2013.
EIRO (2009f): Germany: Industrial Relations Profile. Dublin: Eurofound. Available
at: http://www.eurofound.europa.eu/eiro/country/germany.pdf – downloaded
21 January 2013.
EIRO (2009g): Italy: Industrial Relations Profile. Dublin: Eurofound. Available
at: http://www.eurofound.europa.eu/eiro/country/italy.pdf – downloaded 21
January 2013.
EIRO (2009h): The Netherlands: Industrial Relations Profile. Dublin: Eurofound.
Available at: http://www.eurofound.europa.eu/eiro/country/netherlands.pdf –
downloaded 21 January 2013.
EIRO (2009i): Portugal: Industrial Relations Profile. Dublin: Eurofound. Available
at: http://www.eurofound.europa.eu/eiro/country/portugal.pdf – downloaded
21 January 2013.
EIRO (2009j): Sweden: Industrial Relations Profile. Dublin: Eurofound. Available
at: http://www.eurofound.europa.eu/eiro/country/sweden.pdf – downloaded
21 January 2013.
EIRO (2009k): Wage Formation in the EU. Dublin: Eurofound. Available at:
http://www.eurofound.europa.eu/docs/eiro/tn0808019s/tn0808019s.pdf –
Ellguth, Peter and Kohaut, Susanne (2011): ‘Tarifbindung und betriebliche
Interessenvertretung: Aktuelle Ergebnisse aus dem IAB Betriebspanel 2010’.
WSI Mitteilungen, 64, 242–47.
Ertman, Thomas (1997): Birth of the Leviathan. Building States and Regimes in
Medieval and Early Modern Europe. New York: Cambridge University Press.
Ervasti, Heikki, Fridberg, Torben, Hjerm, Mikael, Kangas, Olli and Ringdal,
Kristen (2008): ‘The Nordic Model’. In Heikki Ervasti, Torben Fridberg, Mikael
Hjerm, and Kristen Ringdal (eds) Nordic Social Attitudes in a European Perspective.
Cheltenham: Edward Elgar, pp. 1–21.
Esping-Andersen, Gøsta (1985): Politics Against Markets: The Social Democratic
Road to Power. Princeton: Princeton University Press.
Esping-Andersen, Gøsta (1990): The Three Worlds of Welfare Capitalism. Cambridge:
Polity Press.
Esping-Andersen, Gøsta (1996): Welfare States in Transition. National Adaptations
in Global Economies. London: Sage.
Esping-Andersen, Gøsta (1997): ‘Hybrid or Unique?: The Japanese Welfare State
Between Europe and America’. Journal of European Social Policy, 7, 179–89.
Esping-Andersen, Gøsta (1999): Social Foundations of Postindustrial Economies.
New York: Oxford University Press.
Estevez-Abe, Margarita (2005): ‘Gender Bias in Skills and Social Policies: The
Varieties of Capitalism Perspective on Sex Segregation’. Social Politics 12,
180–215. Available at: ISI://WOS:000231658800003.
Estevez-Abe, Margarita (2006): ‘Gendering the Varieties of Capitalism – A Study
of Occupational Segregation by Sex in Advanced Industrial Societies. World
Politics 59, 142–75. Available at: ISI://000246731900005.
Estevez-Abe, Margarita (2009): ‘Gender, Inequality, and Capitalism: The
“Varieties of Capitalism” and Women’. Social Politics 16, 182–91. Available at:
ISI://000266970200002.
190 References
Estevez-Abe, Margarita, Iversen, Torben and Soskice, David (2001): ‘Social

Protection and the Formation of Skills: A Reinterpretation of the Welfare State’.
In Peter Hall and David Soskice (eds) ‘Varieties of Capitalism’: The Institutional
Foundations of Comparative Advantage. Oxford: Oxford University Press,
pp. S145–83.
Fargion, Valeria (2001): ‘Italy: Moving from the Southern Model’. In Pete Alcock
and Gary Craig (eds) International Social Policy. Welfare Regimes in the Developed
World. New York: Palgrave Macmillan, pp. 183–202.
Ferge, Zsuzsa and Kolberg, Jon (1992): Social Policy in a Changing Europe. Frankfurt:
Campus-Westview.
Ferrera, Maurozio (1996): ‘The “Southern Model” of Welfare in Social Europe’.
Journal of European Social Policy, 6(1), 17–37.
Ferrera, Maurizio (2010): ‘The South European Countries’. In Francis G. Castles,
pp. 616–29.
Flora, Peter and Alber, Jens (1981): ‘Modernization, Democratization, and the
Development of Welfare States in Western Europe’. In Peter Flora and Arnold
J. Heidenheimer (eds) The Development of Welfare States in Europe and America.
New Brunswick, NJ: Transaction Books, pp. 37–80.
Fulton, Lionel (2011): Worker Representation in Europe. Brussels: ETUI. Website:
http://www.worker-participation.eu.
Gallego, Raquel, Coma, Ricard and Subirats, Joan (2005): ‘From State Welfare
to Regional Welfare?’ In Nicola McEwen and Luis Moreno (eds) The Territorial
Politics of Welfare. New York: Routledge, pp. 103–26.
Giaimo, Susan (2001): ‘Who Pays for Health Care Reform?’ In Paul Pierson
(ed.) The New Politics of the Welfare State. Oxford: Oxford University Press,
pp. 334–67.
Glassmann, Ulrich (2009): ‘Rule-Breaking and Freedom of Rules in National
Production Models: How German Capitalism Departs from the “Rhenish
Equilibrium”’. In Colin Crouch and Helmut Voelzkow (eds) Innovation in Local
Economies. Germany in Comparative Context. Oxford: Oxford University Press,
pp. 22–43.
Glatzer, Miguel (2005): ‘Revisiting “Embedded Liberalism”. Globalization and
the Welfare State in Spain and Portugal’. In Miguel Glatzer and Dietrich
Rueschemeyer (eds) Globalization and the Future of the Welfare State. Pittsburgh,
PA: University of Pittsburgh Press, pp. 106–29.
Goldsmith, Michael and Kingfisher, Catherine (2003): ‘The Welfare State in
Aotearoa/New Zealand’. In Christian Aspalter (ed.) Welfare Capitalism around
the World. Taichung: Casa Verde Publishing, pp. 329–42.
Goodin, Robert E., Headey, Bruce, Muffels, Ruud and Dirven, Henk-Jan (1999): The
Real Worlds of Welfare Capitalism. Cambridge: Cambridge University Press.
Gough, Ian, Jonathan Bradshaw, John Ditch, and Tony, Eardley (1997): ‘Social
Assistance in OECD Countries.’ Journal of European Social Policy 7(1), 17–43.
Grabher, Gernot (1993): The Embedded Firm. On the Socioeconomics of Industrial
Networks. London: Routledge.
Graham, Andrew (1997): ‘The UK 1975–95: Myths and Realities of Conservative
Capitalism’. In Colin Crouch and Wolfgang Streeck (eds) Political Economy of
Modern Capitalism: Mapping Convergence and Diversity. London: Sage, pp. 117–32.
References 191
Granovetter, Mark (1985): ‘Economic Action and Social Structure. The Problem
of Embeddedness’. American Journal of Sociology, 91(3), 481–510.
Grønbjerg, Kirsten A., Street, David and Suttles, Gerald D. (1978): Poverty and
Social Change. Chicago: University of Chicago Press.
Guillén, Ana M. (1992): ‘Social Policy in Spain: From Dictatorship to Democracy
(1939–1982)’. In Zsuzsa Ferge and Jon Kolberg (eds) Social Policy in a Changing
Europe. Frankfurt/New York: Westview Press, pp. 119–42.
Guillén, Ana M. (2010): ‘Defrosting the Spanish Welfare State: The Weight of
Conservative Components’. In Bruno Palier (ed.) A Long Goodbye to Bismarck?
The Politics of Welfare Reforms in Continental Europe. Amsterdam: Amsterdam
Gundelach, Peter (1994): ‘National Value Differences: Modernization or
Institutionalization?’ International Journal of Comparative Sociology, 35, 37–58.
Hacker, Jacob S. and Pierson, Paul (2010): ‘Winner-Take-All Politics: Public Policy,
Political Organization, and the Precipitous Rise of Top Incomes in the United
States’. Politics & Society, 38, 152–204.
Hall, Peter (1986): Governing the Economy: The Politics of State Intervention in Britain
and France. Cambridge: Polity Press.
Hall, Peter (1993): ‘Policy Paradigms, Social Learning, and the State. The Case of
Economic Policymaking in Britain’. Comparative Politics, 25, 275–96.
Hall, Peter (1997): ‘The Political Economy of Adjustment in Germany’. In Bob
Hancké, Ulrich Jürgens, Frieder Naschold and David Soskice (eds) Ökonomische
Leistungsfähigkeit und institutionelle Innovation. Das deutsche Produktions- und
Politikregime im globalen Wettbewerb. Berlin: Edition Sigma, pp. 293–319.
Hall, Peter (1999): ‘The Political Economy of Europe in an Era of Interdependence’.
In Herbert Kitschelt, Peter Lange, Gary Marks and John Stephens (eds)
Continuity and Change in Contemporary Capitalism. Cambridge: Cambridge
Hall, Peter (2001): ‘Organized Market Economies and Unemployment in
Europe: Is It Finally Time to Accept Liberal Orthodoxy?’ In Nancy Bermeo
(ed.) Unemployment in the New Europe. New York: Cambridge University Press,
pp. 52–86.
Hall, Peter (2002): ‘Globalization and Economic Adjustment in Germany’. In Karl
Müller, Ellen Hollingsworth and Roger Hollingsworth (eds) Advancing Socio-
Economics: An Institutionalist Perspective. Lanham, MD: Rowman & Littlefield
Publishers, pp. 253–305.
Hall, Peter (2007): ‘The Evolution of Varieties of Capitalism in Europe’. In Bob
Conflict, Contradictions and Complementarities in the European Economy. Oxford:
Oxford University Press, pp. 39–85.
Hall, Peter (2009): ‘Re-Forming Capitalism’. Archives Européennes de Sociologie, 50,
488–94.
Hall, Peter and Gingerich, Daniel W. (2004): Varieties of Capitalism and Institutional
Complementarities in the Macroeconomy: An Empirical Analysis. MPIfG Discussion
Paper 04/5. Cologne, Max Planck Institute for the Study of Societies. Available at:
http://www.mpifg.de/pu/mpifg_dp/dp04-5.pdf – downloaded 14 March 2006.
Hall, Peter and Gingerich, Daniel W. (2009): ‘Varieties of Capitalism and
Institutional Complementarities in the Political Economy: An Empirical
Analysis’. British Journal of Political Science, 39, 449–82.
192 References
Hall, Peter and Soskice, David (2001a): ‘An Introduction to Varieties of

Capitalism’. In Peter Hall and David Soskice (eds) Varieties of Capitalism: The
Institutional Foundations of Comparative Advantage. Oxford: Oxford University
Press, pp. 1–71.
Hall, Peter and Soskice, David (2001b): Varieties of Capitalism: The Institutional
Foundations of Comparative Advantage. Oxford: Oxford University Press.
Hall, Peter A. and Thelen, Kathleen (2009): ‘Institutional Change in Varieties of
Capitalism’. Socio-Economic Review, 7, pp. 361–91.
Hamilton, Gary and Woolsey, Nicole (1988): ‘Market, Culture and Authority:
A Comparative Analysis of Management and Organizations in the Far East’.
American Journal of Sociology, 94, 52–94.
Hampson, Jack (1997): ‘Social Protection and Social Insurance in Portugal’.
In Jochen Clasen (ed.) Social Insurance in Europe. Bristol: Policy Press,
pp. 151–76.
Hancké, Bob (2009): Debating Varieties of Capitalism: A Reader. Oxford and
New York: Oxford University Press.
Hancké, Bob, Rhodes, Martin and Thatcher, Mark (eds) (2007a): Beyond Varieties
of Capitalism. Conflict, Contradictions and Complementarities in the European
Economy. New York: Oxford University Press.
Hancké, Bob, Rhodes, Martin and Thatcher, Mark (2007b): ‘Introduction: Beyond
Varieties of Capitalism’. In Bob Hancké, Martin Rhodes and Mark Thatcher (eds)
Beyond Varieties of Capitalism. Conflict, Contradictions and Complementarities in
the European Economy. Oxford: Oxford University Press, pp. 3–38.
Harbridge, Raymond and Walsh, Pat (2002): ‘Labour Market Reform in
New Zealand’. In Hedva Sarfati and Giuliano Bonoli (eds) Labor Market and
Social Protection Reforms in International Perspective. Parallel or Converging Tracks?
Aldershot: Ashgate, pp. 198–220.
Harrison, Lawrence and Huntington, Samuel (2000): Culture Matters: How Values
Shape Human Progress. New York: Basic Books.
Hassel, Anke and Schiller, Christof (2010): Der Fall Hartz IV. Wie es zur Agenda
2010 kam und wie es weitergeht. Frankfurt: Campus.
Hay, Colin (2005): ‘Two Can Play at That Game … or Can They? Varieties
of Capitalism, Varieties of Institutionalism’. In David Coates (ed.) Varieties
of Capitalism, Varieties of Approaches. New York: Palgrave Macmillan,
pp. 106–21.
Hemerijck, Anton and Marx, Ive (2010): ‘Continental Welfare at a Crossroads:
The Choice between Activation and Minimum Income Protection in Belgium
and the Netherlands’. In Bruno Palier (ed.) A Long Goodbye to Bismarck? The
Politics of Welfare Reforms in Continental Europe. Amsterdam: Amsterdam
Hemerijck, Anton, Meer, Marc Van der and Visser, Jelle (2000): ‘Innovation
through Co-ordination – Two Decades of Social Pacts in the Netherlands’. In
Giuseppe Fajertag and Philippe Pochet (eds) Social Pacts in Europe. Brussels:
European Trade Union Institute, pp. 257–78.
Hemerijck, Anton, Unger, Brigitte and Visser, Jelle (2000): ‘How Small Countries
Negotiate Change. Twenty-Five Years of Policy Adjustment in Austria, the
Netherlands, and Belgium’. In Fritz Wilhelm Scharpf and Vivien A. Schmidt
(eds) Welfare and Work in the Open Economy. Volume II. Diverse Responses to
Common Challenges. Oxford: Oxford University Press, pp. 175–264.
References 193
Herrigel, Gary (1993): ‘Power and the Redefinition of Industrial Districts.

The Case of Baden-Württemberg’. In Gernot Grabher (ed.) The Embedded
Firm. On the Socioeconomics of Industrial Networks. London: Routledge,
pp. 227–51.
Herrigel, Gary (1996): Industrial Constructions: The Sources of German Industrial
Power. Cambridge: Cambridge University Press.
Hicks, Alexander and Kenworthy, Lane (1998): ‘Cooperation and Political
Economic Performance in Affluent Democratic Capitalism’. American Journal
of Sociology, 103(6), 631–72.
Hicks, Alexander and Kenworthy, Lane (2003): ‘Varieties of Welfare Capitalism’.
Socio-Economic Review 2003, 1, 27–61. Available at: http://ser.oxfordjournals.
org/cgi/reprint/1/1/27 – downloaded 2 April 2006.
Hicks, Alexander and Swank, Duane (1992): ‘Politics, Institutions and Social
Welfare Spending in the Industrialized Democracies 1960–82’. American
Political Science Review, 86, 658–74.
Higgins, Jane (1999): ‘From Welfare to Workfare’. In Jonathan Boston, Paul
Dalziel and Susan St. John (eds) Redesigning the Welfare State in New Zealand.
Problems, Policies, Prospects. Oxford: Oxford University Press, pp. 260–77.
Hinze, Otto (1970a [1930]): Staat und Verfassung. Gesammelte Abhandlungen zur
Allgemeinen Verfassungsgeschichte. Göttingen: Vandenhoeck & Rupprecht.
Hinze, Otto (1970b [1930]): ‘Typologie der ständischen Verfassungen des
Abendlandes’. In Otto Hintze (ed.) Staat und Verfassung. Gesammelte
Abhandlungen zur Allgemeinen Verfassungsgeschichte. Göttingen: Vandenhoeck &
Rupprecht, pp. 120–39.
Hirschman, Albert O. (1977): The Passions and the Interests: Political Arguments For
Capitalism Before its Triumph. Princeton, NJ: Princeton University Press.
Hollingsworth, Roger (1991): ‘The Logic of Coordinating American Manufacturing
Sectors’. In John Campbell, Roger Hollingsworth and Leon Lindberg (eds)
Governance of the American Economy. Cambridge: Cambridge University Press,
pp. 35–73.
Hollingsworth, Roger (1997a): ‘Continuities and Changes in Social Systems of
Production: The Cases of Japan, Germany and the United States’. In Robert Boyer
and J. Rogers Hollingsworth (eds) Contemporary Capitalism. The Embeddedness of
Institutions. Cambridge: Cambridge University Press, pp. 265–310.
Hollingsworth, Roger (1997b): ‘The Institutional Embeddedness of American
Capitalism’. In Colin Crouch and Wolfgang Streeck (eds) Political Economy
of Modern Capitalism: Mapping Convergence and Diversity. London: Sage, pp.
133–47.
Hollingsworth, Roger (1998): ‘New Perspectives on the Spatial Dimensions of
Economic Coordination: Tensions Between Globalization and Social Systems
of Production’. Review of International Political Economy, 5(3), 482–507.
Höpner, Martin (2001): Corporate Governance in Transition: Ten Empirical Findings
on Shareholder Value and Industrial Relations in Germany. MPIfG Discussion
Paper 01/5. Cologne: Max Planck Institute for the Study of Societies. Available
at: http://www.mpi-fg-koeln.mpg.de/pu/mpifg_dp/dp01-5.pdf – downloaded
26 April 2006.
Höpner, Martin (2005a): ‘What Connects Industrial Relations and Corporate
Governance? Explaining Institutional Complementarity’. Socio-Economic
Review, 3(2), 331–58.
194 References
Höpner, Martin (2005b) ‘Epilogue to “Explaining Institutional Complementarity”:

What Have We Learnt? Complementarity, Coherence and Institutional
Change’. Socio-Economic Review, 3(2), 383–8.
Höpner, Martin (2007): Coordination and Organization. The Two Dimensions
of Nonliberal Capitalism, MPIfG Discussion Paper 07/12. Cologne: MPIfG.
Available at: http://www.mpi-fg-koeln.mpg.de/pu/mpifg_dp/dp07-12.pdf –
Höpner, Martin and Jackson, Gregory (2001): An Emerging Market for Corporate
Control? The Mannesmann Takeover and German Corporate Governance, MPIfG
Discussion Paper 01/4. Cologne: Max Planck Institute for the Study of
Societies. Available at: http://www.mpi-fg-koeln.mpg.de/pu/mpifg_dp/dp01-4.
pdf – downloaded 26 April 2006.
Huber, Evelyne and Stephens, John (1996): Political Power and Gender in the
Making of the Social Democratic Service State. Paper delivered at the 1996 Annual
Meeting of the American Political Science Association, San Francisco, 28
August–1 September.
Huber, Evelyne and Stephens, John (2001a): ‘Welfare State and Production
Regimes in the Era of Retrenchment’. In Paul Pierson (ed.) The New Politics of
the Welfare State. Oxford: Oxford University Press, pp. 107–45.
Huber, Evelyne and Stephens, John (2001b): Development and Crisis of the Welfare
State: Parties and Politics in Global Markets. Chicago: University of Chicago
Press.
Huemer, Gerhard, Mesch, Michael and Traxler, Franz (1999): The Role of Employer
Associations and Labour Unions in the EMU: Institutional Requirements for European
Economic Policies. Aldershot: Ashgate.
Huntington, Samuel (1993): ‘The Clash of Civilizations?’ Foreign Affairs, 72(3),
22–49.
Immergut, Ellen (1998): ‘The Theoretical Core of the New Institutionalism’.
Politics and Society, 26(1), 5–34.
Iversen, Thomas and Soskice, David (2000): An Asset Theory of Social Preferences.
Paper presented at the annual meeting of the American Political Science
Association. Wardman Park. Available at: http://www.nd.edu/~kellogg/pdfs/
Iverson.pdf – downloaded 31 March 2006.
Iversen, Torben, Pontusson, Jonas and Soskice, David (2000): Unions, Employers
and Central Banks. Cambridge: Cambridge University Press.
Iversen, Torben and Rosenbluth, Frances McCall (2010): Women, Work, And
Politics: The Political Economy of Gender Inequality, The Yale ISPS Series. New
Haven, CT: Yale University Press.
Iversen, Torben and Soskice, David (2006): ‘Electoral Institutions and the Politics
of Coalitions: Why Some Democracies Redistribute More Than Others’.
American Political Science Review, 100, 165–81.
Jackson, Gregory (2005): ‘Employee Representation in the Board Compared:
A Fuzzy Sets Analysis of Corporate Governance, Unionism and Political
Institutions’. Industrielle Beziehungen, 12, 1–28.
Jackson, Gregory and Deeg, Richard (2006): How Many Varieties of Capitalism?
Comparing the Comparative Institutional Analyses of Capitalist Diversity, MPIfG
Discussion Paper 06/2. Cologne: Max Planck Institute for the Study of Societies.
Jackson, Gregory and Vitols, Sigurt (2001): ‘Pension Regimes and Financial
Systems: Between Social Security, Market Liquidity and Corporate Governance’.
References 195
In Bernhard Ebbinghaus and Philip Manow (eds) Varieties of Welfare Capitalism.

Jessoula, Matteo and Alti, Tiziana (2010): ‘Italy: An Uncompleted Departure
from Bismarck’. In Bruno Palier (ed.) A Long Goodbye to Bismarck? The Politics
of Welfare Reforms in Continental Europe. Amsterdam: Amsterdam University
Press, pp. 157–81.
Jones, Catherine (1993): New Perspectives on the Welfare State in Europe. London:
Routledge.
Kahl, Sigrun (2005): ‘The Religious Roots of Modern Poverty Policy: Catholic,
Lutheran, and Reformed Protestant Traditions Compared’. Archives Européennes
de Sociologie, 46, 91–126.
Kahl, Sigrun (2009): ‘Religious Doctrines and Poor Relief: A Different Causal
Pathway’. In Kees van Kersbergen and Philip Manow (eds) Religion, Class Coalitions
and Welfare States. Cambridge: Cambridge University Press, pp. 267–95.
Karamessini, Maria (2008): ‘Continuity and change in the southern European
social model.’ International Labour Review 147(1), 43–70.
Katzenstein, Peter (1984): Corporatism and Change: Austria, Switzerland, and the
Politics of Industry. Ithaca, NY: Cornell University Press.
Katzenstein, Peter (1985): Small States in World Markets. Ithaca, NY: Cornell
University Press.
Katznelson, Ira and Zolberg, Aristide (1986): Working-Class Formation: Nineteenth-
Century Patterns in Western Europe and the United States. Princeton, NJ: Princeton
University Press.
Kauppinen, Timo (2000): ‘Social Pacts in Finland’. In Giuseppe Fajertag and
Kautto, Mikko (2010): ‘The Nordic Countries’. In Francis G. Castles, Stephan
Leibfried, Jane Lewis, Herbert Obinger and Christopher Pierson (eds) The
Oxford Handbook of the Welfare State. Oxford: Oxford University Press,
pp. 586–600.
Kelly, John and Hamann, Kerstin (2008): ‘Varieties of Capitalism and Industrial
Relations’. In Paul Blyton, Nicolas Bacon, Jack Fiorito and Edmund Heery (eds)
The SAGE Handbook of Industrial Relations. London: Routledge, pp. 129–48.
Kenworthy, Lane (1997): ‘Globalization and Economic Convergence’. Competition
and Change, 2, 1–64.
Kenworthy, Lane (2006): ‘Institutional Coherence and Macroeconomic
Performance’. Socio-Economic Review, 4, 69–91.
Kersbergen, Kees van (2009): ‘Religion and the Welfare State in the Netherlands’.
In Kees van Kersbergen and Philip Manow (eds) Religion, Class Coalitions, and
Welfare States, 1st edn. Cambridge: Cambridge University Press, pp. 119–45.
Kersbergen, Kees van and Manow, Philip (eds) (2009): Religion, Class Coalitions,
and Welfare States. Cambridge: Cambridge University Press.
Kersbergen, Kees van and Manow, Philip (2010): ‘Religion’. In Francis G. Castles,
pp. 265–77.
Kersting, Wolfgang (2003): ‘Gerechtigkeit: Die Selbstverewigung des egalitarist-
ischen Sozialstaat’. In Stephan Lessenich (ed.) Wohlfahrtsstaatliche Grundbegriffe.
Historische und aktuelle Diskurse. Frankfurt: Campus, pp. 105–35.
196 References
Kesting, Stefan and Nielsen, Klaus (2008): ‘Varieties of Capitalism: Theoretical

Critique and Empirical Observations’. In Wolfram Elsner and Gerhard Hanappi
(eds) Varieties of Capitalism and New Institutional Deals. Regulation, Welfare and
the New Economy. Cheltenham: Edward Elgar, pp. 23–51.
Kiely, Gabriel (1999): ‘Introduction: From Colonial Paternalism to National
Partnership: An Overview of Irish Social Policy’. In Gabriel Kiely, Anne
O’Donnell, Patricia Kennedy and Suzanne Quin (eds) Irish Social Policy in
Context. Dublin: Dublin University Press, pp. 1–10.
King, Desmond and Wood, Stewart (1999): ‘The Political Economy of
Neoliberalism: Britain and the United States in the 1980s’. In Herbet
Kitschelt, Peter Lange, Gary Marks and John Stephens (eds) Continuity and
Change in Contemporary Capitalism. Cambridge: Cambridge University Press,
pp. 371–97.
Kitschelt, Herbert, Lange, Peter, Marks, Gary and Stephens, John (1999a):
Continuity and Change in Contemporary Capitalism. Cambridge: Cambridge
University Press.
Kitschelt, Herbert, Lange, Peter, Marks, Gary and Stephens, John (1999b):
‘Conclusions’. In Herbet Kitschelt, Peter Lange, Gary Marks and John Stephens
(eds) Continuity and Change in Contemporary Capitalism. Cambridge: Cambridge
Kitschelt, Herbert and Streeck, Wolfgang (2004): Germany: Beyond the Stable State.
London and Portland: Frank Cass.
Kitson, Michael (2005): ‘Measuring Capitalism: Output, Growth and Economic
Policy’. In David Coates (ed.) Varieties of Capitalism, Varieties of Approaches.
New York: Palgrave Macmillan, pp. 29–46.
Knudsen, Tim and Rothstein, Bo (1994): ‘State Building in Scandinavia’.
Comparative Politics, 26, 203–20.
Konzelmann, Sue, Fovargue-Davies, Marc and Schnyder, Gerhard (2010):
Varieties of Liberalism: Anglo-Saxon Capitalism in Crisis? Working Paper No. 403.
Cambridge: Centre for Business Research, University of Cambridge.
Korpi, Walter (1985): ‘Power Resources Approach vs. Action and Conflict: On
Causal and Intentional Explanations in the Study of Power’. Sociological Theory,
3(2), 31–45.
Korpi, Walter (2000): Faces of Inequality: Gender, Class and Patterns of Inequalities
in Different Types of Welfare States. Luxembourg Income Study Working
Paper No. 224. Syracuse: Maxwell School of Citizenship and Public Affairs.
Available at: http://scholar.google.com/url?sa=U&q=http://www.lisproject.org/
publications/liswps/224.pdf – downloaded 27 May 2006.
Korpi, Walter (2006): ‘Power Resources and Employer-Centred Approaches
in Explanations of Welfare States and Varieties of Capitalism: Protagonists,
Consenters, and Antagonists’. World Politics, 58, 167–206.
Korpi, Walter and Palme, Joakim (1998): ‘The Strategy of Equality and the
Paradox of Redistribution’. American Sociological Review, 63(5), 661–87.
Krasner, Stephen (1984): ‘Approaches to the State’. Comparative Politics, 16(2),
223–46.
Kreps, David (1990): A Course in Microeconomic Theory. Princeton, NJ: Princeton
University Press.
Kudrle, Robert T. and Marmor, Theodore R. (1981): ‘The Development of Welfare
States in North America’. In Peter Flora and Arnold J. Heidenheimer (eds)
References 197
The Development of Welfare States in Europe and America. New Brunswick, NJ:
Transaction Books, pp. 81–121.
Kuhnle, Stein (1980): ‘National Equality and Local Decision-Making: Values in
Conflict in the Development of the Norwegian Welfare State’. Acta Sociologica,
23, 97–111.
Kuhnle, Stein (1981): ‘The Growth of Social Insurance Programs in Scandinavia:
Outside Influences and Internal Forces’. In Peter Flora and Arnold J.
Heidenheimer (eds) The Development of Welfare States in Europe and America.
New Brunswick, NJ: Transaction Books, pp. 125–50.
Kuhnle, Stein and Sander, Anne (2010): ‘The Emergence of the Western Welfare
State’. In Francis G. Castles, Stephan Leibfried, Jane Lewis, Herbert Obinger
and Christopher Pierson (eds) The Oxford Handbook of the Welfare State. Oxford:
La Porta, Rafael, López-de-Silanes, Florencio and Shleifer, Andrei (2008): ‘The
Economic Consequences of Legal Origins’. Journal of Economic Literature, 46,
285–332.
La Porta, Rafael, López-de-Silanes, Florencio, Shleifer, Andrei and Vishny, Robert W.
(1997): ‘Legal Determinants of External Finance’. Journal of Finance, 52, 1131–50.
La Porta, Rafael, López-de-Silanes, Florencio, Shleifer, Andrei and Vishny, Robert
W. (1998): ‘Law and Finance’. Journal of Political Economy, 106, 1113–55.
Lane, David Stuart and Myant, Martin (2007): Varieties of Capitalism in Post-
Communist Countries. Basingstoke: Palgrave Macmillan.
Lauder, Hugh, Brown, Phillip and Ashton, David (2008): ‘Globalisation, Skill
Formation and the Varieties of Capitalism Approach’. New Political Economy,
13, 19–35.
Lavalette, Michael and Penketh, Laura (2003): ‘The Welfare State in the United
Kingdom’. In Christian Aspalter (ed.) Welfare Capitalism around the World.
Taichung: Casa Verde Publishing, pp. 61–86.
Lehmbruch, Gerhard (1984): ‘Concertation and the Structure of Corporatist
Networks’. In John Goldthorpe (ed.) Order and Conflict in Contemporary
Capitalism. Oxford: Clarendon Press, pp. 61–80.
Lehmbruch, Gerhard (2001): ‘The Institutional Embedding of Market Economies.
The German “Model” and Its Impact on Japan’. In Wolfgang Streeck and Kozo
Yamamura (eds) The Origins of Nonliberal Capitalism: Germany and Japan in
Comparison. Ithaca, NY: Cornell University Press, pp. 39–93.
Lehmbruch, Gerhard and Schmitter, Philippe (1979): Trends toward Corporatist
Intermediation. Beverly Hills, CA: Sage.
Lehmbruch, Gerhard and Schmitter, Philippe (1982): Patterns of Corporatist Policy-
Making. Beverly Hills, CA: Sage.
Leibfried, Stephan (1992) ‘Towards a European Welfare State? On Integrating
Poverty Regimes into the European Community’. In Zsuzsa Ferge and Jon
Kolberg (eds) Social Policy in a Changing Europe. Frankfurt: Campus-Westview,
pp. 245–79.
Leisering, Lutz (2001): ‘Germany: Reform from Within’. In Pete Alcock and Gary
Craig (eds) International Social Policy: Welfare Regimes in the Developed World.
New York: Palgrave Macmillan, pp. 161–82.
Lenz, Gabriel and Lipset, Seymour (2000): ‘Corruption, Culture and Markets’. In
Lawrence Harrison and Samuel Huntington (eds) Culture Matters: How Values
Shape Human Progress. New York: Basic Books, pp. 112–24.
198 References
Lessenich, Stephan (1994): ‘Three Worlds of Welfare Capitalism – oder vier?

Strukturwandel arbeits- und sozialpolitischer Regulierungsmuster in Spanien’.
Politische Vierteljahresschrift, 35(2), 224–44.
Lessenich, Stephan (2003): Wohlfahrtsstaatliche Grundbegriffe. Historische und
aktuelle Diskurse. Frankfurt: Campus.
Levine, Daniel (1978): ‘Conservatism and Tradition in Danish Social Welfare
Legislation, 1890–1933: A Comparative View’. Comparative Studies in Society
and History, 20, 54–69.
Levine, Ross (2000): Bank-Based or Market-Based Financial Systems: Which is Better?
Minneapolis: Finance Department Carlson School of Management, University
of Minnesota. Available at: http://scholar.google.com/url?sa=U&q=http://
www.bus.umich.edu/KresgeLibrary/Collections/Workingpapers/wdi/wp442.
pdf – downloaded 2 June 2006.
Levy, Jonah D. (2000): ‘France. Directing Adjustment?’ In Fritz Wilhelm Scharpf and
Vivien A. Schmidt (eds) Welfare and Work in the Open Economy. Volume II. Diverse
Responses to Common Challenges. Oxford: Oxford University Press, pp. 308–51.
Lightman, Ernie and Riches, Graham (2001): ‘Canada: One Step Forwards, Two
Steps Back?’ In Pete Alcock and Gary Craig (eds) International Social Policy. Welfare
Lijphart, Arend (1968): The Politics of Accommodation. Pluralism and Democracy in
the Netherlands. Berkeley: University of California Press.
Lima, Maria da Paz Campos and Naumann, Reinhard (2000): ‘Social Pacts in
Portugal: From Comprehensive Policy Programmes to the Negotiation of
Concrete Industrial Relations Reforms?’ In Giuseppe Fajertag and Philippe
Pochet (eds) Social Pacts in Europe. Brussels: European Trade Union Institute,
pp. 321–42.
Lind, Jens (2000): ‘Recent Issues on the Social Pact in Denmark’. In Giuseppe
Fajertag and Philippe Pochet (eds) Social Pacts in Europe. Brussels: European
Trade Union Institute, pp. 135–59.
Linz, Juan J. (1967): ‘The Party System of Spain: Past and Future’. In Seymour
Martin Lipset and Stein Rokkan (eds) Party Systems and Voter Alignments: Cross-
National Perspectives. New York: Free Press, pp. 197–282.
Lipietz, Alain (1987): Mirages and Miracles: The Crises of Global Fordism. London:
Verso.
Lipset, Seymour Martin and Rokkan, Stein (1967): ‘Cleavage Structures, Party
Systems, and Voter Alignments: An Introduction’. In Seymour Martin Lipset
and Stein Rokkan (eds) Party Systems and Voter Alignments: Cross-National
Perspectives, 2nd edn. New York: Free Press, pp. 1–64.
Lloyd, Caroline and Payne, Jonathan (2002): ‘On the “Political Economy of
Skill”: Assessing the Possibilities for a Viable High Skills Project in the United
Kingdom’. New Political Economy, 7(3), 367–95.
Longstreth, Frank, Steinmo, Sven and Thelen, Kathleen (1992): Structuring Politics.
Historical Institutionalism in Comparative Analysis. Cambridge: Cambridge
University Press.
Lucio, Miguel Martínez (2000): ‘Spain: Regulating Employment and Social
Fragmentation’. In Anthony Ferner and Richard Hyman (eds) Changing
Industrial Relations in Europe, 2nd edn. Oxford: Blackwell, pp. 426–58.
Lütz, Susanne (2003): Governance in der politischen Ökonomie. MPIfG Working
Paper 03/05. Cologne: Max Planck Institute for the Study of Societies. Available
References 199
at: http://www.mpi-fg-koeln.mpg.de/pu/mpifg_dp/dp03-5.pdf – downloaded

15 May 2006.
Lynch, Julia (2009): ‘Italy: A Christian Democratic or Clientelist Welfare State?’
In Kees van Kersbergen and Philip Manow (eds) Religion, Class Coalitions, and
Welfare States, 1st edn. Cambridge: Cambridge University Press, pp. 91–118.
Maclean, Mairi and Harvey, Charles (2008): ‘The Continuing Diversity of
Corporate Governance Regimes: France and Britain Compared’. In Roger
Strange and Gregory Jackson (eds) Corporate Governance and International
Business. Strategy, Performance and Institutional Change. New York: Palgrave
Macmillan, pp. 208–25.
Madsen, Per Kongshøj (2002): ‘The Danish Model of Flexicurity: A Paradise –
With Some Snakes’. In Hedva Sarfati and Giuliano Bonoli (eds) Labor Market
and Social Protection Reform in International Perspective. Parallel or Converging
Tracks? Aldershot: Ashgate, pp. 243–65.
Maino, Franca (2003): ‘The Welfare State in Italy’. In Christian Aspalter (ed.)
Welfare Capitalism around the World. Taichung: Casa Verde Publishing,
pp. 199–217.
Maître, Bertrand, Nolan, Brian and Whelan, Christopher (2005): ‘Welfare Regimes
and Household Income Packaging in the European Union’. Journal of European
Social Policy, 15(2), 157–71.
Mallin, Chris A. (2010): Corporate Governance. Oxford, New York: Oxford
University Press.
Mancur Olson, (1982): ‘The Rise and Decline of Nations: Economic Growth’.
Stagflation, and Social Rigidities, Yale University Press, 1982.
Mann, Michael (1993): The Sources of Social Power: Volume 2, The Rise of Classes
and Nation States 1760–1914. Cambridge: Cambridge University Press.
Manning, Nick and Shaw, Ian (1998): ‘The Transferability of Welfare Models:
A Comparison of the Scandinavian and State Socialist Models in Relation to
Finland and Estonia’. Social Policy & Administration, 32, 572–90.
Manow, Philip (1997): Social Insurance and the German Political Economy. MPIfG
Discussion Paper 97/2. Cologne.
Manow, Philip (2001a): ‘Welfare State Building and Coordinated Capitalism
in Japan and Germany’. In Wolfgang Streeck and Kozo Yamamura (eds) The
Origins of Nonliberal Capitalism: Germany and Japan in Comparison. Ithaca, NY:
Cornell University Press, pp. 94–120.
Manow, Philip (2001b): ‘Business Coordination, Wage Bargaining and the
Welfare State. Germany and Japan in Comparative Historical Perspective’. In
Bernhard Ebbinghaus and Philip Manow (eds) Varieties of Welfare Capitalism.
Manow, Philip (2001c): ‘Comparative Institutional Advantages of Welfare
State Regimes and New Coalitions in Welfare State Reforms’. In Paul Pierson
pp. 146–64.
Manow, Philip (2009): ‘Electoral Rules, Class Coalitions and Welfare State
Regimes, or How to Explain Esping-Andersen with Stein Rokkan’. Socio-
Economic Review, 7, 101–21.
Manow, Philip and Palier, Bruno (2009): ‘A Conservative Welfare State Regime
without Christian Democracy? The French État-providence, 1880–1960’.
In Kees van Kersbergen and Philip Manow (eds) Religion, Class Coalitions,
200 References
and Welfare States, 1st edn. Cambridge: Cambridge University Press,

pp. 146–75.
Marc, Céline and Zajdela, Hélène (2005): Emploi des Mères et Politique Familiale:
Doit-on S’Inspirer du Modèle Suédois? Working Paper at the CEE, no. 45.
Mares, Isabella (2001a): ‘Firms and the Welfare State: When, Why, and How
Does Social Policy Matter to Employers?’ In Peter Hall and David Soskice (eds)
‘Varieties of Capitalism’: The Institutional Foundations of Comparative Advantage.
Mares, Isabella (2001b): ‘Strategic Bargaining and Social Policy Development:
Unemployment Insurance in France and Germany’. In Bernhard Ebbinghaus
and Philip Manow (eds) Varieties of Welfare Capitalism. London: Routledge,
pp. 52–75.
Martin, Cathie (2005): ‘Beyond Bone Structure: Historical Institutionalism and
the Style of Economic Growth’. In David Coates (ed.) Varieties of Capitalism,
Varieties of Approaches. New York: Palgrave Macmillan, pp. 47–62.
Martin, Cathie and Swank, Duane (2001): ‘Employers and the Welfare State: The
Political Economic Organization of Firms and Social Policy in Contemporary
Capitalist Democracies’. Comparative Political Studies, 34(8), 889–923.
Martin, Cathie Jo and Swank, Duane (2008): ‘The Political Origins of Coordinated
Capitalism: Business Organizations, Party Systems, and State Structure in the
Age of Innocence’. American Political Science Review, 102, 181–98.
Martin, Cathie Jo and Swank, Duane (2012): The Political Construction of Business
Interests. Cambridge: Cambridge University Press.
Martin, Cathie Jo and Thelen, Kathleen (2007): ‘The State and Coordinated
Capitalism: Contributions of the Public Sector to Social Solidarity in
Postindustrial Societies’. World Politics, 60, 1–36.
McAllister, Ian and Vowles, Jack (1994): ‘The Rise of New Politics and Market
Liberalism in Australia and New Zealand’. British Journal of Political Science, 24,
381–402.
McBride, Stephen (2005): ‘Exporting and Internalizing Globalization: Canada and
the Global System of Power’. In Susanne Soederberg, Georg Menz and Philip
Cerny (eds) Internalizing Globalization: The Rise of Neoliberalism and the Decline of
National Varieties of Capitalism. New York: Palgrave Macmillan, pp. 106–22.
McLaughlin, Eugene (2001): ‘Ireland: From Catholic Corporatism to Social
Partnership’. In Allan Cochrane, John Clarke and Sharon Gewirtz (eds)
Comparing Welfare States. London: Sage, pp. 223–60.
McVeigh, Paul (2005): ‘Embedding Neoliberalism in Spain: from Franquismo to
Neoliberalism’. In Susanne Soederberg, Georg Menz and Philip Cerny (eds)
Internalizing Globalization: The Rise of Neoliberalism and the Decline of National
Varieties of Capitalism. New York: Palgrave Macmillan, pp. 90–105.
Meer, Marc van der (1996): ‘Aspiring Corporatism? Industrial Relations in Spain’.
In Joris Van Ruysseveldt and Jelle Visser (eds) Industrial Relations in Europe.
Traditions and Transitions. London: Sage, pp. 310–36.
Menz, Georg (2005a): ‘Auf Wiedersehen, Rhineland Model: Embedding
Neoliberalism in Germany’. In Susanne Soederberg, Georg Menz and Philip
Cerny (eds) Internalizing Globalization: The Rise of Neoliberalism and the Decline
of National Varieties of Capitalism. New York: Palgrave Macmillan, pp. 33–48.
Menz, Georg (2005b): ‘Making Thatcher Look Timid: The Rise and the Fall of the
New Zealand Model’. In Susanne Soederberg, Georg Menz and Philip Cerny
References 201
(eds) Internalizing Globalization: The Rise of Neoliberalism and the Decline of

National Varieties of Capitalism. New York: Palgrave Macmillan, pp. 47–62.
Michelotti, Marco and Nyland, Chris (2008): ‘Varieties of Capitalism and
Diversity in Labour Standards Regulation: The Case of Italy’. European Journal
of Industrial Relations, 14, 177–95.
Miller, David (1999): Principles of Social Justice. Cambridge, MA: Harvard
University Press.
Molina, Óscar and Rhodes, Martin (2007a): ‘Industrial Relations and the Welfare
State in Italy: Assessing the Potential of Negotiated Change’. West European
Politics 30, 803–29.
Molina, Óscar and Rhodes, Martin (2007b): ‘The Political Economy of Adjustment
in Mixed Market Economies. A Study of Spain and Italy’. In Bob Hancké,
Martin Rhodes and Mark Thatcher (eds) Beyond Varieties of Capitalism, 1st edn.
Moreno, Luis (2001): ‘Spain, a Via Media of Welfare Development’. In Peter
Taylor-Gooby (ed.) Welfare States Under Pressure. London: Sage, pp. 100–23.
Moschella, M. (2011): ‘Different Varieties of Capitalism? British and Italian
Recapitalization Policies in Response to the Sub-Prime Crisis’. Comparative
European Politics 9, 76–99.
Myles, John (1996): ‘When Markets Fail: Social Welfare in Canada and the United
States’. In Gøsta Esping-Andersen (ed.) Welfare States in Transition: National
Adaptations in Global Economies. London: Sage, pp. 116–40.
Myles, John (1998): ‘How to Design a “Liberal” Welfare State: A Comparison of
Canada and the United States’. Social Policy and Administration, 32, 341–64.
Myles, John and Pierson, Paul (2001): ‘The Comparative Political Economy of
Pension Reform’. In Paul Pierson (ed.) The New Politics of the Welfare State.
Negrelli, Serafino (2000): ‘Social Pacts in Italy and Europe: Similar Strategies and
Structures; Different Models and National Stories’. In Giuseppe Fajertag and
Nölke, Andreas and Vliegenthart, Arjan (2009): ‘Enlarging the Varieties of
Capitalism: The Emergence of Dependent Market Economies in East Central
Europe’. World Politics, 61, 670–702.
Nordlund, A. (2000): ‘Social Policy in Harsh Times. Social Security Development
in Denmark, Finland, Norway and Sweden During the 1980s and 1990s’.
International Journal of Social Welfare, 9, 31–42.
Nørgaard, Asbjørn Sonne (2000): ‘Party Politics and the Organization of the
Danish Welfare State, 1890–1920: The Bourgeois Roots of the Modern Welfare
State’. Scandinavian Political Studies, 23, 183–215.
North, Douglass (1990): Institutions, Institutional Change and Economic Performance.
Cambridge: Cambridge University Press.
Nygård, Mikael (2006): ‘Welfare-Ideological Change in Scandinavia: A Comparative
Analysis of Partisan Welfare State Positions in Four Nordic Countries,
1970–2003’. Scandinavian Political Studies, 29, 356–85.
Obinger, Herbert (2009): ‘Religion and the Consolidation of the Swiss Welfare
State, 1848–1945’. In Kees van Kersbergen and Philip Manow (eds) Religion,
Class Coalitions, and Welfare States, 1st edn. Cambridge: Cambridge University
Press, pp. 176–209.
202 References
O’Donnell, Rory (2004): ‘Ireland: Social Partnership and the “Celtic Tiger”
Economy’. In Jonathan Perraton and Ben Clift (eds) Where are National
Capitalisms Now? New York: Palgrave Macmillan, pp. 50–69.
OECD (2001): Contributions of Financial Systems to Growth in OECD Countries.
Economics Department Working Papers No. 280. Paris: Organization for
Economic Development and Cooperation. URL: http://www.oecd.org/
dataoecd/38/43/1888670.pdf – 28 May 2006
OECD (2004): Employment Outlook 2004. Paris: Organization for Economic
Development and Cooperation.
OECD (2005a): Pensions at a Glance: Public Policies across OECD Countries. Paris:
Organization for Economic Development and Cooperation.
OECD (2005b): Employment Outlook 2005. Paris: Organization for Economic
OECD (2005c): OECD in Figures 2005. Paris: Organization for Economic Development
and Cooperation. Available at: http://www.oecd.org/document/62/0,2340,en_
2649_34489_2345918_1_1_1_1,00.html – downloaded 3 June 2006.
OECD (2006): OECD Factbook 2006. Paris: Organisation for Economic
Development and Cooperation. Available at: http://titania.sourceoecd.org/
vl=2492909/cl=11/nw=1/rpsv/factbook/ – downloaded 3 June 2006.
OECD (2009): OECD Employment Outlook 2009. Paris: Organization for Economic
Oesch, Daniel (2007): ‘Weniger Koordination, mehr Markt? Kollektive
Arbeitsbeziehungen und Neokorporatismus in der Schweiz seit 1990’. Swiss
Political Science Review, 13, 337–68.
Offe, Claus (1984): ‘Arbeitsgesellschaft’: Strukturprobleme und Zukunftsperspektiven.
Frankfurt: Campus.
Ogus, AI (1982): ‘Great Britain’. In Peter A. Kohler and Hans F. Zacher (eds)
The Evolution of Social Insurance, 1881–1981. New York: St. Martin’s Press,
pp. 150–264.
Olson, Mancur (1982): The Rise and Decline of Nations: Economic Growth,
Stagflation, and Social Rigidities. Yale, CT: Yale University Press.
Orloff, Ann Shola and Skocpol, Theda (1984): ‘Why Not Equal Protection?
Explaining the Politics of Public Social Spending in Britain, 1900–1911, and
the United States, 1880s–1920’. American Sociological Review, 49, 726–50.
Palier, Bruno (2010a): ‘Continental Western Europe’. In Francis G. Castles,
pp. 601–15.
Palier, Bruno (2010b): ‘The Dualizations of the French Welfare System’. In
Bruno Palier (ed.) A Long Goodbye to Bismarck? The Politics of Welfare Reforms in
Continental Europe. Amsterdam: Amsterdam University Press, pp. 73–99.
Palier, Bruno (ed.) (2010c): A Long Goodbye to Bismarck? The Politics of Welfare
Reforms in Continental Europe. Amsterdam: Amsterdam University Press.
Palier, Bruno (2010d): ‘Ordering Change: Understanding the “Bismarckian”
Welfare Reform Trajectory’. In Bruno Palier (ed.) A Long Goodbye to Bismarck?
The Politics of Welfare Reforms in Continental Europe. Amsterdam: Amsterdam
Palier, Bruno (2011): Understanding the Reform Trajectory of the Bismarckian Welfare
Systems. Paper for presentation at the Max Planck Institute for the Study of
References 203
Societies. With excerpts from Bruno Palier (ed.) A Long Goodbye to Bismarck?
The Politics of Welfare Reforms in Continental Europe.
Pasture, Patrick T. (1993): ‘The April 1944 “Social Pact” in Belgium and Its
Significance for the Post-War Welfare State’. Journal of Contemporary History,
28, 695–714.
Peck, Jamie and Theodore, Nik (2007): ‘Variegated Capitalism’. Progress in Human
Geography, 31, 731–72.
Peng, Ito and Wong, Joseph (2010): ‘East Asia’. In Francis G. Castles, Stephan
Leibfried, Jane Lewis, Herbert Obinger and Christopher Pierson (eds) The Oxford
Handbook of the Welfare State. Oxford: Oxford University Press, pp. 656–70.
Pérez, Sofía A. (2000): ‘Social Pacts in Spain’. In Giuseppe Fajertag and Philippe
Pochet (eds) Social Pacts in Europe. Brussels: European Trade Union Institute,
pp. 343–63.
Pesonen, Pertti and Riihinen, Olavi (2002): Dynamic Finland. The Political System
and the Welfare State. Helsinki: Finnish Literature Society.
Pierson, Paul (1994): Dismantling the Welfare State? Reagan, Thatcher and the
Politics of Retrenchment. Cambridge: Cambridge University Press.
Pierson, Paul (1996): ‘The New Politics of the Welfare State’. World Politics, 48(2),
143–79.
Pierson, Paul (2000): ‘Three Worlds of Welfare State Research’. Comparative
Political Studies, 33(6/7), 791–821.
Pierson, Paul (2001a): The New Politics of the Welfare State. Oxford: Oxford
University Press.
Pierson, Paul (2001b): ‘Coping with Permanent Austerity: Welfare State
Restructuring in Affluent Democracies’. In Paul Pierson (ed.) The New Politics
of the Welfare State. Oxford: Oxford University Press, pp. 410–56.
Pierson, Paul (2001c): Explaining Institutional Origins and Change. Manuscript.
Available at: http://www.polisci.berkeley.edu/faculty/bio/permanent/Pierson,
P/ch4.pdf – downloaded 28 May 2006.
Piore, Michael and Sabel, Charles (1984) The Second Industrial Divide. New York:
Basic Books.
Polanyi, Karl (1944): The Great Transformation. New York: Farrar & Rinehart.
Pontusson, Jonas (1997): ‘Between Neo-Liberalism and the German Model:
Swedish Capitalism in Transition’. In Colin Crouch and Wolfgang Streeck
(eds) Political Economy of Modern Capitalism: Mapping Convergence and Diversity.
London: Sage, pp. 55–70.
Pontusson, Jonas (2005a): Inequality and Prosperity: Social Europe vs. Liberal
America. Ithaca, NY: Cornell University Press.
Pontusson, Jonas (2005b): ‘Varieties and Commonalities of Capitalism’. In David
Coates (ed.) Varieties of Capitalism, Varieties of Approaches. New York: Palgrave
Macmillan, pp. 163–88.
Porta, Rafael, Botero, Juan, Djankov, Simeon, López-de-Silanes, Florencio and
Shleifer, Andrei (2004): ‘The Regulation of Labour’. The Quarterly Journal of
Economics, 119(4), 1339–82.
Porter, Michael (1990): The Competitive Advantage of Nations. New York: Free
Press.
Porter, Michael (2000): ‘Attitudes, Values, Beliefs and the Microeconomics
of Prosperity’. In Lawrence Harrison and Samuel Huntington (eds) Culture
Matters: How Values Shape Human Progress. New York: Basic Books, pp. 15–28.
204 References
Putnam, Robert (1993): Making Democracy Work. Civic Traditions in Modern Italy.
Princeton, NJ: Princeton University Press.
Redding, Gordon and Witt, Michael (2009): ‘China’s Business System and its
Future Trajectory’. Asia Pacific Journal of Management, 26, 381–99.
Regini, Marino (1997): ‘Social Institutions and Production Structure: The Italian
Variety of Capitalism in the 1980s’. In Colin Crouch and Wolfgang Streeck
(eds) Political Economy of Modern Capitalism: Mapping Convergence and Diversity.
London: Sage, pp. 102–16.
Rhodes, Martin (2000): ‘Restructuring the British Welfare State: Between
Domestic Constraints and Global Imperatives’. In Fritz Wilhelm Scharpf
and Vivien A. Schmidt (eds) Welfare and Work in the Open Economy. Volume II.
Diverse Responses to Common Challenges. Oxford, Oxford University Press,
pp. 19–68.
Rhodes, Martin (2005): ‘Varieties of Capitalism and the Political Economy of
European Welfare States’. New Political Economy, 10(3), 363–70.
Ricardo, David (1821): On the Principles of Political Economy and Taxation.
London.
Roche, William K. (2007): ‘Social Partnership in Ireland and New Social Pacts.
Industrial Relations’. A Journal of Economy and Society, 46, 395–425.
Rogow, Robert (1993): ‘Canada’. In Miriam Rothman, Dennis R. Briscoe and
Raoul C. D. Nacamulli (eds) Industrial Relations around the World. Labor Relations
for Multinational Companies. Berlin: De Gruyter, pp. 67–86.
Rothstein, Bo (1992): ‘Labour-Market Institutions and Working-Class Strengths’.
In Frank Longstreth, Sven Steinmo and Kathleen Thelen (eds) Structuring Politics.
Historical Institutionalism in Comparative Analysis. Cambridge: Cambridge
Rothstein, Bo (1998): Just Institutions Matter: The Moral and Political Logic of the
Universal Welfare State. Cambridge: Cambridge University Press.
Royo, Sebastián (2007): ‘Varieties of Capitalism in Spain: Business and the Politics
of Coordination’. European Journal of Industrial Relations, 13, 47–65.
Sachs, Jeffrey (2000): ‘Notes on a New Sociology of Economic Development’. In
Lawrence Harrison and Samuel Huntington (eds) Culture Matters: How Values
Shape Human Progress. New York: Basic Books, pp. 29–43.
Salonen, Tapio (2001): ‘Sweden: Between Model and Reality’. In Pete Alcock
and Gary Craig (eds) International Social Policy. Welfare Regimes in the Developed
World. New York: Palgrave Macmillan, pp. 143–60.
Scharpf, Fritz (1999): The Viability of Advanced Welfare States in the International
Economy: Vulnerabilities and Options. MPIfG Working Paper 99/9. Cologne: Max
Planck Institute for the Study of Societies. Available at: http://www.mpi-fg-koeln.
mpg.de/pu/workpap/wp99-9/wp99-9.html – downloaded 22 October 2005.
Scharpf, Fritz (2000): ‘Economic Changes, Vulnerabilities and Institutional
Capabilities’. In Welfare and Work in the Open Economy. Volume I: From
Vulnerability to Competitiveness. Oxford: Oxford University Press, pp. 21–124.
Scharpf, Fritz and Schmidt Vivien (2000a): ‘Introduction’. In Welfare and Work
in the Open Economy. Volume I: From Vulnerability to Competitiveness. Oxford:
Scharpf, Fritz and Schmidt, Vivien (2000b): Welfare and Work in the Open Economy.
Volume I: From Vulnerability to Competitiveness. Oxford: Oxford University
Press.
References 205
Scharpf, Fritz and Vivien Schmidt (2000c): Welfare and Work in the Open Economy.
Volume II: Diverse Responses to Common Challenges. Oxford: Oxford University
Press.
Scheurell, Robert P. (2003): ‘The Welfare State in the United States’. In Christian
Aspalter (ed.) Welfare Capitalism around the World. Taichung: Casa Verde
Publishing, pp. 27–44.
Schmidt, Vivien (2000): ‘Still Three Models of Capitalism? The Dynamics of
Economic Adjustment in Britain, Germany and France’. In Roland Czada
and Susanne Lütz (eds) Die politische Konstitution von Märkten. Opladen:
Westdeutscher Verlag, pp. 38–72.
Schmidt, Vivien (2002): The Futures of European Capitalism. Oxford: Oxford
University Press.
Schneider, Ben Ross (2009): ‘Hierarchical Market Economies and Varieties of
Capitalism in Latin America’. Journal of Latin American Studies, 41, 553–75.
Schneider, Martin R. and Paunescu, Mihai (forthcoming 2012): ‘Changing
Varieties of Capitalism and Revealed Comparative Advantages from 1990 to
2005: A Test of the Hall and Soskice Claims’. Socio-Economic Review, advance
access. Available at: http://ser.oxfordjournals.org/content/early/2012/02/06/
ser.mwr038.abstract.
Schumpeter, Joseph Alois (2003 [1943]): Capitalism, Socialism and Democracy,
e-library, Taylor & Francis. Available at: http://www.routledge.com/books/
details/9780415567893.
Schwartz, Herman (2000): ‘Internationalization and Two Liberal Welfare States.
Australia and New Zealand’. In Fritz Wilhelm Scharpf and Vivien A. Schmidt
(eds) Welfare and Work in the Open Economy. Volume II. Diverse Responses to
Common Challenges. Oxford: Oxford University Press, pp. 69–130.
Scruggs, Lyle and James, Allan (2006): ‘Welfare-state decommodification in 18
OECD countries: a replication and revision.’ Journal of European Social Policy
16(1), 55–72.
Scruggs, Lyle and Pontusson, Jonas (2008): New Dimensions of Welfare State
Regimes in Advanced Democracies, Paper presented at the 2008 Annual Meeting
of the American Political Science Association. Available at: http://sp.uconn.
edu/~scruggs/apsa08.pdf – downloaded 21 January 2013.
Shalev, Michael (1999): Limits and Alternatives to Multiple Regression in Macro-
Comparative Research. European Forum Paper WS/80. Florence: European
University Institute.
Shonfield, Andrew (1965): Modern Capitalism. The Changing Balance of Public and
Private Power. New York: Oxford University Press.
Shonfield, Andrew (1969): Modern Capitalism. New York: Oxford University
Press.
Smith, Adam (1979 [1759]): The Theory of Moral Sentiments. Oxford: Clarendon
Press.
Smith, Adam (1979 [1776]): The Wealth of Nations, ed. Andrew Skinner.
Baltimore: Penguin.
Smith, Adam (1993 [1776]): An Inquiry into the Nature and Causes of the Wealth of
Nations, A Selected Edition. Oxford: Oxford University Press.
Soederberg, Susanne, Menz, Georg and Cerny, Philip (eds) (2005): Internalizing
Globalization: The Rise of Neoliberalism and the Decline of National Varieties of
Capitalism. New York: Palgrave Macmillan.
206 References
Sørensen, Aage B. (1998): ‘On Kings, Pietism and Rent-seeking in Scandinavian

Welfare States’. Acta Sociologica, 41, 363–75.
Sorge, Arndt (2005): The Global and the Local. Understanding the Dialectics of
Business Systems. Oxford: Oxford University Press.
Soskice, David (1999): ‘Divergent Production Regimes: Coordinated and
Uncoordinated Market Economies in the 1980s and 1990s’. In Herbert
Kitschelt, Peter Lange, Gary Marks and John Stephens (eds) Continuity and
Change in Contemporary Capitalism. Cambridge: Cambridge University Press,
pp. 101–34.
Soskice, David (2005): ‘Varieties of Capitalism and Cross-national Gender
Differences’. Social Politics, 12, 170–9. Available at: ISI://000231658800002.
Soskice, David (2007): ‘Macroeconomics and Varieties of Capitalism’. In Bob
Conflict, Contradictions and Complementarities in the European Economy. Oxford:
Soskice, David (2009): ‘Varieties of Capitalism; Varieties of Reform’. In Anton
Hemerijck, Ben Knapen and Ellen Van Doorn (eds) Aftershocks: Economic
Crisis and Institutional Choice. Amsterdam: Amsterdam University Press,
pp. 133–41.
Soskice, David and Iversen, Torben (2001): ‘An Asset Theory of Social Policy
Preferences’. American Political Science Review, 95, 875–93.
Starke, Peter, Obinger, Herbert and Castles, Francis G. (2008): ‘Convergence
Towards Where: In What Ways, If Any, Are Welfare States Becoming More
Similar?’ Journal of European Public Policy, 15, 975–1000.
Steinmo, Sven (2005): ‘The Evolution of the Swedish Model’. In Susanne
Soederberg, Georg Menz and Philip Cerny (eds) Internalizing Globalization: The
Rise of Neoliberalism and the Decline of National Varieties of Capitalism. New York:
Palgrave Macmillan, pp. 149–64.
Steinmo, Sven (2010): The Evolution of Modern States: Sweden, Japan, and the United
States. New York: Cambridge University Press.
Streeck, Wolfgang (1992a): ‘Productive Constraints: On the Institutional
Conditions of Diversified Quality Production’. In Wolfgang Streeck (ed.) Social
Institutions and Economic Performance. Studies of Industrial Relations in Advanced
Capitalist Economies. London: Sage, pp. 1–40.
Streeck, Wolfgang (1992b): Social Institutions and Economic Performance. Studies of
Industrial Relations in Advanced Capitalist Economies. London: Sage.
Streeck, Wolfgang (1997a): ‘Beneficial Constraints: On the Economic Limits
of Rational Voluntarism’. In Robert Boyer and J. Rogers Hollingsworth
(eds) Contemporary Capitalism. The Embeddedness of Institutions. Cambridge:
Cambridge University Press, pp. 197–219.
Streeck, Wolfgang (1997b): ‘German Capitalism. Does it Exist? Can it Survive?’
In Colin Crouch and Wolfgang Streeck (eds) Political Economy of Modern
Capitalism: Mapping Convergence and Diversity. London: Sage, pp. 33–54.
Streeck, Wolfgang (1999): Korporatismus in Deutschland. Zwischen Nationalstaat
und Europäischer Union. Frankfurt: Campus.
Streeck, Wolfgang (2001a): ‘Introduction: Explorations into the Origins of
Nonliberal Capitalism in Germany and Japan’. In Wolfgang Streeck and Kozo
Yamamura (eds) The Origins of Nonliberal Capitalism: Germany and Japan in
Comparison. Ithaca, NY: Cornell University Press, pp. 1–38.
References 207
Streeck, Wolfgang (2001b): ‘Preface to “Comparing Welfare Capitalisms”’. In

Bernhard Ebbinghaus and Philip Manow (eds) Varieties of Welfare Capitalism.
Social Policy and Political Economy in Europe, Japan and the USA. London:
Routledge, pp. xv–xx.
Streeck, Wolfgang (2001c): The Transformation of Corporate Organization in Europe:
An Overview. MPIfG Working Paper 01/8. Cologne: Max Planck Institute for
the Study of Societies. Available at: http:// www.mpi-fg-koeln.mpg.de/pu/
workpap/wp01-8/wp01-8.html – downloaded 14 June 2005.
Streeck, Wolfgang (2002): Note On Complementarity: How It Comes About and
How We Should Analyse It. Paper Presented at the Seminar ‘Institutional
Complementarity and the Dynamics of Economic Systems’, Paris, 5–6 April.
Streeck, Wolfgang (2009): Re-Forming Capitalism: Institutional Change in the
German Political Eonomy. Oxford and New York: Oxford University Press.
Streeck, Wolfgang and Thelen, Kathleen (2005a): Beyond Continuity: Institutional
Change in Advanced Political Economies. Oxford: Oxford University Press.
Streeck, Wolfgang and Thelen, Kathleen (2005b): ‘Introduction: Institutional
Change in Advanced Political Economies’. In Wolfgang Streeck and Kathleen
Thelen (eds) Beyond Continuity: Institutional Change in Advanced Political
Economies. Oxford: Oxford University Press, pp. 1–39.
Svallfors, Stefan (1997): ‘Worlds of Welfare and Attitudes to Redistribution: A
Comparison of Eight Western Nations’. European Sociological Review, 13, 283–304.
Svallfors, Stefan (2010): ‘Public Attitudes’. In Francis G. Castles, Stephan Leibfried,
Jane Lewis, Herbert Obinger and Christopher Pierson (eds) The Oxford Handbook
of the Welfare State. Oxford: Oxford University Press, pp. 241–51.
Swank, Duane (2001): ‘Political Institutions and Welfare State Restructuring: The
Impact of Institutions on Social Policy Change in Developed Democracies’.
In Paul Pierson (ed.) The New Politics of the Welfare State. Oxford: Oxford
Swank, Duane (2002): Global Capital, Political Institutions, and Policy Change in
Developed Welfare States. Cambridge: Cambridge University Press.
Swenson, Peter (2002): Capitalists Against Markets: The Making of Labor Markets and
Welfare States in the United States and Sweden. Oxford: Oxford University Press.
Teague, Paul (2009): ‘Path Dependency and Comparative Industrial Relations:
The Case of Conflict Resolution Systems in Ireland and Sweden’. British Journal
of Industrial Relations, 47, 499–520.
Thatcher, Mark (2007): ‘Reforming National Regulatory Institutions. The EU
and Cross-national Variety in European Network Industries’. In Bob Hancké,
Martin Rhodes and Mark Thatcher (eds) Beyond Varieties of Capitalism, 1st edn.
Thelen, Kathleen (2000): ‘Why German Employers Cannot Bring Themselves to
Dismantle the German Model’. In Torben Iversen, Jonas Pontusson and David
Soskice (eds) Unions, Employers and Central Banks. Cambridge: Cambridge
Thelen, Kathleen (2010): ‘Economic Regulation and Social Solidarity: Conceptual
and Analytic Innovations in the study of Advanced Capitalism’. Socio-Economic
Review, 8, 187–207.
Thelen, Kathleen (2012): ‘Varieties of Capitalism: Trajectories of Liberalization
and the New Politics of Social Solidarity’. Annual Review of Political Science, 15,
137–59.
208 References
Thorkildsen, Dag (1997): ‘Religious Identity and Nordic Identity’. In Øystein

Sørensen and Bo Stråth (eds) The Cultural Construction of Norden. Oslo:
Scandinavian University Press, pp. 138–60.
Tipton, Frank (2009): ‘Southeast Asian Capitalism: History, Institutions, States,
and Firms’. Asia Pacific Journal of Management, 26, 401–34.
Titmuss, Richard M. (1974) Social Policy. An Introduction. London: Allen & Unwin.
Trampusch, Christine (2009a): Der erschöpfte Sozialstaat. Transformation eines
Politikfeldes. Frankfurt: Campus.
Trampusch, Christine (2009b): ‘Religion, Parteien und Industrielle Beziehungen.
Die Entstehung der Arbeitslosenversicherung in den Niederlanden’. Zeitschrift
für Sozialreform, 55, 293–316.
Trampusch, Christine (2010a): ‘Co-evolution of Skills and Welfare in Coordinated
Market Economies? A Comparative Historical Analysis of Denmark, the Netherlands
and Switzerland’. European Journal of Industrial Relations, 16, 197–220.
Trampusch, Christine (2010b): ‘The Welfare State and Trade Unions in Switzerland:
An Historical Reconstruction of the Shift from a Liberal to a Post-liberal Welfare
Regime’. Journal of European Social Policy, 20, 58–73.
Tricker, R. Ian (2009): Corporate Governance: Principles, Policies and Practices.
Oxford: Oxford University Press.
Trigilia, Carlo and Burroni, Luigi (2009): ‘Italy: Rise, Decline and Restructuring of
a Regionalized Capitalism’. Economy and Society, 38, 630–53.
Uzuhashi, Takafumi Ken (2001): ‘Japan: Bidding Farewell to the Welfare Society’.
In Pete Alcock and Gary Craig (eds) International Social Policy. Welfare Regimes
in the Developed World. New York: Palgrave Macmillan, pp. 104–23.
Valocchi, Steve (1992): ‘The Origins of the Swedish Welfare State: A Class Analysis
of the State and Welfare Politics’. Social Problems, 39, 189–200.
Van Kersbergen, Kies (1995): Social Capitalism: A Study of Christian Democracy and
the Welfare State. London: Routledge.
Van Leeuwen, Marco H. D. (1997): ‘Trade Unions and the Provision of Welfare in
the Netherlands, 1910–1960’. The Economic History Review, 50, 764–91.
Van Rie, Tim, Marx, Ive and Horemans, Jeroen (2011): ‘Ghent Revisited:
Unemployment Insurance and Union Membership in Belgium and the Nordic
Countries’. European Journal of Industrial Relations, 17, 125–39.
Venn, Danielle (2009): Legislation, Collective Bargaining and Enforcement: Updating
the OECD Employment Protection Indicators. OECD Social, Employment and
Migration Working Papers, no 89. Paris: OECD. Available at: http://www.oecd.
org/els/workingpapers – downloaded 21 January 2013.
Visser, Jelle (2011): Codebook of the Data Base on Institutional Characteristics of Trade
Unions, Wage Setting, State Intervention and Social Pacts, 1960–2010 (ICTWSS)
Version 3.0. Amsterdam: Amsterdam Institute for Advanced Labour Studies.
Vitols, Sigurt (2001a): ‘The Origins of Bank-based and Market-based Financial
Systems: Germany, Japan and the United States’. In Wolfgang Streeck and
Kozo Yamamura (ed.) The Origins of Nonliberal Capitalism: Germany and Japan in
Comparison. Ithaca, NY and London: Cornell University Press, pp. 171–99.
Vitols, Sigurt (2001b): ‘Varieties of Corporate Governance: Comparing Germany
and the UK’. In Peter Hall and David Soskice (eds) Varieties of Capitalism: The
Institutional Foundations of Comparative Advantage. Oxford: Oxford University
Press, pp. 337–60.
Vogel, Steven (2001): ‘The Crisis of German and Japanese Capitalism. Stalled on the
Road to the Liberal Market Model?’ Comparative Political Studies, 34(10), 1103–33.
References 209
Weber, Max (1949): Max Weber on the Methodology of the Social Sciences, translated
and edited by Shils, Edward and Finch, Henry. New York: The Free Press.
Weber, Max (1958 [1904]): The Protestant Ethic and the Spirit of Capitalism. New York:
Charles Scribner’s Sons.
Weber, Max (1965 [1905]): ‘Die protestantische Ethik und der Geist des
Kapitalismus’. In Johannes Winckelmann (ed.) Die protestantische Ethik I. Eine
Aufsatzsammlung. Hamburg: Siebenstern Taschenbuch Verlag, pp. 27–317.
Westney, Eleanor (1996): ‘The Japanese Business System: Key Features and
Prospects for Change’. Journal of Asian Business, 12(1), 21–50.
Whitley, Richard (1998): ‘Internationalization and Varieties of Capitalism: The
Limited Effects of Cross-National Coordination of Economic Activities on the
Nature of Business Systems’. Review of International Political Economy, 5(3),
445–81.
Whitley, Richard (1999): Divergent Capitalisms: The Social Structuring and Change
of Business Systems. Oxford: Oxford University Press.
Whitley, Richard (2007): Business Systems and Organizational Capabilities: The
Institutional Structuring of Competitive Competences. Oxford: Oxford University
Press.
Wilensky, Harold and Lebeaux, Charles (1958): Industrial Society and Social
Welfare. New York: Free Press.
Williamson, Oliver (1975): Markets and Hierarchies, Analysis and Antitrust
Implications: A Study in the Economics of Internal Organization. New York: Free
Press.
Winckelmann, Johannes (1965): Die protestantische Ethik I. Eine Aufsatzsammlung.
Hamburg: Siebenstern Taschenbuch Verlag.
Wincott, Daniel (2001): ‘Reassessing the Social Foundations of Welfare (State)
Regimes’. New Political Economy, 2001, 6(3), 409–25.
Witt, Michael A. (2010): China: What Variety of Capitalism? INSEAD Working
Paper.
Wood, Stewart (2001a): ‘Business, Government and Patterns of Labour Market
Policy in Britain and the Federal Republic of Germany’. In Peter Hall and David
Soskice (eds) Varieties of Capitalism: The Institutional Foundations of Comparative
Advantage. Oxford: Oxford University Press, pp. 247–74.
Wood, Stewart (2001b): ‘Labour Market Regimes under Threat? Sources of
Continuity and Change in Germany, Britain and Sweden’. In Paul Pierson
pp. 368–409.
World Federation of Exchanges (2004): ‘Annual Report and Statistics’. World
Federation of Exchanges. Available at: http://www.world-exchanges.org/WFE/
home.asp?menu=11&document=2816 – downloaded 28 May 2006.
World Bank (2006): ‘World Development Indicators’. Available at: http://devdata.
worldbank.org/wdi2006/contents/toc.htm – downloaded 28 May 2006.
Zolberg, Aristide (1986): ‘How Many Exceptionalisms?’ In Ira Katznelson and
Aristide Zolberg (eds) Working-Class Formation: Nineteenth-Century Patterns in
Western Europe and the United States. Princeton, NJ: Princeton University Press,
pp. 397–456.
Zugehör, Rainer (2003): Die Zukunft des deutschen Kapitalismus. Unternehmen
zwischen Kapitalmarkt und Mitbestimmung. Opladen: Leske und Budrich.
Zysman, John (1994): ‘How Institutions Create Historically Rooted Trajectories of
Growth’. Industrial and Corporate Change, 1, 243–83.
Appendix
A Data used
AT AUS BE CA CH DK ESP FIN FR
Hall and 1.00 0.74 0.13 0.51 0.70 0.57 0.72 0.69
Gingerich
coordination1
Hicks and 0.96 0.67 0.04 0.55 0.72 0.88 0.97
Kenworthy
corporatism2
Traditional 1.48 –1.23 1.22 –0.90 –1.39 –0.94 0.46 1.16
conservative
Hicks and
Kenworthy3
Social –0.13 –0.58 0.35 –0.88 –0.88 0.98 0.81 –0.73
democratic
Hicks and
Kenworthy4
Conservatism 8 0 8 2 0 2 6 8
Esping-Andersen5
Liberalism 4 10 4 12 12 6 4 8
Esping-Andersen6
Socialism 2 4 4 4 4 8 6 2
Esping-Andersen7
Dummy for 0 0 1 0 1 1 1 1 1
coordination
Dummy for 0 1 0 1 1 0 0 0 0
liberal welfare
state
Dummy for 0 0 0 0 0 1 0 1 0
social democratic
welfare state
Dummy for 1 0 1 0 0 0 1 0 1
conservative
welfare state
Gini World 29.1 35.2 33.0 32.6 33.7 24.7 34.7 26.9 32.7
Bank8
Gini CIA9 31.0 35.2 25.0 33.1 33.1 23.2 32.5 26.9 32.7
Lowest 1010 3.3 2.0 3.4 2.6 2.9 2.6 2.6 4.0 2.8
Lowest 2011 8.6 5.9 8.5 7.2 7.6 8.3 7.0 9.6 7.2
Richest 2012 37.8 41.3 41.4 39.9 41.3 35.8 42.0 36.7 40.2
Richest 1013 23.0 25.4 28.1 24.8 25.9 21.3 26.6 22.6 25.1
Unemployment14 5.3 5.5 7.4 7.2 4.4 5.3 11.0 8.9 9.6
Labour 70.2 73.6 65.3 78.2 81.0 80.2 69.7 73.8 69.5
participation15
Share industry16 27.8 21.4 24.9 22.3 23.7 23.7 30.5 25.7 23.0
Share service17 67.2 74.9 73.1 75.0 72.6 73.1 64.0 69.3 72.6
ISIC 118 17.6 13.6 13.8 13.6 15.4 13.5 18.9 11.9 12.4
ISIC 619 13.9 15.4 16.5 14.0 8.6 20.2 14.6 17.1 15.5
210
211
GER IRE IT JA NL NOR NZ POR SWE UK US
0.95 0.29 0.87 0.74 0.66 0.76 0.21 0.72 0.69 0.07 0.00
0.80 0.07 0.44 0.77 0.58 0.96 0.14 0.40 0.10 0.02
1.06 0.65 1.58 0.01 –0.35 0.03 –0.10 –0.95 –0.91 –0.87
0.04 0.17 –0.90 –1.04 0.20 1.89 0.63 1.88 0.03 –1.83
8 4 8 4 4 4 2 0 0 0
6 2 6 10 8 0 2 0 6 12
4 2 0 2 6 8 4 8 4 0
1 0 1 1 1 1 0 1 1 0 0
0 1 0 0 0 0 1 0 0 1 1
0 0 0 0 0 1 0 0 1 0 0
1 0 1 1 1 0 0 1 0 0 0
28.3 34.3 36.0 24.9 30.9 25.8 36.2 38.5 25.0 36.0 40.8
28.3 35.9 36.0 37.9 30.9 25.8 36.2 38.5 25.0 36.8 45.0
3.2 2.9 2.3 4.8 2.5 3.9 2.2 2.0 3.6 2.1 1.9
8.5 7.4 6.5 10.6 7.6 9.6 6.4 5.8 9.1 6.1 5.4
36.9 42.0 42.0 35.7 38.7 37.2 43.8 45.9 36.6 44.0 45.8
22.1 27.2 26.8 21.7 22.9 23.4 27.8 29.8 22.2 28.5 29.9
9.9 4.4 8.1 4.9 4.7 4.5 4.0 7.0 6.6 4.7 5.6
72.7 68.6 62.5 72.2 76.6 79.1 76.6 72.9 78.7 76.2 75.4
31.0 27.7 31.0 28.4 20.3 20.9 22.7 31.4 22.6 22.3 20.0
66.6 65.9 64.5 67.1 76.6 75.6 69.8 56.5 75.2 76.4 78.4
11.8 12.0 16.4 12.7 14.2 10.4 15.4 17.2 12.0 15.4 15.5
15.7 13.3 14.3 17.4 17.2 17.9 13.0 17.7 20.3 17.4 15.8
(continued)
212
Appendix A Continued
ISIC 6 21.35 15.40 23.95 23.40 16.20 19.30 14.85 14.60

employment20
Employment 31.20 15.50 31.12 18.59 15.87 33.80 15.56 25.50 21.52
public21
Employer –0.09 –0.52 –0.09 –1.68 –0.09 –0.09 1.11 –0.09
centralization22
Employer –1.13 –1.13 0.00 –1.13 1.13 1.13 1.13 0.00
coordination23
Enterprise 0.44 –1.01 0.44 –1.01 0.44 0.44 0.90 –0.20
cooperation24
Union power25 1.65 –0.24 0.28 –0.56 –0.74 0.82 0.43 –1.06
Collective –0.30 0.83 0.29 –1.02 –0.30 1.13 0.68 –0.30
bargaining 26
Union density27 37 25 56 28 18 74 15 76 10
Coverage 95 80 90 32 40 80 80 90 90
collective wage
bargaining28
Income 64 46 62 66 77 79 65 71
replacement
unemployment29
Working hours30 1.550 1.814 1.542 1.733 1.556 1.475 1.800 1.669 1.431
Tax family31 29.4 13.1 40.2 20.5 17.9 31.3 31.0 38.8 39.4
Tax single32 44.7 23.1 55.6 30.2 29.5 44.2 37.9 45.9 48.3
Unemployment 32 22 42 15 33 50 36 36 24
benefit33
Pension 93 52 63 57 67 54 88 79 68
replacement
rate34
Taxes worker35 44.9 28.6 54.2 32.3 28.8 41.5 38.0 43.8 47.4
Tax revenue36 43.1 31.6 45.4 33.8 29.5 48.3 34.9 44.8 43.4
Social 25.96 18.00 24.72 17.81 26.41 29.22 19.57 24.80 28.45
expenditure37
Share means 5.9 3.2 18.9 6.6 3.0 11.5 11.5 3.1
tested38
Share private 0.30 0.33 0.29 0.30 0.41 0.17 0.29 0.23 0.15
health39
Expenditure of 57.21 36.24 55.28 14.08 55.74 76.49 41.21 60.81 62.18
GDP40
Employment 2.20 1.50 2.50 1.10 1.60 1.80 3.10 2.10 2.60
protection41
Index employee 1 0 0 0 0 1 0 0 1
appointment42
Index employee 1 0 1 0 0 1 1 1 0
participation
law43
Index works 1 0 1 0 0 1 1 1 0
council decision
power44
Employment 30 48 34 36 25 70 55 42
laws45
Index level of 0.50 0.25 0.50 0.00 0.25 0.25 0.50 1.00 0.50
bargaining46
213
17.05 20.80 17.10 22.70 19.85 18.80 21.90 19.10 16.60 19.70 21.90
15.70 20.02 14.96 8.70 26.48 37.64 19.81 14.27 34.37 19.98 16.60
1.11 –0.09 –0.09 –0.09 –0.09 1.11 –0.09 1.11 –0.09 –1.68
1.13 –1.13 –1.13 0.00 0.00 1.13 –1.13 1.13 –1.13 –1.13
0.90 –1.01 1.04 1.48 –0.92 0.44 –1.01 0.44 –0.92 –0.89
–0.60 –0.08 0.07 –0.71 –0.18 1.24 –0.34 1.47 –0.32 –0.95
–0.30 –0.30 1.10 –0.30 0.58 1.60 –1.40 1.35 –0.52 –1.15
25 38 35 22 23 54 23 24 79 31 13
90 80 15 80 70 25 80 90 30 14
66 43 47 56 76 70 43 73 32 58
1.441 1.646 1.591 1.801 1.354 1.338 1.813 1.678 1.563 1.672 1.822
32.6 12.8 35.6 20.4 32.4 27.2 16.8 24.2 41.1 17.8 19.4
50.7 25.8 46.2 24.2 42.3 37.0 19.6 32.5 48.6 29.7 30.0
29 38 34 8 53 34 28 41 39 16 14
72 37 89 59 84 65 40 80 69 48 51
50.7 23.8 45.7 26.6 43.6 36.9 20.7 32.6 48.0 31.2 29.6
35.5 29.7 43.1 25.3 38.8 43.4 34.9 37.1 50.6 35.6 25.6
27.39 13.75 24.45 16.89 21.75 23.90 18.53 21.10 29.78 21.82 14.73
9.6 26.6 4.3 3.7 11.4 3.7 100.0 8.8 12.4 15.8 39.8
0.22 0.25 0.25 0.18 0.38 0.16 0.21 0.30 0.24 0.17 0.56
55.04 50.25 50.97 46.68 57.04 63.83 39.04 42.32 78.15 19.90 42.87
2.50 1.30 2.40 1.80 2.30 2.60 1.30 3.50 2.90 1.10 0.70
1 0 0 0 0 1 0 0 0 0 0
1 0 0 0 1 0 0 1 1 0 0
1 0 1 0 1 0 0 1 1 0 0
51 49 59 37 54 41 32 79 50 28 22
0.50 0.75 0.25 0.00 0.50 0.50 0.00 0.75 0.25 0.00 0.00
(continued)
214
Appendix A Continued
Index 0.83 0.00 0.67 0.00 0.00 1.00 0.67 0.50 0.33
codetermination47
Investors 2 8 8 8 1 7 4 6 10
protection48
Bank credit49 0.83 0.47 0.37 0.48 1.44 0.42 0.66 0.67 0.82
Market 29.4 121.8 79.9 120.4 231.0 62.7 90.5 98.8 90.7
capitalization
200450
Market 0.07 0.43 0.26 0.46 0.71 0.22 0.18 0.18 0.20
capitalization51
Employment 0.50 0.35 0.51 0.26 0.45 0.57 0.74 0.74 0.74
laws52
Collective 0.36 0.37 0.42 0.20 0.42 0.42 0.59 0.32 0.67
relations laws53
Social security 0.71 0.78 0.62 0.79 0.82 0.87 0.77 0.79 0.78
laws54
Left power55 0.24 0.35 0.09 0.69 0.69 0.74 0.31 0.79 0.34
Factor 156 0.589 –0.897 0.734 –1.232 –0.654 1.149 0.227 1.178 0.506
Factor 257 –0.468 0.138 –0.379 0.831 0.790 1.550 –1.583 0.460 –0.316
Factor 358 0.283 –0.690 –0.997 –0.143 3.001 –0.744 0.560 –0.043 –0.114
Factor 459 0.425 0.181 0.304 0.371 1.138 0.028 0.746 –0.261 0.059
1. Degree of coordination for different economies (Hall and Gingerich 2004: 14).
2. Degree of corporatism for different societies (Hicks and Kenworthy 1998).
3. Degree of conservatism for different societies (Hicks and Kenworthy 2003: 33).
4. Degree of socialism as opposed to liberalism for different societies (Hicks and Kenworthy
2003: 33).
5. Degree of conservatism for different societies (Esping-Andersen 1990: 74).
6. Degree of liberalism for different societies (Esping-Andersen 1990: 74).
7. Degree of socialism for different societies (Esping-Andersen 1990: 74).
8. Gini Index – data for 1993–2000 (World Bank 2006).
9. Distribution of family income, Gini index – data for 1993–2000 (CIA (data for 2006),
URL: http://www.cia.gov/cia/publications/factbook/fields/2172.html).
10. Percentage share of income or consumption for the poorest 10 per cent of the population –
data for 1993–2000 (World Bank 2006).
11. Percentage share of income or consumption for the poorest 20 per cent of the population –
12. Percentage share of income or consumption for the richest 20 per cent of the population –
13. Percentage share of income or consumption for the richest 10 per cent of the population –
14. Unemployment rate – data for 2004 (OECD 2005b: 238).
15. Labour force participation rate for persons aged 15–64 years – data for 2004 (OECD
2005b: 238).
16. Employment in industry as share of all employment – data for 2004 (OECD (data for
2005), URL: http://dx.doi.org/10.1787/808800743257).
17. Employment in the service sector as share of all employment – data for 2004 (OECD
2005c, URL: http://dx.doi.org/10.1787/808800743257).
18. Contribution to total economy gross value added: wholesale and retail trade, restaurants
and hotels (ISIC 1) – data for 2003 (OECD 2005c, URL: http://dx.doi.org/10.1787/8728
52212688).
215
1.00 0.00 0.17 0.00 0.67 0.33 0.00 0.50 0.67 0.00 0.00
5 9 7 6 4 7 10 7 2 10 7
0.86 0.51 1.04 0.74 0.48 0.41 0.63 0.44 0.74 0.65
43.6 62.8 47.1 79.6 107.5 56.6 44.2 43.8 108.8 132.6 139.4
0.19 0.27 0.12 0.73 0.41 0.15 0.40 0.08 0.38 0.76 0.58
0.70 0.34 0.65 0.16 0.73 0.69 0.16 0.81 0.74 0.28 0.22
0.61 0.46 0.63 0.63 0.46 0.65 0.25 0.65 0.54 0.19 0.26
0.67 0.71 0.76 0.64 0.63 0.83 0.72 0.74 0.84 0.69 0.65
0.29 0.00 0.32 0.01 0.26 0.71 0.46 0.09 0.85 0.28 0.71
0.839 –0.775 0.421 –0.683 0.359 1.003 –1.455 0.213 1.526 –1.267 –1.781
–0.654 –0.695 –1.568 0.042 0.107 1.226 0.214 –2.143 1.427 0.494 0.527
0.891 –0.827 –0.420 1.702 0.396 –0.417 –1.403 –0.115 –0.134 –0.812 0.024
–0.931 –1.141 0.090 –3.050 1.315 –0.801 –0.434 0.452 0.361 –0.323 1.470
19. Contribution to total economy gross value added: education, health, social work and
other services (ISIC 6) – data for 2003 (OECD 2005c, URL: http://dx.doi.org/10.1787/
872852212688).
20. ISIC 6 employment (CE and DE) – data for 1997–2003. (OECD 2006, URL: http://www1.
oecd.org/scripts/cde/DoQuery.asp).
21. Employment in the public sector (government, social security institutions, non-profit
sector and public enterprises) as share of all employment – data for 1997–2003 (ILO data
for 2003, http://laboursta.ilo.org).
22. Employer centralization – average for 1979–95 (Martin and Swank 2002: 899).
23. Employer coordination – average for 1979–95 (Martin and Swank 2002: 899).
24. Enterprise cooperation – average for 1979–95 (Martin and Swank 2002: 899).
25. Central union power density – average for 1979–95 (Martin and Swank 2002: 899).
26. Collective bargaining – average for 1979–95 (Martin and Swank 2002: 899).
27. Trade union members as percentage of total workforce – data for 2000 (OECD 2004:
145).
28. Coverage of collective wage bargaining – data for 2000 (OECD 2004: 145).
29. Average net income replacement rate for fully ensured single worker with APW wage
and couple with two children and one APW wage – data for 1999 (Allan and Scruggs
2004).
30. Average annual hours actually worked per person in employment – data for 2003 (OECD
2005b: 255).
31. Deduction (per cent) of wage due to taxes and social security contributions for mar-
ried couple, two children, one average wage – data for 2001 (OECD, URL: http://www.
nationmaster.com/graph-T/tax_tot_tax_wed_sin_wor).
32. Deduction (per cent) of wage due to taxes and social security contributions for single
worker, no children, average wage – data for 2001 from (OECD, URL http://www.
nationmaster.com/graph-T/tax_tot_tax_wed_sin_wor).
33. OECD summary measure of unemployment benefit as percentage of old wage – data for
2003 (OECD, URL: http://www.oecd.org/dataoecd/25/31/34008592.xls).
216
34. Pension replacement rate – net mandatory pension programs replacement rates as per-
centage of pre-retirement net-earning for men with average wage – data for 2003 (OECD
2005a).
35. Taxes (including all social security contributions) on the average production worker
(average wage, no children) – data for 2004 (OECD 2006).
36. Total tax revenue (no social security payments) as share of GDP – data for 2003 (OECD
2006, URL: http://puck.sourceoecd.org/vl=7267625/cl=26/nw=1/rpsv/factbook/data/
09-03-01-t01.xls).
37. Public social expenditure (cash benefits, provision of goods and services and tax breaks
with social purposes) – data for 2001 (OECD 2006, URL: http://thesius.sourceoecd.org/
vl=30620998/cl=14/nw=1/rpsv/factbook/data/09-02-01-t01.xls).
38. Share of means-tested social transfers on all social transfers in per cent – data for second
half of 1990s (Eurostat data for 2003 and Gough et al. 1997 for non-European countries).
39. Private healthcare expenditure as share of total healthcare expenditure – data for 2003
(OECD 2005c – http://dx.doi.org/10.1787/530538806724).
40. Total government expenditure as share of GDP – data for 2004 (CIA World Factbook
2005, URL: www.cia.gov).
41. Employment protection (composite index of legal restrictions on the ability of employers
to lay off regular employees) – data for 2003 (OECD 2004: 117).
42. Index: Employees appoint board members, 1 yes; 0 no – data for 2002 (OECD Survey data
for 2003, URL: www.oecd.org/dataoecd/58/27/21755678.pdf, page 47f.).
43. Index: Employee participation required by law 1 yes; 0 no – data for 2002 (OECD Survey
data for 2003, URL: www.oecd.org/dataoecd/58/27/21755678.pdf, page 47f.).
44. Index: Works council decision power 0 none; 0.5 consultation; 1 co-decision – data for
2002 (OECD Survey data for 2003, URL: www.oecd.org/dataoecd/58/27/21755678.pdf,
page 48f.).
45. Employment law rigidities, index by Porta et al. (2004).
46. Index of dominant level of bargaining: 0 company or individual; 25 company/plant; 50
industry; 75 industry and central; 100 central. Data for 2004 – (EIRO data fo 2005, URL:
http://www.eiro.eurofound.eu.int/ and OECD 2004: 151).
47. Co-determination index, made of indices: employees appoint board members; employee
participation required by law; works councils decision rights – data for 2002 (OECD
Survey 2003, URL: www.oecd.org/dataoecd/58/27/21755678.pdf, page 47f.).
48. Index of protection for investors (ownership disclosure, measures that reduce expropria-
tion, and disclosures to help investors) – data for 2005 (World Bank 2006).
49. Deposit money bank credits to the private sector as a percentage of GDP – average for
1980–1995 (Levine 2000: 43).
50. Overall size of the stock market in US dollars as a percentage of GDP – data for 2004,
2000 for Belgium (World Bank 2006).
51. Value of domestic equities listed on domestic exchanges divided by GDP – average for
1980–1995 (Levine 2000: 43).
52. Employment law rigidities, index by Porta et al. (2004).
53. Collective relations laws index to measure the power of labour unions over working
conditions (Porta et al. 2004: 1362f.).
54. Social security laws index to measure decommodification by the pension- health- and
unemployment system (Porta et al. 2004: 1362f.).
55. Years in which chief executive and largest party in congress have left or centre political
orientation from 1928–95 (Porta et al. 2004: 1362f.).
56. First factor of principal factor analysis.
57. Second factor of principal factor analysis.
58. Third factor of principal factor analysis.
59. Fourth factor of principal factor analysis.
B Factor loadings
1 2 3 4
Labour force participation rate for persons aged 15–64 – data for 2004 –.116 .767 .211 .154
(OECD 2005b: 238).
Employment in industry as share of all employment – data for 2004 .264 –.781 .209 –.316
(OECD data for 2005. URL: http://dx.doi.org/10.1787/808800743257)
Employment in the service sector as share of all employment – data for 2004 –.178 .813 –.114 .266
(OECD 2005c, URL: http://dx.doi.org/10.1787/808800743257)
Contribution to total economy gross value added: wholesale and retail trade, –.273 –.545 .095 .480
restaurants and hotels (ISIC 1) – data for 2003 (OECD 2005c, URL: http://dx.
doi.org/10.1787/872852212688)
Contribution to total economy gross value added: education, health, social work .494 .244 –.386 –.186
and other services (ISIC 6) – data for 2003 (OECD 2005c, URL: http://dx.doi.org/
10.1787/872852212688)
Isic 6 employment (CE and DE) data for 1997–2003 (OECD 2006, –.754 .081 .362 –.013
Structure of the labour market

URL: http://www1.oecd.org/scripts/cde/DoQuery.asp)
Employment in the public sector (government, social security institutions, .615 .518 –.372 .182
non-profit sector and public enterprises) as share of all employment – data for
1997–2003 (ILO data for 2003, URL: http://laboursta.ilo.org).
Employment protection (composite index of legal restrictions on the ability of .770 –.462 .092 .054
employers to lay off or fire regular employees) – data for 2003 (OECD 2004: 117).
Employment law rigidities, index by Porta et al. (2004). .411 –.704 .040 .073
Social security laws index to measure decommodification by the pension, health, .348 .384 –.025 .128
regulation
and unemployment system (Porta et al. 2004: 1362f.).
Labour market
Employment laws index to measure the protection of employed workers .845 –.278 .068 .279
217
(Porta et al. 2004: 1362f.).
(continued)
Appendix B Continued
218
1 2 3 4
Gini Index – data for 1993–2000 (World Bank 2006). –.661 –.491 –.192 .456
Percentage share of income or consumption for poorest 20 per cent of .587 .346 .304 –.540
population – data for 1993–2000 (World Bank 2006).
Income
Percentage share of income or consumption for richest 20 per cent of –.634 –.450 –.248 .406
distribution
population – data for 1993–2000 (World Bank 2006).
Employer centralization – average for 1979–95 (Martin and Swank 2001: 899). .740 .034 .060 –.339
Employer coordination – average for 1979–95 (Martin and Swank 2001: 899). .704 .355 .401 –.077
Enterprise cooperation – average for 1979–95 (Martin and Swank 2001: 899). .637 –.088 .436 –.390
Central union power density – average for 1979–95 (Martin and Swank 2001: 899). .649 .202 –.276 .036
Collective bargaining – Average for 1979–95 (Martin and Swank 2001: 899). .751 .146 –.109 .002
Trade union members as percentage of total workforce – data for 2000 .645 .459 –.331 –.106
(OECD 2004: 145).
Coverage of collective wage bargaining – data for 2000 (OECD 2004: 145). .839 –.280 –.171 .194
Index of dominant level of bargaining: 0 company or individual; .25 .601 –.389 –.036 .042
company/plant; .5 industry; .75 industry and central; 1 central. Data for 2004 –
(EIRO data for 2005, URL: http://www.eiro.eurofound.eu.int/ and OECD Employment
Industrial relations
Outlook 2004: 151).
Co-determination index, made of indices: employees appoint board members; .814 –.083 –.009 .163
employee participation required by law; works council’s decision rights –
Data for 2002 (OECD Survey 2003, URL: www.oecd.org/dataoecd/58/27/
21755678.pdf, page 47f.).
Collective relations laws index to measure the power of labour unions .561 –.429 .265 –.284
over working conditions (Porta et al. 2004: 1362f.).
Average net income replacement rate for fully insured single worker with APW .598 .333 .444 .320
wage and couple with two children and one APW wage – data for 1999
(Allan and Scruggs 2004: 500)
Deduction (per cent) of wage due to taxes and social security contributions for .848 –.057 –.046 .166
married couple, two children, one average wage – data for 2001 (OECD 2006,
URL: http://www.nationmaster.com/graph-T/tax_tot_tax_wed_sin_wor
Deduction (per cent) of wage due to taxes and social security contributions for .835 –.071 –.039 .216
single worker, no children, average wage – data for 2001 (OECD 2006, URL:
http://www.nationmaster.com/graph-T/tax_tot_tax_wed_sin_wor
OECD summary measure of unemployment benefit as percentage of old wage – data .660 –.096 –.149 .356
for 2003 (OECD 2006, URL: http://www.oecd.org/dataoecd/25/31/34008592.xls).
Pension replacement rate – net mandatory pension systems replacement rates as .604 –.439 .353 .312
percentage of pre-retirement net-earning for men with average wage – data for 2003
(OECD 2005a).
Taxes (including all social security contributions) on the average production worker .814 –.081 –.035 .214
Welfare state
(average wage, no children) – data for 2004 (OECD 2006).
Total tax revenue (no social security payments) as share of GDP – data for 2003 .835 .166 –.383 .174
(OECD 2006, URL: http://puck.sourceoecd.org/vl=7267625/cl=26/nw=1/rpsv/
factbook/data/09-03-01-t01.xls).
Public social expenditure (cash benefits, provision of goods and services and tax .785 .216 .135 .163
breaks with social purposes) – data for 2001 (OECD 2006, URL: http://thesius.
sourceoecd.org/vl=30620998/cl=14/nw=1/rpsv/factbook/data/09-02-01-t01.xls).
Share of means-tested social transfers on all social transfers in per cent – data for second –.569 .080 –.368 .017
half of 1990s (Eurostat data for 2003 and Gough et al. 1997 for non-European countries).
Private health care expenditure as share of total health care expenditure – –.398 –.059 .296 .679
data for 2003 (OECD 2005c, URL: http://dx.doi.org/10.1787/530538806724).
219
(continued)
Appendix B Continued
220
1 2 3 4
Total government expenditure as share of GDP – data for 2004 .790 .236 .113 .005
(CIA World Factbook 2005, https://www.cia.gov/library/publications/the-world-
factbook/, www.cia.gov).
Index of protection for investors (ownership disclosure, measures that reduce –.396 –.077 –.697 –.300
expropriation, and disclosures to help investors) – data for 2005 (World Bank 2006).
Deposit money bank credits to the private sector as a share of GDP – average –.162 –.064 .888 –.077
for 1980–95 (Levine 2000: 43).
Overall size of the stock market in US dollars as a percentage of GDP – data for –.376 .421 .502 .408
Financial system
2004, 2000 for Belgium (World Bank 2006)
Value of domestic equities listed on domestic exchanges divided by GDP – –.667 .446 .351 –.098
Average for 1980–95 (Levine 2000: 43).
Years in which chief executive and largest party in congress have left or .159 .724 .048 .363
centre political orientation from 1928–95 (Porta et al. 2004: 1362f.).
power
Political
Index
Asian capitalism, 39, 43, 45 Corporatism, 15–16, 156

Australia, 13–15, 114–15 Culture, 72–4, 76–7, 87, 93–6, 102–4,
Austria, 122–4 117, 173–5
Banks Decommodification, 6, 33, 75–81

see coordinated market economy Dendrogram, 35–6
Belgium, 126–8 Denmark, 72, 148–51, 155–7
Beneficial constraints, 81 Dualization
Beveridge, 6, 53, 100 see conservatively coordinated
Bismarck, 6, 120, 122, 128, 133, 136, capitalism
140, 145, 153, 156
Britain, 98–102 England
see United Kingdom
Calvinism, 96–9, 124, 130, 143–57 Employer associations, 74–80
Canada, 108–10 Employment protection, 21–3, 73–4
Capital markets, 21–2 Employment rates, 10–12, 77–8
Catholicism, 96, 106, 118–42 also see gender
Civil law, 174 Esping-Andersen
Cleavages, 118–52, 171, 173 see welfare regime research
Cluster analysis, 34–7 Exceptions to typologies, 51–4
Collective bargaining, 20–4, 64–5,
72–86 Financial sector, 17
Common law, 174 Finland, 151–3, 155–7
Complementarity, 4, 23–30, 47, France, 38–40, 48, 53, 110, 128–30
63–91, 117, 172–4 Franco, 135–6
Concertation, 132–48 Functionalism, 83
Conservatism, 7, 14–15, 31–5, 65–7,
106–7, 119–38, 159, 164–70 Gender
Conservatively coordinated capitalism, female elite decision makers,
3–4, 6–7, 14–15, 28, 33, 35–6, 62, 166–7
118–20, 140–2, 169–73 female employment, 165
Convergence, 28–30 Germany, 38, 40, 47, 120–2
Coordinated market economy, 16–19 Ghent unemployment insurance,
also see conservatively coordinated 84–5, 126–7
capitalism Gradual change, 27
also see social democratically Great Britain
coordinated capitalism see United Kingdom
Coordination, 16–25, 31–2, 39, 53, Group-based coordination
62, 68, 70, 72–90, 96–7, 163–4 see conservatively coordinated
coordination versus solidarity, capitalism
168–71
Corporate Governance, 18, 47, 68–70, Ideal type, 14–15, 26, 55
83–4 Incoherent capitalisms, 50–7
221
222 Index
Incremental, innovation, 16 Prisoner’s dilemma, 17, 81

Individualism, 98–9, 103–4 Proportional representation
Industrial Relations, 16–17, 64–8 see voting systems
Innovation, 16 Protestantism
Ireland, 106–8 see Calvinism
Italy, 48, 133–5 also see Lutheranism
Japan, 13–14, 35–6, 39, 47, 56, 72, Radical innovation, 16

140–2 Regulation theory, 39–42
Religion
Liberal capitalism, 3–4, 6, 28, 33, see Calvinism
35–6, 45, 62, 91–9, 115–18, also see Catholicism
168–71, 173 also see Lutheranism
Liberalization, 14–15, 28–30, 43, 85–7, Retirement, 81–4, 45–6
97–8, 100, 107, 121, 168–71, 173 early retirement, 81–4
varieties of, 168–71 pension replacement rate, 81–4
Liberal market economy, 17–19 decommodification of, 83
also see liberal capitalism Retrenchment, 85–7
Lutheranism, 96, 144–5, 155–6 also see liberalization
Majoritarian representation Scandinavia

see voting systems see social democratically
Mediterranean welfare and capitalism, coordinated capitalism
13–15, 35–6, 43–5, 54, 72, 133–40 Service sector, 11
Methodological nationalism, 28–30 Skills
general and specific, 70–4
Netherlands, 130–3 Social democratically coordinated
New Zealand, 13, 111–14 capitalism, 3–4, 6–9, 14–15, 28,
Norway, 153–7 33, 35–6, 45, 62, 132, 143–5,
155–7, 169–73
Outsider-insider policy Social democratic welfare state
see conservatively coordinated see welfare regimes
capitalism Social embeddedness, 47
also see Mediterranean capitalism Social inequality, 9–10
Social pacts, 152, 154
Pension Spain, 135–8
see retirement Stock market, 21–2, 68–70
Pillarization Sweden, 145–8, 155–7
see Belgium Switzerland, 35–6, 54, 124–6
also see Netherlands
Policy style, 4, 33, 46, 53–4, 62, 93–6, Trade union strength, 64–8
104–5, 113, 115–17, 130, 134–5, Typologies, 37–50
137–8, 148, 155–6, 170–3 integrated in one scheme, 59–62
Political systems
see voting systems Unemployment insurance, 8–9, 80,
Portugal, 138–40 70–1, 78–9
Power Ressources, 5–6, 84–7 also see Ghent unemployment
Principal factor analyis insurance
see cluster analysis United Kingdom, 38, 48–9, 98–102
Index 223
United States, 49, 102–6 Wage earner welfare state, 13,

Universalism, 6–7, 145, 148–9, 154 112
Wage replacement rates, 8–9
Varieties of capitalism, 15–25 Welfare regime research, 5–13
weaknesses of, 25–30, 42 weaknesses of, 13–15, 28–30,
Voting systems, 85, 174 42
Welfare state research,
Wage bargaining, 20–4, 78 see welfare regime research
Wage deductions, 7–8 Works councils, 18–20

(Work and Welfare in Europe) Martin Schröder (Auth.) - Integrating Varieties of Capitalism and Welfare State Research - A Unified Typology of Capitalisms (2013, Palgrave Macmillan UK)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Work and Welfare in Europe) Martin Schröder (Auth.) - Integrating Varieties of Capitalism and Welfare State Research - A Unified Typology of Capitalisms (2013, Palgrave Macmillan UK)

Uploaded by

Copyright:

Available Formats

Work and Welfare in Europe

Series Editors: Denis Bouget, University of Nantes, France; Jochen Clasen,

Sonja Drobnic and Ana M. Guillén Rodriguez

Colette Fagan, Maria González Menèndez and Silvia Gómez Ansón

Neil Fraser, Rodolfo Gutierrez and Ramon Pena-Cassas

Patricia Frericks and Robert Maier

Paolo Graziano, Sophie Jacquot and Bruno Palier

Karl Hinrichs and Matteo Jessoula

Colin Lindsay and Donald Houston

Ive Marx and Kenneth Nelson

Birgit Pfau-Effinger and Tine Rostgaard

Work and Welfare in Europe

List of Figures and Tables viii

4.7 Varieties of capitalism and power resources 84

1.1 Taxes and social security contributions for different

4.5 Pension replacement rate of prior wage and coordination 82

1.1 Three welfare regimes in comparison 6

I would like to thank Aaron Boesenecker, Jason Beckfield, Peter Hall,

welfare typology analyses what benefits and duties individuals have in

welfare state and the production system to a common principle, that

production systems coincide. Trying to explain this, Chapter 4 shows

This chapter introduces Esping-Andersen’s welfare and Hall and

1.1 Welfare state research

Drawing on Esping-Andersen (1999: 33), one can define welfare state

Table 1.1 Three welfare regimes in comparison

Liberal Social Conservative

Role of – Family Marginal Marginal Central

Source: Esping-Andersen (1999: 85).

social equality (Esping-Andersen 1999: 78f.). Their welfare state is

One-earner married couple at 100% of average earnings, 2 children

Figure 1.2 Benefit entitlements in the first year of unemployment in 2007

Figure 1.5 Service sector employment as share of total employment (civilians

states do not damage service sector employment as much, because they

type.7 This provoked criticism again Esping-Andersen’s typology. The

1.1.1 Weaknesses of Esping-Andersen’s welfare regime typology

There are also good theoretical reasons to use Esping-Andersen’s three-

or they circumvent markets, be it by stressing social solidarity, be it

1.2 Varieties of capitalism

In contrast to previous conceptions of capitalist diversity – which

Hall and Soskice draw heavily on the neo-corporatism literature,

an incentive to cooperate, associations, networks or the state can diffuse

easy to imagine how an internet start-up, a young biotechnology firm

Table 1.2 Characteristics of coordinated market economies and liberal market

Coordinated market Liberal market economies

Industrial Cooperative Antagonistic

councils are a disadvantage in liberal countries, where consultation

Figure 1.7 Rights of works councils in 2010

Hall and Soskice describe a similar clustering between liberal and

Again, there is a fairly clear clustering in liberal and non-liberal

countries, which should allow companies to be flexible in the hiring

Figure 1.10 OECD indicator of employment protection in 2008

in all respects. What is important though is that the different parts of

should tend to develop complementary practices in other spheres as

Coverage of collective bargaining

between the indicators. Note however, that Ireland is excluded, since

1.2.2 The coordination index

1.2.3 Weaknesses of Hall and Soskice’s varieties of capitalism

behaviour, if occurring en masse, forms a production system and it then

most of the views Streeck attributes to that approach reflect a crude

1.3 Why it makes sense to combine nations into regimes

Criticism that is pertinent already if addressed to varieties of capital-

Overall, academics have a paradoxical, one might even say cannibalis-

2.1 Verifying diversity empirically