Three World of Welfare Capitalism or More? A State-of-the-Art

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/258136528
Three world of welfare capitalism or more? A State-of-the-Art-

Report
Article in Journal of European Social Policy · May 2002

DOI: 10.1177/0952872002012002114
CITATIONS READS
1,093 6,692
2 authors:
Wil Arts John Gelissen

Tilburg University Tilburg University
107 PUBLICATIONS 2,537 CITATIONS 56 PUBLICATIONS 3,430 CITATIONS
SEE PROFILE SEE PROFILE
All content following this page was uploaded by John Gelissen on 13 January 2014.
The user has requested enhancement of the downloaded file.

Journal of European Social Policy
http://esp.sagepub.com
Three worlds of welfare capitalism or more? A state-of-the-art report

WIL Arts and John Gelissen
Journal of European Social Policy 2002; 12; 137
DOI: 10.1177/0952872002012002114
The online version of this article can be found at:

http://esp.sagepub.com/cgi/content/abstract/12/2/137
Published by:
http://www.sagepublications.com
Additional services and information for Journal of European Social Policy can be found at:
Email Alerts: http://esp.sagepub.com/cgi/alerts
Subscriptions: http://esp.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.co.uk/journalsPermissions.nav
Citations http://esp.sagepub.com/cgi/content/refs/12/2/137
Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Article
Three worlds of welfare capitalism or more?

A state-of-the-art report
Wil Arts and John Gelissen*, Tilburg University, The Netherlands
Summary This paper surveys the debate Résumé Nous présentons un état des lieux
regarding Esping-Andersen’s typology of des débats qui ont entouré la typologie des
welfare states and reviews the modified or Etats providences proposée par Esping-
alternative typologies ensuing from this Andersen ainsi que des typologies modifiées ou
debate. We confine ourselves to the classifica- alternatives qui ont été présentées par la suite.
tions which have been developed by Esping- Nous nous limiterons aux classifications qui
Andersen’s critics in order to cope with the ont été proposées par les critiques du travail
following alleged shortcomings of his typol- d’Esping Andersen qui visaient à dépasser les
ogy: (1) the misspecification of the Mediter- «prétendues» limitations de sa typologie.1)
ranean welfare states as immature Continental une mauvaise spécification des Etats Providence
ones; (2) the labelling of the Antipodean méditerranéens comme des Etats providences
welfare states as belonging to the ‘liberal’ continentaux inachevés 2) la labelisation des
regime type; (3) a neglect of the gender- Etats providences des Antipodes comme
dimension in social policy. We reconstruct appartenant au régime de type «libéral» 3) la
several typologies of welfare states in order to non prise en compte de l’effet «genre» dans les
establish, first, whether real welfare states are politiques sociales. Nous avons reconstruit dif-
quite similar to others or whether they are férentes typologies d’Etats Providences afin
rather unique specimens, and, second, whether d’établir, tout d’abord si la réalité des Etats
there are three ideal-typical worlds of welfare Providences est unique ou si certains sont fort
capitalism or more. We conclude that real semblables et ensuite s’il existe trois idéal type
welfare states are hardly ever pure types and d’Etats providences ou davantage?
are usually hybrid cases; and that the issue of Nous sommes arrivé à la conclusion que:
ideal-typical welfare states cannot be satisfac- premièrement les Etats providences réels sont
torily answered given the lack of formal theo- rarement des cas purs mais constituent bien
rizing and the still inconclusive outcomes of plus des hybridations. En deuxième lieu, on ne
comparative research. In spite of this conclu- peut répondre de manière satisfaisante à la ques-
sion there is plenty of reason to continue to tion de savoir le nombre d’idéal types d’Etats
work on and with the original or modified Providences du fait de l’absence de théorisa-
typologies. tion adéquate et parce que les résultats des
recherches comparatives ne permettent pas de
Key words comparative social policy conclure dans un sens ou un autre. Troisième-
analysis, typology, welfare state regimes, ment, en dépit de cette conclusion, il existe de
worlds of welfare capitalism multiples raisons pour continuer à travailler sur
et à partir des typologies originales ou modifiées.
* Author to whom correspondence should be sent: John Gelissen, Department of Sociology, Tilburg
University, PO Box 90153, 5000 LE Tilburg, The Netherlands.
Journal of European Social Policy 0958-9287 (200205)12:2 Copyright © 2002 SAGE Publications, London, Thousand
Oaks and New Delhi, Vol 12 (2): 137–158; 023114

138 Arts and Gelissen
Introduction to theoretical and methodological shortcom-

ings (cf. Lessenich and Ostner, 1998). The
When Esping-Andersen (1990) published his more hostile critics feel that typologies as such
Three Worlds of Welfare Capitalism it was have no explanatory power and, therefore, his
received with applause (Offe, 1991; Cnaan, scheme does not contribute to proper theoriz-
1992; Hicks, 1991; Kohl, 1993). After a few ing about what is happening with and within
years, the book even became a modern classic. welfare states (cf. Baldwin, 1996).
Nevertheless, Esping-Andersen (1993; 1994; In this paper, we give an overview of the
1996; 1997; 1999) has been forced time after debate about Esping-Andersen’s typology
time by both his critics and his adherents to during the last decade and review the state of
elaborate on his original arguments. The tenet the art of typifying welfare states at the turn
of Esping-Andersen’s treatise of the welfare of the millennium. The pivotal questions are
state was that, for a long time in both the the- as follows: How and why has the welfare state
oretical and empirical literature, too little developed? How and why do national welfare
attention had been given to cross-national dif- systems differ from one another – or are they
ferences in welfare state structures.1 In spite of similar to each other? Do welfare states
this putative lack of attention he could – theo- cluster into different regime types and, if so,
retically and empirically – stand ‘on the shoul- how and why? It is not our intention to raise
ders of giants’. Theoretically, the work of new questions. Our objective is to settle
Marshall (1950; 1963; 1965; 1981) and affairs, for the time being, by giving an
Titmuss (1958; 1974) laid the foundations for overview of what we think is the gist of the
Esping-Andersen’s typology (Boje, 1996: 19). discussion, by weighting the most important
Empirically, he could profit from the compar- arguments and taking stock of the modelling
ative research by, among others, Wilensky business. To achieve this goal, we elaborate on
(1975), Flora and Heidenheimer (1981), earlier overviews by Abrahamson (1999), Arts
Mommsen (1981) and Flora (1983; 1986). He and Gelissen (1999) and Kohl (2000). First,
argued that we are entering better times we review the debate with respect to Esping-
because ‘the most intensive activity of welfare Andersen’s typology and modified versions of
state theorising at the moment has become it. Second, we try to establish whether there
identifying diversity, specifying welfare state are three ideal-typical worlds of welfare capi-
typologies’ (Esping-Andersen, 1994: 715). talism or more. Third, we outline which
Research has to follow theory’s lead because authors identify which national states as
‘only comparative empirical research will ade- belonging to a particular type of welfare state.
quately disclose the fundamental properties This means that we will not only look at ideal
that unite or divide modern welfare states’ worlds of welfare capitalism, but also at real
(Esping-Andersen, 1990: 3). In his ‘seminal’ ones.
book he suited the action to the word by con-
structing today’s best-known and most fre-
quently used typology of welfare states, and Ideal-types
by empirically testing whether distinct welfare
states that resemble his ideal-types can be Do typologies based on ideal-types have theo-
observed. retical and empirical value as Esping-Andersen
For accomplishing this feat, he not only assumes? The conclusion emerging from the
received wide critical acclaim and constructive philosophy of science literature is clear: not if
critism, but also some negative criticism. The ideal-types are goals in themselves, but only if
more amicable critics argue that his typology they are the means to a goal; namely, the rep-
has merits but is neither exhaustive nor exclu- resentation of a reality, which cannot (yet) be
sive and therefore needs revising. Others refer described using laws (Klant, 1984). This
Journal of European Social Policy 2002 12 (2)

Three worlds of welfare capitalism or more? 139
means that typologies are only fruitful to an see the forest rather than the myriad of unique
empirical science that is still in its infancy. In trees. However, he warns of the danger that
contrast, a mature empirical science empha- the resulting forest may bear little resemblance
sizes the construction of theories and not the to reality.
formulation of typologies. There are good Looking for a more detailed answer to the
reasons to argue that the comparative macro- question of the theoretical status of ideal-types
sociology of welfare states is still in statu and typologies, we must return to the locus
nascendi. Therefore the formulation of typolo- classicus: Weber’s methodological essays.
gies could be useful. Before answering the Weber (1949; 1968 [1922]) deals with two
question of whether welfare state typologies different kinds of ideal-types, individualistic
based on ideal-types are not only useful but and holistic ones (cf. Hempel, 1965 [1952];
also have explanatory value, we first consider Rogers, 1969; Watkins, 1969 [1953]). Esping-
what Esping-Andersen himself says about the Andersen’s ideal-types of welfare state regimes
methodological status of his typology. are holistic. They propose to give a bird’s eye
He addresses the question of whether the view of the broad characteristics of a social or
welfare state is merely the sum total of a historical situation. The ideality of such types
nation’s social policy repertoire, or whether it lies in their simplification and aloofness from
is an institutional force above and beyond a detail. They emphasize the ‘essential’ features
given policy array. His answer is straightfor- of a situation considered as a whole. By com-
ward: the welfare state cannot be regarded as paring an impure welfare state with an ideal-
the sum total of social policies, it is more than typical one – both considered as a whole – the
a numerical cumulation of discrete programmes deviations of the former from the latter are
(Esping-Andersen, 1994: 712). Therefore, in thrown into relief. It is the simultaneous
the relation between state and economy, he knowledge of both the ideal-type and the real-
defines welfare state regimes as a complex of type that enables holistic ideal-types to be
legal and organizational features that are sys- used ‘as conceptual instruments for compari-
tematically interwoven. Esping-Andersen son with and measurement of reality’
(1990: 3, 26, 32) boldly suggests that when (Watkins, 1969 [1953]: 458–9).
we focus on the principles embedded in From a logical point of view (von Kempski,
welfare states, variations are not linearly dis- 1972), the general term ‘welfare state’ is a
tributed around a common denominator. They label for a certain class of democratic indus-
are clustered around three highly diverse trial capitalist societies, characterized by
regime-types, each organized according to its certain properties (i.e. social citizenship or the
own discrete logic of organization, stratifica- fact that more or less extensive welfare provi-
tion, and societal integration. Therefore, we sions are legally provided, or, in still other
can identify three models, or ideal-types of words, the fact that the state plays a principal
welfare states: conservative, liberal and social- part in the welfare mix alongside the market,
democratic. These ideal-types owe their civil society, and the family). Welfare states
origins to different historical forces and they have seldom been established as a result of big
follow qualitatively different developmental plans or big fights, but mostly as results of
trajectories. Contrary to the ideal world of complex processes and successive steps of
welfare states, the real world is likely to social and political engineering in the history
exhibit hybrid forms. There are no one-dimen- of democratic industrial capitalist societies. In
sional nations in the sense of a pure case. spite of the largely incremental emergence of
Today, every country presents a system mix. welfare states, Esping-Andersen is of the
Esping-Andersen (1997: 171) argues that opinion that this class of societies does not
despite this it is fruitful to construct ideal- consist of a great number of unique cases,
types for the sake of economy: to be able to but that they cluster together in three distinct

subclasses. Each of the three types he identi- tested later on in this paper. The second con-
fies with a deep tradition in political mobiliza- dition is that the typology is a means to an
tion and political philosophy (conservatism, end – explanation – and not an end in itself.
liberalism and socialism respectively) which Esping-Andersen uses the regime types not
then link to particular features of contempo- only as dependent variables but also as inde-
rary social policy (and broader political pendent variables to explain cross-national
economy) configurations. variations in dependent variables such as
To determine the characteristics of these social behaviour and social attitudes. He also
subclasses without going back in history, two uses the typology to postulate and explain the
indicators are crucial: decommodification and occurrence of positive feedback loops. Accord-
stratification. Together they define a two- ing to him, existing institutional welfare
dimensional property-space. Although real arrangements heavily determine, maybe even
welfare states are most of the time not unique, over-determine, national trajectories (Esping-
they certainly are never completely similar. Andersen, 1999: 4). This suggests path-
This means that they are almost always dependency. This is because the stratification
impure types. The consequence is that outcomes of particular arrangements shape
although they cluster together in three sub- class coalitions, which tend to reproduce the
classes it is not always easy to classify all cases original institutional matrix and welfare out-
unambiguously. In practice it is possible that comes. This means that policies provide incen-
different judges assign a particular welfare tives that encourage individuals and groups to
state to different subclasses. Interjudge valid- act in ways that lock in a particular path of
ity can be accomplished by assessing which policy development (cf. Pierson, 1993).
ideal-type – the extreme limiting cases in this A third condition for accepting typologizing
ordering – they approximate best. By compar- as a legitimate endeavour is that theory con-
ing impure real welfare states to ideal-types, struction on welfare states is still in an early
the deviations of the ‘impure’ real-types are stage. This is also true. Boje (1996: 18) argues
contrasted with the ‘purity’ of the ideal-type. that the present ‘state of the art’ in most
This simultaneous recognizability of both the welfare state research is characterized by a
ideal and the real-type make it possible to use lack of theory. Few theoretical alternatives are
holistic ideal-types as conceptual instruments available. At the most, one can think of
for comparison and for the empirical determi- marxist explanations (see, for example
nation of reality (Watkins, 1969 [1953]). But Therborn, 1995) or de Swaan’s (1988) synthe-
accurate ordering is not enough. After all, sis of rational-choice theories and figurational
ideal-types are also instruments for providing sociology. The construction of ideal-types can
explanations. If we use them to satisfy this be fruitful under the condition that these will
objective, they should not only be understood eventually lead to theories. We will return to
as a conceptual system but also as a system this issue in the final section.
of theoretical statements. These should
encompass testable, general hypotheses or, at
least, provide a framework for interpretation Three worlds of ‘welfare
(Hempel, 1965 [1952]). capitalism’ . . .
To what conclusion does the preceding
reflection lead? In reply to Esping-Andersen’s The central explanatory questions Esping-
fiercest critics, we can say that their criticism Andersen (1990: 4, 105) asks are: Why is the
is unjust if certain conditions are met. The world composed of three qualitatively differ-
first condition is that the typology is a valid ent welfare-state logics? Why do nations crys-
and reliable instrument for classifying welfare tallize into distinct regime-clusters? These
states. Whether this condition holds will be questions demand a theoretical answer. Since

he is of the opinion that the existing theoreti- accordance with this theoretical expectation,
cal models of the welfare state are inadequate, he succeeds in empirically identifying three
reconceptualization and retheorization are closely paralleled models – ideal-types – of
necessary. In answering these questions he regime-types on both the stratification and the
starts with the orienting statement that history decommodification dimension. There appears
and politics matter. Or, more specifically: ‘The to be a clear coincidence of high decommodi-
historical characteristics of states, especially fication and strong universalism in the
the history of political class coalitions as the Scandinavian, social-democratically influenced
most decisive cause of welfare-state variations, welfare states. There is an equally clear coinci-
have played a determinate role in forging the dence of low decommodification and strong
emergence of their welfare-statism’ (1990: 1). individualistic self-reliance in the liberal
What are the historical and political forces Anglo–Saxon nations. Finally, the Continental
behind the regime differences? According to European countries group closely together as
Esping-Andersen (1990: 29), three interacting corporatist and etatist, and are also modestly
factors are significant: the nature of class decommodifying (Esping-Andersen, 1990: 77).
mobilization (especially of the working class), In spite of anomalies such as the
class-political action structures, and the his- Netherlands and Switzerland, the overall
torical legacy of regime institutionalization. picture is convincing, at least at first glance.
The provisional answer to his central ques- This empirical success permits a more exten-
tions is therefore: If you look at the history of sive description of these three worlds of
so-called welfare states you find three ideal- welfare capitalism. First, there is the liberal
typical trajectories, a liberal, a conservative type of welfare capitalism, which embodies
and a social-democratic one. Fortunately, one individualism and the primacy of the market.
does not have to go back in history, however, The operation of the market is encouraged by
in order to typify ‘real’ welfare states. We can the state, either actively – subsidizing private
characterize them, as we have mentioned welfare schemes – or passively by keeping
before, by looking at their positions on two (often means tested) social benefits to a
fundamental dimensions of welfare statism: modest level for the demonstrably needy.
There is little redistribution of incomes within
1. The degree of decommodification, i.e. the this type of welfare state and the realm of
degree to which a (social) service is ren- social rights is rather limited. This welfare
dered as a matter of right, and the degree regime is characterized by a low level of
to which a person can maintain a liveli- decommodification. The operation of the
hood without reliance on the market. liberal principle of stratification leads to divi-
2. The kind of social stratification and soli- sion in the population: on the one hand, a
darities, i.e. which social stratification minority of low-income state dependants and,
system is promoted by social policy and on the other hand, a majority of people able
does the welfare state build narrow or to afford private social insurance plans. In this
broad solidarities? type of welfare state, women are encouraged
to participate in the labour force, particularly
What are the characteristics of the three dis- in the service sector.
tinct regime-types to which the historical Second, there is a world of conservative-
forces lead? To answer this question, Esping- corporatist welfare states, which is typified by
Andersen (1990: 73) argues that although the a moderate level of decommodification. This
before-mentioned dimensions are conceptually regime type is shaped by the twin historical
independent, according to his ‘theory’ he legacy of Catholic social policy,2 on the one
would expect that there is sufficient covaria- side, and corporatism and etatism on the
tion for distinct regime clusters to emerge. In other side. This blend had three important

consequences in terms of stratification. In the typologies or added one or more types to

first place, the direct influence of the state is existing classifications for greater empirical
restricted to the provision of income mainte- refinement. From this vast array of welfare
nance benefits related to occupational status. state typologies we have selected six classifica-
This means that the sphere of solidarity tions, which we think draw attention to inter-
remains quite narrow and corporatist. esting characteristics of welfare states not
Moreover, labour market participation by directly included in Esping-Andersen’s classifi-
married women is strongly discouraged, cation. All these typologies and their main
because corporatist regimes – influenced by characteristics are summarized in Table 1.
the Church – are committed to the preserva- These alternative classifications relate to
tion of traditional family structures. Another three important criticisms of Esping-
important characteristic of the conservative Andersen’s classification (for these and other
regime type is the principle of subsidiarity: the points of critique see Schmidt, 1998; Gough,
state will only interfere when the family’s 2000b).3 First, the misspecification of the
capacity to service its members is exhausted Mediterranean welfare states; second, labelling
(Esping-Andersen, 1990: 27). the Antipodean welfare states as belonging to
Finally, Esping-Andersen recognizes a the ‘liberal’ welfare state regime; and finally,
social-democratic world of welfare capitalism. the neglect of the gender-dimension in social
Here, the level of decommodification is high, policy. In the following sections, we will
and the social-democratic principle of stratifi- discuss these criticisms in more detail and
cation is directed towards achieving a system present some of the alternative classifications
of generous universal and highly distributive developed by his critics.
benefits not dependent on any individual con-
tributions. In contrast to the liberal type of
welfare states, ‘this model crowds out the The Mediterranean
market and, consequently, constructs an
essentially universal solidarity in favour of the One important criticism of Esping-Andersen’s
welfare state’ (Esping-Andersen, 1990: 28). classification is that he did not systematically
Social policy within this type of welfare state include the Mediterranean countries.
is aimed at a maximization of capacities for Specifically, in The Three Worlds of Welfare
individual independence. Women in particular Capitalism Italy belongs – according to him –
– regardless of whether they have children or to the family of the corporatist welfare state
not – are encouraged to participate in the regimes, whereas Spain, Portugal and Greece
labour market, especially in the public sector. are not covered by his typology. Although he
Countries that belong to this type of welfare admits that these countries have some impor-
state regime are generally dedicated to full tant characteristics in common – i.e. a
employment. Only by making sure that as Catholic imprint (with the exception of
many people as possible have a job is it possi- Greece) and a strong familialism (Esping-
ble to maintain such a high-level solidaristic Andersen, 1997: 180) – he seems to include
welfare system. them in the continental/corporatist model. His
omission of a systematic treatment of the
Mediterranean has brought about a lively
. . . Or more? debate about the existence of a ‘Southern’ or
‘Latin Rim’ model of social policy. For
In the Introduction, we indicated the tremen- example, Katrougalos (1996) supports
dous impact of Esping-Andersen’s work on Esping-Andersen’s position by arguing that
comparative social policy analysis. Since then, the Mediterranean countries ‘do not form a
several authors have developed alternative distinct group but rather a subcategory, a

Table 1 An overview of typologies of welfare states
Types of welfare states and their characteristics Indicators/dimensions
Esping-Andersen (1990) 1. Liberal: Low level of decommodification; market-differentiation of welfare • Decommodification
2. Conservative: Moderate level of decommodification; social benefits mainly dependent on • Stratification
former contributions and status
3. Social-democratic: High level of decommodification; universal benefits and high degree of
benefit equality
Leibfried (1992) 1. Anglo–Saxon (Residual): Right to income transfers; welfare state as compensator of last resort • Poverty, social insurance
and tight enforcer of work in the market place and poverty policy

2. Bismarck (Institutional): Right to social security; welfare state as compensator of first resort and
employer of last resort

3. Scandinavian (Modern): Right to work for everyone; universalism; welfare state as employer of
first resort and compensator of last resort
4. Latin Rim (Rudimentary): Right to work and welfare proclaimed; welfare state as a semi-
institutionalized promise
Castles & Mitchell (1993) 1. Liberal: Low social spending and no adoption of equalizing instruments in social policy • Welfare expenditure
2. Conservative: High social expenditures, but little adoption of equalizing instruments in social • Benefit equality
policy
3. Non-Right Hegemony: High social expenditure and use of highly equalizing instruments in • Taxes
social policy
4. Radical: Achievement of equality in pre-tax, pre-transfer income (adoption of equalizing
instruments in social policy), but little social spending
Siaroff (1994) 1. Protestant Liberal: Minimal family welfare, yet relatively egalitarian gender situation in the • Family welfare orientation
labour market; family benefits are paid to the mother, but are rather inadequate
2. Advanced Christian-democratic: No strong incentives for women to work, but strong incentives • Female work desirability
to stay at home
3. Protestant Social-democratic: True work–welfare choice for women; family benefits are high and • Extent of family benefits
always paid to the mother; importance of Protestantism being paid to women
4. Late Female Mobilization: Absence of Protestantism; family benefits are usually paid to the
father; universal female suffrage is relatively new
Ferrera (1996) 1. Anglo–Saxon: Fairly high welfare state cover; social assistance with a means test; mixed system • Rules of access (eligibility)
of financing; highly integrated organizational framework entirely managed by a public
administration
continued over
143
144
Table 1 continued
Types of welfare states and their characteristics Indicators/dimensions
2. Bismarck: strong link between work position (and/or family state) and social entitlements; • Benefit formulae
benefits proportional to income; financing through contributions; reasonably substantial social
assistance benefits; insurance schemes mainly governed by unions and employer organizations
3. Scandinavian: social protection as a citizenship right; universal coverage; relatively generous • Financing regulations
fixed benefits for various social risks; financing mainly through fiscal revenues; strong
organizational integration
4. Southern: fragmented system of income guarantees linked to work position; generous benefits • Organizational–managerial
without articulated net of minimum social protection; health care as a right of citizenship; arrangements
particularism in payments of cash benefits and financing; financing through contributions and
Arts and Gelissen

fiscal revenues
Bonoli (1997) 1. British: Low percentage of social expenditure financed through contributions (Beveridge); low • Bismarck and Beveridge
social expenditure as a percentage of GDP model
2. Continental: High percentage of social expenditure financed through contributions (Bismarck);
high social expenditure as a percentage of GDP
3. Nordic: Low percentage of social expenditure financed through contributions (Beveridge); high • Quantity of welfare state
social expenditure as a percentage of GDP expenditure
4. Southern: High percentage of social expenditure financed through contributions (Bismarck);
low social expenditure as a percentage of GDP
Korpi & Palme (1998) 1. Basic Security: Entitlements based on citizenship or contributions; application of the flat-rate
benefit principle
2. Corporatist: Entitlements based on occupational category and labour force participation; use of • Bases of entitlement
the earnings-related benefit principle
3. Encompassing: Entitlement based on citizenship and labour force participation; use of the flat- • Benefit principle
rate and earnings-related benefit principle
4. Targeted: Eligibility based on proved need; use of the minimum benefit principle • Governance of social
5. Voluntary State Subsidized: Eligibility based on membership or contributions; application of the insurance programme
flat-rate or earnings-related principle
variant of the Continental model. They are risks of life. The right to social protection is
merely underdeveloped species of the attributed on the basis of citizenship. The
Continental model, welfare states in their Anglo–Saxon family of welfare states is also
infancy, with the main common characteristics characterized by a highly inclusive social secu-
being the immaturity of the social protection rity coverage, but only in the area of health
systems and some similar social and family care can one speak of fully universal risk cov-
structures’ (Katrougalos, 1996: 43). However, erage. Also flat-rate benefits and means testing
according to other commentators (Leibfried, play an important role. In the third group of
1992; Ferrera, 1996; Bonoli, 1997; Trifiletti, countries, the relationship between social
1999) it seems logical to see the South security entitlements, a person’s labour market
European countries as a separate cluster. status and role within the family (breadwinner
They have developed classifications of or not) is still clearly visible. Contributions
European welfare states which try to show the play an important role in financing the
existence of a separate ‘southern model’ of various schemes. Almost everybody has social
social policy. insurance coverage through their own or
Leibfried (1992) distinguishes four social derived rights. Finally, the social protection
policy or poverty regimes within the countries systems of Southern countries are highly frag-
of the European Community: the Scandina- mented and, although there is no articulated
vian welfare states, the ‘Bismarck’ countries, net of minimum social protection, some bene-
the Anglo–Saxon countries and the Latin Rim fits levels are very generous (such as old age
countries. These policy regimes are based on pensions). Moreover, in these countries health
different policy models – modern, institu- care is institutionalized as a right of citizen-
tional, residual and rudimentary – in which ship. However, in general, there is relatively
social citizenship has developed in different little state intervention in the welfare sphere.
and sometimes incomplete ways. Within these Another important feature is the high level of
policy regimes, welfare state institutions have particularism with regard to cash benefits and
a different function in combating poverty. financing, expressed in high levels of clien-
However, it is particularly important that telism. The most important features of each
Leibfried adds a fourth category – the ‘Latin type are summarized in Table 1.
Rim’ countries – to Esping-Andersen’s original Bonoli (1997) uses the Mediterranean coun-
classification. He emphasizes as an important tries – among others – to develop the final
characteristic of these countries the lack of an classification we wish to discuss in this
articulated social minimum and a right to section. He is especially critical of the decom-
welfare. modification approach. According to him, it
Ferrera (1996) also argues explicitly for the does not allow one to discriminate effectively
inclusion of a ‘Southern model’ of social between the Bismarckian and the Beveridgean
policy (1996: 4–7). He concentrates on four approaches to social policy. As an alternative,
dimensions of social security systems: the rules he combines two approaches to the classifica-
of access (eligibility rules), the conditions tion of welfare states. One concentrates on
under which benefits are granted, the regula- the ‘how much’ dimension (emphasized in the
tions to finance social protection and, finally, Anglo–Saxon literature) and the other on the
the organizational-managerial arrangements ‘how’ dimension of social policy (emphasized
to administrate the various social security in the Continental-European or French tradi-
schemes. Based on these dimensions, he makes tion). As an empirical indicator of the first
a distinction between the Scandinavian, dimension, he uses social expenditure as a pro-
Anglo–Saxon, Bismarckian and Southern portion of GDP, and of the second dimension
countries. The Scandinavian countries are the percentage of social expenditure financed
characterized by universal coverage for the through contributions. These indicators lead

him to identify four types of countries: the In a discussion of their study, Hill (1996)
British countries, the Continental European points out that Castles and Mitchell’s critique
countries, the Nordic countries and the of Esping-Andersen’s work essentially follows
Southern countries, thus giving credit to the two lines. In the first place, they draw atten-
proposal of a ‘Southern model’. tion to the fact that political activity from the
Upon examining the combined arguments Left may have been introduced into those
of Leibfried, Ferrera and Bonoli, as presented countries in the achievement of equality in
in Table 1, it appears that a strong similarity pre-tax, pre-transfer income rather than in the
exists among their first three types and those pursuit of equalization through social policy.
of Esping-Andersen. However, all three Second, they argue – again about Australia
authors add a fourth – Mediterranean – type but also with relevance to the United King-
of welfare state regime to the original Esping- dom – that the Esping-Andersen approach dis-
Andersenian classification. Using empirical regards the potential for income-related
evidence, they argue that this is a prototype benefits to make an effective contribution to
rather than a subcategory of the continental/ redistribution. Australian income maintenance
corporatist model. is almost entirely means-tested. It uses an
approach that neither concentrates on a liberal-
type redistribution to the very poor, nor
The Antipodes resembles the more universal social-democratic
and hierarchical solidaristic conservative
Esping-Andersen also discusses the Antipodean ideal-types highlighted in Esping-Andersen’s
countries (i.e. Australia and New Zealand) as study (Hill, 1996: 46). This is the reason why
representatives of the liberal welfare state Castles and Mitchell develop an alternative,
regime. This is because of their marginal com- four-way classification of welfare states:
mitment to public welfare and strong reliance Liberal, Conservative, Non-Right Hegemony
on means testing. However, according to and Radical. This utilizes the level of welfare
Castles (1998), Australia and New Zealand expenditure (i.e. household transfers as a per-
have a more particular and a more inclusive centage of GDP); average benefit equality; and
approach to social protection than the stan- income and profit taxes as a percentage of
dard liberal form. Thresholds are set at com- GDP.
paratively high levels, so that a large part of Other evidence for the exceptional position
the population receives some means-tested of the Antipodean countries, specifically
benefits. The result is that the Antipodes Australia, is found when countries are classi-
exhibit the world’s most comprehensive systems fied according to the typology developed by
of means-tested income support benefits. Korpi and Palme (1998). This is based on
Redistribution has been traditionally pursued institutional characteristics of welfare states.
through wage controls and employment secu- They try to investigate the causal factors
rity rather than social programmes. Income which influence the institutional aspects of the
guarantees, realized by using market regulation welfare state on the one hand, and the effects
thus play an important role in the institutional of institutions on the formation of interests,
set-up of these welfare states. It therefore seems preferences and identities – as well as on the
that the Antipodean countries represent a sep- degree of poverty and inequality in a society –
arate social policy model. It led Castles and on the other hand. They argue that institu-
Mitchell (1993) to question whether ‘social tional structures can be expected to reflect the
spending is the only route to greater income role of conflicts among interest groups, while
redistribution’, implying that there may be other they are also likely to form important frame-
ways than income maintenance by which states works for the definitions of interests and iden-
may mitigate the effects of market forces. tities among citizens. They can thereby be

expected to influence coalition formation, the state and the market provide welfare, but
which is significant for income redistribution also families. A further omission is that there
and poverty. As the basis of their classifica- is no serious treatment of the degree to which
tion, Korpi and Palme take the institutional women are excluded from or included in the
structures of two social programmes – old age labour market.4 Instead of employing the all-
pensions and sickness cash benefits – which or-nothing words ‘inclusion’ and ‘exclusion’
they consider to lie at the heart of the welfare to gender differences, it seems sensible to
state. The institutional structures of the two stress the importance of partial citizenship
programmes are classified according to three (Bulmer and Rees, 1996: 275). Women
aspects: the bases of entitlements, the princi- obtained full civil and political rights a consid-
ples applied to determine benefit levels (to erable time ago, but with regard to social
what extent social insurance should replace rights, women are still discriminated against,
lost income), and the governance of a social sometimes formally, and nearly always infor-
insurance programme (whether or not repre- mally because of different labour market posi-
sentatives of employers and employees partici- tions, linked to different gender roles.
pate in the governing of a programme). Based According to many feminist authors, it is the
on these three aspects, they discriminate sexual division of paid and unpaid work –
among five different ideal-types of institu- especially care and domestic labour – that
tional structures: the targeted (empirically needs incorporating in the typology (Lewis,
exemplified by the Australian case), voluntary 1992; O’Connor, 1993; Orloff, 1993;
state subsidized, corporatist, basic security Sainsbury, 1996; O’Connor et al., 1999).
and encompassing model. In Table 1, these With respect to another issue, social care,
ideal-types and their most important features Daly and Lewis (2000: 289) argue that differ-
are delineated. Again, the Esping-Andersen ent styles of social policy have incorporated
model stands. However, a number of countries the key element of social care differently. They
are no longer considered to belong to a sub- identify certain tendencies concerning care in
category of his three prototypes, but to a new specific welfare states. For example, the
prototype. Scandinavian countries form a distinct group
in that they have strongly institutionalized
care for both the elderly and children. In the
Gender, familialism and late female Mediterranean welfare states, care tends to be
mobilization privatized to the family whereas, in Germany,
it is seen as most appropriately a function of
By explicitly incorporating gender, several voluntary service providers. In France, a
authors have tried to reconceptualize the strong distinction is made between care for
dimensions of welfare state variation. Sub- children and for the elderly, with a strong col-
jecting the mainstream welfare state typolo- lective sector in the former and little voluntary
gies to an analysis of the differential places of involvement. Another form is found in the
men and women within welfare states would, Beveridge-oriented welfare states – Great
according to them, produce valuable insights. Britain and Ireland – where a strong distinc-
This does not mean, however, that the charac- tion is also made between caring for children
teristics used to construct the typologies are and caring for (elderly) adults. In the former –
exhaustive (Sainsbury, 1996: 41). Gender as opposed to the latter – little collectivization
analysis suggests that there are whole areas of has taken place. Although they do not really
social policy that Esping-Anderson simply classify welfare states into actual clusters,
misses. What seems to be particularly lacking Daly and Lewis make a strong case for using
is a systematic discussion of the family’s place social care as a critical dimension for
in the provision of welfare and care. Not only analysing variations.

As far as the gender gap in earnings is con- of welfare capitalism as defined by Esping-
cerned, Gornick and Jacobs (1998) found that Andersen. For example, Bonoli’s Continental
Esping-Andersen’s regime-types do capture type is very much like Esping-Andersen’s
important distinctions among contemporary Conservative type; in both types contributions
welfare states. Their results showed that play a rather important role. Equally, Castles
the size of the public sector, the extent of and Mitchell’s Non-Right Hegemony type
the public-sector earnings premium and the shows a large amount of congruence with
impact of the public sector on gender differen- Esping-Andersen’s Social-democratic type,
tials in wages all varied more across regimes because of the high degree of universalism and
than within them. In this way, they showed equalization in social policy. We could go on,
the fruitfulness of emphasizing the gender per- but we would like to raise another issue.
spective in Esping-Andersen’s classification of One may wonder whether, if the relation-
welfare states. Moreover, Trifiletti (1999) ship among the different typologies is as
incorporated a gender perspective into Esping- strong as we assume, this close correspon-
Andersen’s classification by showing that a dence of types will also be apparent in the
systematic relationship exists between the actual clustering of countries. Although not
level of decommodification and whether the every classification developed by these authors
state treats women as wives and mothers or as covers the same nations, there is a rather large
workers. The latter is also an important overlap which makes it possible to answer this
dimension identified by Lewis (1989). question. For that purpose, Table 2 shows the
Finally, Siaroff (1994) also argues that the extent to which the ideal-types – constructed
existing literature does not pay enough atten- by Esping-Andersen’s critics – coincide with
tion to how gender inequality is embedded in his own ideal-types. We then added the ideal-
social policy and welfare states. In order to types, proposed by these critics, placing
arrive at a more gender-sensitive typology of related ideal-types, when possible, under one
welfare state regimes, he examines a variety of heading. This results in five – instead of the
indicators of gender equality and inequality in original three – worlds of welfare capitalism
work and welfare. He compares the work– and answers our original question. Next, in
welfare choice of men and women (i.e. Table 2 we arranged the real-types according
whether to partake in the welfare state or to to the different ideal-types, thereby following
engage in paid labour) across countries. This the suggestions of the different authors.
allows him to distinguish among a Protestant It appears that, even when one uses differ-
social-democratic, a Protestant liberal, an ent indicators to classify welfare states, some
Advanced Christian-democratic and a Late countries emerge as standard examples,
Female Mobilization welfare state regime. approximating certain ideal-types. The United
Although the labels suggest otherwise, this States is, according to everyone’s classifica-
typology also shows a strong overlap with the tion, the prototype of a welfare state which
Esping-Andersenian classification. Only the can best be denoted as liberal (with or without
latter type – the Late Female Mobilization the suffix: Protestant, Anglo–Saxon or basic
welfare state regime – is an addition, which security). Germany approaches the
resembles the previously distinguished Bismarckian/Continental/conservative ideal-
Mediterranean type of welfare states. type and Sweden approximates the social-
democratic ideal-type (Scandinavian/Nordic).
However, consensus seems to end here. For
Ideal and real-types example, according to some, Italy can best be
assigned to the second, corporatist/continen-
In Table 1, we ordered the types discussed tal/conservative type, but belongs, according
above broadly in accordance with the worlds to others, along with Greece, Spain and

Table 2 Classification of countries according to seven typologies
Type
I II III IV V
Esping-Andersen Liberal Conservative Social-democratic
(Decommodification)
• Australia • Italy • Austria
• Canada • Japan • Belgium
• United States • France • Netherlands
• New Zealand • Germany • Denmark
• Ireland • Finland • Norway

• United Kingdom • Switzerland • Sweden
Leibfried Anglo–Saxon Bismarck Scandinavian Latin Rim

• United States • Germany • Sweden • Spain
• Australia • Austria • Norway • Portugal
• New Zealand • Finland • Greece
• United Kingdom • Denmark • Italy
• France
Castles & Mitchell Liberal Conservative Non-Right Hegemony Radical
• Ireland • West-Germany • Belgium • Australia
• Japan • Italy • Denmark • New Zealand
• Switzerland • Netherlands • Norway • United Kingdom
• United States • Sweden
Siaroff Protestant Liberal Advanced Christian-democratic Protestant Social-democratic Late Female Mobilization
• Australia • Austria • Denmark • Greece

• Canada • Belgium • Finland • Ireland
• New Zealand • France • Norway • Italy
• United Kingdom • West-Germany • Sweden • Japan
• United States • Luxembourg • Portugal
• Netherlands • Spain
• Switzerland
Ferrera (Europe only) Anglo–Saxon Bismarckian Scandinavian Southern
• United Kingdom • Germany • Sweden • Italy
• Ireland • France • Denmark • Spain
• Belgium • Norway • Portugal
• Netherlands • Finland • Greece
• Luxembourg
149
• Austria
• Switzerland continued over
150
Table 2 continued
Type
I II III IV V
Bonoli (Europe only) British Continental Nordic Southern
• United Kingdom • Netherlands • Sweden • Italy
• Ireland • France • Finland • Switzerland
Arts and Gelissen

• Belgium • Norway • Spain
• Germany • Denmark • Greece
• Luxembourg • Portugal
Korpi & Palme Basic Security Corporatist Encompassing Targeted
• Canada • Austria • Finland • Australia
• Denmark • Belgium • Norway
• Netherlands • France • Sweden
• New Zealand • Germany
• Switzerland • Italy
• Ireland • Japan
• United Kingdom
• United States
Note: Underlined countries indicate a prototype.

Portugal to a distinctive Mediterranean type. welfare states along the dimensions of degree
The same holds for Australia which may of decommodification and the modes of strati-
either be classified as liberal or is the proto- fication, three qualitatively different clusters
type of a separate, radical welfare state. will appear. Alongside the more fundamental
Nevertheless, as far as these countries and criticism of his three-way classification – that
types are concerned, consensus is stronger Esping-Andersen employs faulty criteria to
than was initially assumed. One must, demarcate a regime – the empirical fit of his
however, recognize that discussions are mainly three-way classification has also been ques-
concerned with whether certain types of tioned. Several authors have tested the good-
welfare states are either separate categories or ness-of-fit of the three-way regime typology.
are subgroups of certain main types. In the following, we discuss their findings,
Hybrid cases are a bigger problem. The which are presented in Table 3.
Netherlands and Switzerland are clear exam- In an effort to evaluate the possible extent
ples of this. If we take, for example, a closer to which quantitative techniques – OLS
look at the Dutch case, we see that Esping- regression and cluster-analysis – suggest the
Andersen (1990) originally assigned the same conclusions as alternative qualitative
Netherlands to the social-democratic type, approaches – such as ‘BOOLEAN’ compara-
whereas Korpi and Palme see it as liberally tive analysis – Kangas (1994) found some
oriented; the basic security type. However, support for the existence of Esping-Andersen’s
most authors place the Netherlands in the different welfare state regimes. Specifically,
second category of corporatist/continental/ cluster-analyses of data on characteristics of
conservative welfare states. This is also the health insurance schemes in OECD countries
choice of Visser and Hemerijck (1997), in 1950 and 1985 corroborated his conjec-
perhaps the foremost specialists on the Dutch tures. However, the results also showed the
welfare state. Curiously enough, this is done existence of two subgroups within the group
using Esping-Andersen’s work as a constant, of liberal welfare states, which largely
positive reference. If we have another look at accorded with the classification of Castles and
Esping-Andersen’s work, this is not as surpris- Mitchell (1993).
ing as one would expect. It is true that the Ragin (1994) also tested Esping-Andersen’s
Netherlands is rated relatively high on social- claim of a three-world classification. By apply-
democratic characteristics, but not exception- ing a combination of cluster-analysis and
ally low on liberal and conservative ‘BOOLEAN’ comparative analysis to charac-
characteristics. Recently, Esping-Andersen has teristics of pension systems, he determined
called the Netherlands the ‘Dutch enigma’ which, if any, of Esping-Andersen’s three
because of its Janus-faced welfare regime worlds of welfare capitalism each country
(1999: 88). The Netherlands is indeed more a fitted best. His cluster analysis suggested the
hybrid case than a prototype of a specific existence of a social-democratic cluster, a cor-
ideal-type. If one attaches more importance to poratist cluster and, finally, a rather large
certain attributes than to others – and adds ‘spare’ cluster, which accommodates cases that
other characteristics or substitutes previous do not conform to Esping-Andersen’s three
ones – then it is easy to arrive at different clas- worlds. On the basis of his findings, Ragin
sifications. concludes that the three-worlds scheme does
not capture existing diversity as adequately as
one would wish.
Empirical robustness of the three-way Shalev (1996) applied factor analysis to 14
classification pension policy indicators collected by Esping-
Andersen, to test for the presence of liberal,
Esping-Andersen claims that if we rate real social-democratic and corporatist regime-

152
Table 3 Empirical robustness of the three-worlds typology

Number of clusters and cluster assignment Method of analysis
Kangas (1994) 1. Liberal: United States, Canada Cluster analysis

2. Conservative: Austria, Germany, Italy, Japan, the Netherlands
3. Social-democratic: Denmark, Finland, Norway, Sweden
4. Radical: Australia, Ireland, New Zealand, United Kingdom
Ragin (1994) 1. Liberal: Australia, Canada, Switzerland, United States BOOLEAN comparative analysis
2. Corporatist: Austria, Belgium, Finland, France, Italy
3. Social-democratic: Denmark, Norway, Sweden
4. Undefined: Germany, Ireland, Japan, the Netherlands, New Zealand,
Arts and Gelissen

United Kingdom
Shalev (1996) 1. Liberal: United States, Canada, Switzerland, Japan Factor analysis
2. Conservative: Italy, France, Belgium, Austria, Ireland
3. Social-democratic: Sweden, Norway, Denmark, Finland
4. Undefined: Germany, the Netherlands, United Kingdom, Australia,
New Zealand
Obinger & Wagschal (1998) 1. Liberal: United States, Canada, Japan, Switzerland Cluster analysis
2. European: Belgium, Germany, Finland, Ireland, United Kingdom,
the Netherlands
3. Conservative: France, Italy, Austria
4. Social-democratic: Denmark, Norway, Sweden
5. Radical: Australia, New Zealand
Wildeboer Schut et al. (2001) 1. Liberal: United States, Canada, Australia, United Kingdom Principal component analysis
2. Conservative: France, Germany, Belgium
3. Social-democratic: Sweden, Denmark, Norway
4. Undefined: the Netherlands
types. This factor analysis showed that the embody any particular ideal-type (Goodin et
intercorrelations among these social policy al., 1999: 56). Third, if one looks at other
indicators were dependent on two dimensions. social programmes than the ones applied by
The first factor measured the level of social- Esping-Andersen, it becomes clear that they
democratic features, whereas the second do not conform so easily – if at all – to his
dimension measured corporatist features of welfare regime patterns (Gough, 2000a: 4).
welfare states. Based on the assignment of
factor scores to individual nations, Shalev
concluded that his findings were in close cor- Conclusion and discussion
respondence with Esping-Andersen’s charac-
terizations of the three welfare state regimes. Before we reach our conclusions, let us
He admitted, however, that some countries are examine how Esping-Andersen himself has
difficult to classify. reacted to the various attempts to amend his
Using cluster-analysis, Obinger and Wagschal typology. The problem is that after consider-
(1998) tested Esping-Andersen’s classification able discussion it seems impossible for him to
of welfare state regimes using the stratifica- make up his mind once and for all. Initially,
tion-criterion. After a detailed re-analysis of Esping-Andersen (1997) reacted, for example,
Esping-Andersen’s original data on stratifica- positively to Castles and Mitchell’s proposal
tion, they concluded that these data are best to add a fourth type – a radical welfare state
described by five regime-clusters. In addition regime – to his typology. He recognized that
to Esping-Andersen’s conservative, liberal and the residual character and the matter of a
social-democratic types, they distinguish a means test are just one side of the coin of the
radical and a hybrid European cluster. Antipodean welfare states. However, he felt
The most recent attempt to empirically cor- that a powerfully institutionalized collection
roborate Esping-Andersen’s classification has of welfare guarantees, which operate through
been undertaken by Wildeboer Schut et al. the market itself, could not be neglected. Later
(2001). This study examined the actual simi- on, however, he argued that the passage of
larities and differences among the welfare time is pushing Australia, Great Britain and
state regimes of the countries originally New Zealand towards what appears to be
included in Esping-Andersen’s classification. prototypical liberalism (Esping-Andersen,
For these countries, 58 characteristics of the 1999). At first he also partially supported the
labour market, tax regime and social protec- proposal to add a separate Mediterranean
tion system at the beginning of the 1990s were type to his typology (Esping-Andersen, 1996:
collected. These were submitted to a non- 66; 1997: 171). He acknowledged the – some-
linear principal component-analysis. The times generous – benefits which are guaran-
results largely confirmed the three-regime teed by certain arrangements, the near absence
typology of Esping-Andersen. of social services and, especially, the Catholic
Summing up, Esping-Andersen’s original imprint and high level of familialism. From
three-worlds typology neither passes the the feminist critics he learned not so much the
empirical tests with flying colours, nor dis- overarching salience of gender as the analyti-
mally fails them. The conclusion is, first, that cal power that a re-examination of the family
his typology has at least some heuristic and can yield. Recently he argued that the acid test
descriptive value, but also that a case can be of a distinct Mediterranean model depends on
made for extending the number of welfare whether families are the relevant focus of
state regimes to four, or even five. Second, social aid, and whether families will fail just
these analyses show that a significant number as markets and states can fail (Esping-
of welfare states must be considered hybrid Andersen, 1999: 90).
cases: no particular case can ever perfectly All in all, Esping-Andersen is very reluctant

to add more regime-clusters to his original theory is difficult to determine. The work of
three. Against the benefits of greater refine- some of the other authors we discussed in this
ment, more nuance and more precision, he overview has a strong empiricist flavour.
weighs the argument of analytical parsimony, However, if we are searching for an underly-
stressing that ‘the peculiarities of these cases ing theoretical notion, it can be found in the
are variations within a distinct overall logic, rather general statement that similar causes
not a wholly different logic per se’ (Esping- have similar consequences. Considering the
Andersen, 1999: 90). labels these authors have put to the prototypes
The answer to the question of whether they distinguish, which are predominantly
Esping-Andersen’s three-type or a derivative geographical and ideological in nature, the
or alternative four, or five-type typology is most important causes are seen to be the pres-
preferable depends, however, not only on par- sure of functional exigencies and the diffusion
simony and verisimilitude. It also depends on of innovations (Goldthorpe, 2000: 54). The
whether these typologies lead to a theoreti- first factor could be translated into a ‘chal-
cally more satisfying and empirically more lenge response’ hypothesis. The challenges
fruitful comparative analysis of welfare state produced by the force of similar circumstances
regimes. As far as theory construction is con- (characteristics of pre-industrial social struc-
cerned, Baldwin (1996: 29) has argued that tures, political institutions, degree of homo-
when asking about typologies, whether of geneity of population, culture, problem
welfare states or anything else, we must ask perceptions and preferences) lead to compara-
not just what but also why. Esping-Andersen’s ble welfare state regimes (responses) (Kuhnle
tentative answer to the question of why three and Alestalo, 2000: 7). The second factor
different welfare state regime types emerged could be put in terms of learning effects in
has been sketched earlier on in this paper. policy making. New ideas, new solutions are
Different welfare regimes are shaped by differ- often a product of a diffusion process. They
ent class coalitions within a context of inher- hit political systems and societies at different
ited institutions. This answer is embedded in a points in ‘developmental time’. As far as this
power-resources mobilization paradigm. The latter factor is concerned, Boje (1996: 15)
tentative answer to the question of why argues that the fact that most welfare states
regime shifts are scarce is that a national state are confronted with huge social problems has
cannot easily escape its historical inheritance. necessitated politicians finding alternative
Institutional inertia is one factor why different procedures, which may solve these problems
welfare state regime types persist, and path more efficiently. Politicians have come to
dependency is another (Kohl, 2000: 125; realize that much may be learned from other
Kuhnle and Alestalo, 2000: 9). Korpi and welfare states.
Palme – and some feminist authors – work in Castles (1993; 1998) too underscores the
the same power-resources mobilization tradi- importance of both factors. He argues that it
tion as Esping-Andersen. It would be worth- is likely that policy similarities and differences
while to develop a theoretical reconstruction among welfare states can be attributed to both
of the different contributions of this paradigm the force of circumstances and to diffusion. As
(for an initial impetus to such an endeavour, far as the latter factor is concerned we can dis-
see Schmidt, 1998: 215–28). Only then could tinguish the institutional arrangements and
the explanatory value of the typology become culture of prototypical welfare states and their
apparent. transmission and diffusion to other countries.
Whether there is, within the welfare model- Regarding the former factor, we can observe
ling business, an alternative available to the immediate impact of economic, political
Esping-Andersen’s power-resource mobiliza- and social variables identified in the contem-
tion cum institutional inertia/path dependency porary public policy literature. Whether these

very general ‘challenge response’ and ‘diffu- their intended social stratification, a tautologi-
sion’ hypotheses will be further developed cal element easily sneaks into the explana-
remains to be seen. For the moment, we can tions. Positive exceptions are Goodin et al.
conclude that, given the empiricist nature of (1999) and Korpi and Palme (1998). Using
the work of the authors who provided alterna- panel-data, Goodin et al. (1999) show that
tive typologies, there should be hardly any welfare state regimes do not only have
objection – for the time being – to the incor- intended results, but also generate unintended
poration of their findings into a power– consequences. As intended and expected, the
resources mobilization paradigm. social-democratic regime succeeds best in real-
Finally, we arrive at the empirical fruitful- izing its fundamental value: minimizing inequal-
ness of the typology. In his overview, ity. But this regime is also at least as good in
Abrahamson (1999) concludes that as an promoting the goals to which other regimes
organizing principle for comparative studies ostensibly attach most importance. Specifically,
of welfare states the typologies have proven to the social-democratic regime also does very
be a very robust and convincing tool. Within well in reducing poverty – a goal which is pri-
the power–resources mobilization paradigm oritized by the liberal welfare state regime –
(Korpi, 1983; Esping-Andersen and Korpi, and in promoting stability and social integra-
1984; Esping-Andersen, 1990) it has been tion, which is the home ground of the corpo-
proposed that the nature of the welfare state ratist welfare state regime. Korpi and Palme
regime would decisively influence support for (1998) find that institutional differences lead
certain forms of social policy. A type that is to a paradox of redistribution: the more bene-
characterized by universalism would generate fits are targeted at the poor and the more the
the strongest support, whereas arrangements creation of equality through equal public trans-
which apply only to minorities would not fers to all is a matter of priority, the less poverty
succeed in winning the support of majorities. and equality will be reduced. Thus, institu-
Tests of this hypothesis (Papadakis and Bean, tional arrangements characteristic of certain
1993; Peillon, 1996; Gelissen, 2000; Gevers et welfare state regimes not only have unintended
al., 2000) have shown some empirical support, consequences, but even perverse effects.
but the evidence is not really encouraging. All in all, these conclusions provide suffi-
More encouraging were the results of an effort cient impetus to continue the work concerning
(Gundelach, 1994) to explain cross-national the resulting welfare state typology. A better
differences in values with respect to welfare formulation of the theory on which it is based
and care using the Esping-Andersenian welfare deserves priority. Only then can predictions be
state regimes. Also, Svallfors’s (1997) and Arts logically – instead of impressionistically –
and Gelissen’s (2001) tests of the hypothesis deduced from theory. Only then is a strict test
that different welfare state regimes matter for of the theory possible and only then will
people’s attitudes towards income-redistribu- the heuristic and explanatory value of the
tion, were strongly endorsed. What especially typology become apparent.
matters to us here is that Svallfors distributive
justice and solidarity had included not only Acknowledgements
Esping-Andersen’s regime-types, but also
other types and Arts and Gelissen. We would like to thank Ian Gough and four
It is more difficult to draw a conclusion anonymous reviewers for valuable comments.
concerning the influence that welfare state
regimes have on social behaviour and their Notes
effects. Much of this research has a bearing on
1 This is an exaggeration. Schmidt (1998),
the distributive effects of welfare state regimes.
Abrahamson (1999) and Arts and Gelissen
Because they are often described in terms of

(1999) have convincingly argued that the evi- Bulmer, M. and Rees, A. M. (1996) ‘Conclusion:
dence is otherwise. Citizenship in the Twenty-first Century’, in M.
2 The importance of Catholicism is emphasized by Bulmer and A. M. Rees (eds) Citizenship Today.
van Kersbergen (1995) in his discussion includ- The Contemporary Relevance of T. H. Marshall,
ing Christian-democratic nations such as pp. 269–84. London: UCL Press.
Germany, Italy and the Netherlands in main- Castles, F. G. (1993) ‘Introduction’, in F. G. Castles
stream welfare state typologies. (ed.) Families of Nations: Patterns of Public
3 For reasons of conciseness we refrain from the Policy in Western Democracies. Aldershot:
debate regarding Esping-Andersen’s classifica- Dartmouth.
tion of Japan as a liberal welfare state. For a Castles, F. G. (1998) Comparative Public Policy:
reaction to this critique see Esping-Andersen Patterns of Post-war Transformation. Cheltenham:
(1997; 1999). Becker (1996) and Goodman and Edward Elgar.
Peng (1996) are even of the opinion that Japan Castles, F. G. and Mitchell, D. (1993) ‘Worlds of
belongs to a sixth prototype of welfare state Welfare and Families of Nations’, in F. G. Castles
regimes, the so-called East-Asian welfare states. (ed.) Families of Nations: Patterns of Public
We acknowledge the importance of these argu- Policy in Western Democracies. Aldershot:
ments, but cannot engage with them here Dartmouth Publishing Company.
(Gough, 2000a; 2000b). Cnaan, R. A. (1992) ‘Three Worlds of Welfare
4 Gornick and Jacobs (1998: 691) point out that Capitalism’, Acta Sociologica 35 (1): 69–71.
Esping-Andersen himself argues that each Daly, M. and Lewis, J. (2000) ‘The Concept of Social
regime-type is associated with women’s employ- Care and the Analysis of Contemporary Welfare
ment levels. Specifically, he (Esping-Andersen, States’, British Journal of Sociology 51 (2): 281–98.
1990) expects that women’s employment rates de Swaan, A. (1988) In Care of the State: Health
will be highest in social-democratic countries, Care, Education and Welfare in Europe and
whereas in liberal welfare states, moderate levels the USA in the Modern Era. Cambridge: Polity
of female employment will be found. The lowest Press.
levels of women’s employment will be found in Esping-Andersen, G. (1990) The Three Worlds of
the conservative welfare states. Welfare Capitalism. Oxford: Polity Press.
Esping-Andersen, G. (1993) ‘The Comparative
References Macro-sociology of Welfare States’, in L. Moreno
(ed.) Social Exchange and Welfare Development,
pp. 123–36. Madrid: Consejo Superior de
Abrahamson, P. (1999) ‘The Welfare Modelling Investigaciones Científicas.
Business’, Social Policy & Administration 33 (4): Esping-Andersen, G. (1994) ‘Welfare States and the
394–415. Economy’, in N. J. Smelser and R. Swedberg
Arts, W. A. and Gelissen, J. (1999) ‘Verzorgings- (eds) The Handbook of Economic Sociology, pp.
staten in soorten: Op zoek naar ideaal- en reële 711–32. Princeton/New York: Princeton
typen’, Mens & Maatschappij 74 (2): 143–65. University Press/Russel Sage Foundation.
Arts, W. A. and Gelissen, J. (2001) ‘Welfare States, Esping-Andersen, G. (1996) ‘Welfare States without
Solidarity and Justice Principles: Does the Type Work: the Impasse of Labour Shedding and
Really Matter?’ Acta Sociologica 44 (4): Familialism in Continental European Social
283–300. Policy’, in G. Esping-Andersen (ed.) Welfare
Baldwin, P. (1996) ‘Can We Define a European States in Transition, pp. 66–87. London: Sage.
Welfare State Model?’, in B. Greve (ed.) Esping-Andersen, G. (1997) ‘Hybrid or Unique?
Comparative Welfare Systems: the Scandinavian The Japanese Welfare State between Europe and
Model in a Period of Change, pp. 29–44. America’, Journal of European Social Policy 7
London: Macmillan. (3): 179–89.
Becker, U. (1996) ‘Over de typologie van welvaarts- Esping-Andersen, G. (1999) Social Foundations of
bestellen: Esping-Andersens theorie in discussie’, Post-industrial Economies. Oxford: Oxford
Beleid & Maatschappij 23 (1): 19–30. University Press.
Boje, T. (1996) ‘Welfare State Models in Esping-Andersen, G. and Korpi, W. (1984) ‘Social
Comparative Research: Do the Models Describe Policy as Class Politics in Post-war Capitalism’,
the Reality?’, in B. Greve (ed.) Comparative in J. Goldthorpe (ed.) Order and Conflict in Con-
Welfare Systems: the Scandinavian Model in a temporary Capitalism. Oxford: Oxford University
Period of Change, pp. 13–27. London: Press.
Macmillan Press. Ferrera, M. (1996) ‘The “Southern” Model of
Bonoli, G. (1997) ‘Classifying Welfare States: a Welfare in Social Europe’, Journal of European
Two-dimension Approach’, Journal of Social Social Policy 6 (1): 17–37.
Policy 26 (3): 351–72.

Flora, P. (1983) The Growth of Mass Democracies Hicks, A. (1991) ‘The Three Worlds of Welfare
and Welfare States, Vol. 1. Frankfurt am Main: Capitalism’, Contemporary Sociology 20 (3):
Campus Verlag. 399–401.
Flora, P. (1986) Growth to Limits: the Western Hill, M. (1996) Social Policy: a Comparative Ana-
European Welfare States since World War II, 5 lysis. London: Prentice Hall/Harvester Wheatsheaf.
vols. Berlin: de Gruyter. Kangas, O. E. (1994) ‘The Politics of Social Security:
Flora, P. and Heidenheimer, A. J. (1981) The On Regressions, Qualitative Comparisons, and
Development of Welfare States in Europe and Cluster Analysis’, in T. Janoski and A. M. Hicks
America. New Brunswick, NJ: Transaction (eds) The Comparative Political Economy of the
Books. Welfare State, pp. 346–64. Cambridge:
Gelissen, J. (2000) ‘Popular Support for Cambridge UP.
Institutionalised Solidarity: a Comparison among Katrougalos, G. S. (1996) ‘The South European
European Welfare States’, International Journal Welfare Model: the Greek Welfare State in Search
of Social Welfare 9 (4): 285–300. of an Identity’, Journal of European Social Policy
Gevers, J., Gelissen, J., Arts, W. and Muffels, R. 6 (1): 39–60.
(2000) ‘Public Health Care in the Balance: Klant, J. J. (1984) The Rules of the Game: the
Exploring Popular Support for Health Care Logical Structure of Economic Theories.
Systems in the European Union’, International Cambridge: Cambridge University Press.
Journal of Social Welfare 9 (4): 301–21. Kohl, J. (1993) ‘Der Wohlfahrtsstaat in vergleichen-
Goldthorpe, J. H. (2000) On Sociology: Numbers, der Perspective’, Zeitschrift für Sozialreform 39:
Narratives, and the Integration of Research and 67–82.
Theory. Oxford: Oxford University Press. Kohl, J. (2000) ‘Der Sozialstaat: Die deutsche
Goodin, R. E., Heady, B., Muffels, R. and Dirven, Version des Wohlfahrtsstaates – Überlegungen zu
H-J. (1999) The Real Worlds of Welfare seiner typologischen Verortung’, in S. Leibfried
Capitalism. Cambridge: Cambridge University and U. Wagschal (eds) Der deutsche Sozialstaat:
Press. Bilanzen- Reformen-Perspectiven, pp. 115–52.
Goodman, R. and Peng, I. (1996) ‘The East Asian Frankfurt/New York: Campus.
Welfare States: Peripatetic Learning, Adaptive Korpi, W. (1983) The Democratic Class Struggle.
Change, and Nation-building’, in G. Esping- London: Routledge & Kegan Paul.
Andersen (ed.) Welfare States in Transition, pp. Korpi, W. and Palme, J. (1998) ‘The Paradox of
192–224. London: Sage. Redistribution and Strategies of Equality: Welfare
Gornick, J. C. and Jacobs, J. A. (1998) ‘Gender, the State Institutions, Inequality and Poverty in the
Welfare State, and Public Employment: a Western Countries’, American Sociological
Comparative Study of Seven Industrialized Review 63 (5): 661–87.
Countries’, American Sociological Review 63 (5): Kuhnle, S. and Alestalo, M. (2000) ‘Growth,
688–710. Adjustments and Survival of European Welfare
Gough, I. (2000a) ‘Welfare Regimes in East Asia States’, in S. Kuhnle (ed.) Survival of the
and Europe: Comparisons and Lessons’, retrieved European Welfare State, pp. 3–18. London/New
9 March 2001 from the World Wide Web: York: Routledge.
http://www.bath.ac.uk/Faculties/HumSocSci/ Leibfried, S. (1992) ‘Towards a European
IFIPA/GSP/. University of Bath: Institute for welfare state? On Integrating Poverty Regimes
International Policy Analysis. into the European Community’, in Z. Ferge
Gough, I. (2000b) ‘Welfare Regimes: on Adapting and J. E. Kolberg (eds) Social Policy in a Changing
the Framework to Developing Countries’, Europe. Frankfurt am Main: Campus Verlag.
retrieved 9 March 2001 from the World Wide Lessenich, S. and Ostner, I. (1998) Welten des
Web: http://www.bath.ac.uk/Faculties/HumSocSci/ Wolhlfahrtskapitalismus: Der Sozialstaat in ver-
IFIPA/GSP/. University of Bath: Institute for gleichender Perspektive. Frankfurt/Main: Campus
International Policy Analysis. Verlag.
Gundelach, P. (1994) ‘National Value Differences: Lewis, J. (1989) ‘Lone Parent Families: Politics and
Modernization or Institutionalization?’, Interna- Economics’, Journal of Social Policy 18 (4):
tional Journal of Comparative Sociology 35 595–600.
(1–2): 37–58. Lewis, J. (1992) ‘Gender and the Development of
Hempel, C. G. (1965) [1952] ‘Typological Methods Welfare Regimes’, Journal of European Social
in the Natural and the Social Sciences’, in C. G. Policy 2 (3): 159–73.
Hempel (ed.) Aspects of Scientific Explanation Marshall, T. H. (1950) Citizenship and Social Class
and Other Essays in the Philosophy of Science, and other Essays. Cambridge: Cambridge UP.
pp. 155–72. New York/London: The Free Marshall, T. H. (1963) Sociology at the Crossroads
Press/Collier-Macmillan. and other Essays. London: Heinemann.

Marshall, T. H. (1965) Social Policy. London: Policy? Occupational Welfare and the Welfare
Hutchinson. State in America, Scandinavia and Japan.
Marshall, T. H. (1981) The Right to Welfare and London: Macmillan.
other Essays. London: Heinemann. Siaroff, A. (1994) ‘Work, Welfare and Gender
Mommsen, W. J. (1981) The Emergence of the Equality: a New Typology’, in D. Sainsbury (ed.)
Welfare State in Britain and Germany. London: Gendering Welfare States, pp. 82–100. London:
Croom Helm. Sage.
Obinger, H. and Wagschal, U. (1998) ‘Das Svallfors, S. (1997) ‘Worlds of Welfare and
Stratifizierungskonzept in der Clusteranalytischen Attitudes to Redistribution: a Comparison of
Überprüfung’, in S. Lessenich and I. Ostner (eds) Eight Western Nations’, European Sociological
Welten des Wohlfahrtskapitalismus: Der Review 13 (3): 283–304.
Sozialstaat in vergleichender Perspektive, pp. Therborn, G. (1995) European Modernity and
109–35. Frankfurt/Main: Campus Verlag. Beyond. The Trajectory of European Societies
O’Connor, J. S. (1993) ‘Gender, Class and 1945–2000. London: Sage.
Citizenship in the Comparative Analysis of Titmuss, R. (1958) Essays on the Welfare State.
Welfare State Regimes: Theoretical and London: Allen & Unwin.
Methodological Issues’, The British Journal of Titmuss, R. (1974) Social Policy. London: Allen &
Sociology 44 (3): 501–18. Unwin.
O’Connor, J. S., Orloff, A. and Shaver, S. (1999) Trifiletti, R. (1999) ‘Southern European Welfare
States, Markets, Families. Gender, Liberalism and Regimes and the Worsening Position of Women’,
Social Policy in Australia, Canada, Great Britain Journal of European Social Policy 9 (1): 49–64.
and the United States. Cambridge: Cambridge van Kersbergen, K. (1995) Social Capitalism: a
UP. Study of Christian Democracy and the Welfare
Offe, C. (1991) ‘The Three Worlds of Welfare State. London: Routledge.
Capitalism’, American Journal of Sociology 96 Visser, J. and Hemerijck, A. C. (1997) A Dutch
(6): 1,555–7. Miracle: Job Growth, Welfare Reform and
Orloff, A. S. (1993) ‘Gender and the Social Rights Corporatism in the Netherlands. Amsterdam:
of Citizenship: a Comparative Analysis of Gender Amsterdam University Press.
Relations and Welfare States’, American von Kempski, J. (1972) ‘Zur Logik der
Sociological Review 58 (3): 303–28. Ordungsbegriffe, besonders in den Sozialwissen-
Papadakis, E. and Bean, C. (1993) ‘Popular Support schaften’, in H. Albert (ed.) Theorie und Realität.
for the Welfare State: a Comparison among Ausgewählte Aufsätze zur Wissenschaftslehre der
Institutional Regimes’, Journal of Public Policy Sozialwissenschaften, pp. 115–38. Tübingen: J.
13 (3): 227–54. C. B. Mohr (Paul Siebeck).
Peillon, M. (1996) ‘A Qualitative Comparative Watkins, J. W. N. (1969) [1953] ‘Methodological
Analysis of Welfare State Legitimacy’, Journal of Individualism and Non-Hempelian Ideal Types’,
European Social Policy 6 (3): 175–90. in L. J. Kimmerman (ed.) The Nature and Scope
Pierson, P. (1993) ‘When Effect becomes Cause: of Social Science: a Critical Anthology. New
Policy Feedback and Political Change’, World York: Appleton-Century-Crofts.
Politics 45 (4): 595–628. Weber, M. (1949) The Methodology of the Social
Ragin, C. (1994) ‘A Qualitative Comparative Sciences. Glencoe, IL: Free Press.
Analysis of Pension Systems’, in T. Janoski and A. Weber, M. (1968) [1922] Gesammelte Aufsätze zur
M. Hicks (eds) The Comparative Political Wissenschaftslehre. Tübingen: J. C. B. Mohr
Economy of the Welfare State, pp. 320–45. (Paul Siebeck).
Cambridge: Cambridge UP. Wildeboer Schut, J. M., Vrooman, J. C. and de Beer,
Rogers, R. E. (1969) Max Weber’s Ideal Type P. (2001) On Worlds of Welfare. Institutions and
Theory. New York: Philosophical Library. their Effects in Eleven Welfare States. The Hague:
Sainsbury, D. (1996) Gender, Equality and Welfare Social and Cultural Planning Office of the
States. Cambridge: Cambridge UP. Netherlands.
Schmidt, M. G. (1998) Sozialpolitik in Deutschland: Wilensky, H. L. (1975) The Welfare State and
Historische Entwicklung und internationaler Equality: Structural and Ideological Roots of
Vergleich. Opladen: Leske & Budrich. Public Expenditures. Berkeley/Los Angeles:
Shalev, M. (1996) The Privatization of Social University of California Press.
View publication stats

Three World of Welfare Capitalism or More? A State-of-the-Art

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

Three World of Welfare Capitalism or More? A State-of-the-Art

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Three world of welfare capitalism or more? A State-of-the-Art-

Article in Journal of European Social Policy · May 2002

Wil Arts John Gelissen

SEE PROFILE SEE PROFILE

The user has requested enhancement of the downloaded file.

Three worlds of welfare capitalism or more? A state-of-the-art report

The online version of this article can be found at:

Email Alerts: http://esp.sagepub.com/cgi/alerts

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Three worlds of welfare capitalism or more?

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Introduction to theoretical and methodological shortcom-

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

consequences in terms of stratification. In the typologies or added one or more types to

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Three worlds of welfare capitalism or more?

employer of last resort

Arts and Gelissen

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Three worlds of welfare capitalism or more?

Leibfried Anglo–Saxon Bismarck Scandinavian Latin Rim

• Australia • Austria • Denmark • Greece

Arts and Gelissen

Note: Underlined countries indicate a prototype.

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Table 3 Empirical robustness of the three-worlds typology

Kangas (1994) 1. Liberal: United States, Canada Cluster analysis

Arts and Gelissen

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

Journal of European Social Policy 2002 12 (2)

Downloaded from http://esp.sagepub.com at RITSUMEIKAN UNIV LIBRARY on April 12, 2009

View publication stats

You might also like