The Emergence of the East Asian Welfare State

Content

1. Introduction
1.1 Formulation of the Research Questions
1.2 Content

2. Methodological Part
2.1 Comments on Terminology and Definitions
2.2 Data Sources
2.3 The Comparative Method in General
2.4 Quantitative and Qualitative Comparisons of Welfare States
2.5 Selection of Cases

3. Theoretical Part
3.1 Functionalist Explanation Models: The Logic of Industrialism
3.2 Conflict-Theoretical Explanation Models
3.3 The Three Worlds of Welfare Capitalism by Esping-Andersen
3.4 Applicability of Esping-Andersen’s Approach to East Asia

4. Development of Welfare Systems in East Asia
4.1 Overview Historical Development
4.2 Development of the South Korean Welfare System
4.2.1 Health Insurance
4.2.2 Occupational Injury
4.2.3 The Pension Program
4.2.4 Unemployment Insurance
4.3 The Development of the Taiwanese Welfare System
4.3.1 Health Care
4.3.2 Occupational Injury
4.3.3 The Pension Program
4.3.4 Unemployment Insurance
4.3.5 Social Assistance
4.4 The Development the Welfare System in Hong Kong
4.5 The Development of the Singaporean Welfare System
4.6 Funding: Similarities and Differences

5. East Asian and South European Welfare Systems Compared
5.1 The South European Countries
5.2 Historical Comparison
5.3 Comparison in the Qualitative Perspective
5.4 Quantitative Comparison

6. Explanations for Different Development
6.1 Conflict-Theoretical Approach: Democratization and Welfare State Development
6.2 Cultural Explanation: Confucian Type of Welfare State
6.3 Developmental Approach
6.4 Discussion

7. Conclusion

8. References

1. Introduction

A change takes place in the exploration of social policy in non-Western regions since the 1990s. The questions of social security in societies in transition has primarily been the subject of the developing countries research, yet in the last years, sociologists and political scientists, who have done research on Western welfare states turn to these countries as well (Midgley, 1986; Jones, 1990, 1993; Esping-Andersen, 1997; Goodman et al., 1996, 1998; Wilding, 2000; Holliday, 2000; Tang, 2000, Gough 2000). This disciplinal change has a real foundation. In some transition societies, formal social security systems develop since the 80s and even more since the 90s of the twentieth century. Their ground structure is comparable with those of the Western countries. We call these states the emerging welfare states. Social security is not merely the duty of the family, local networks or subsidy anymore.

Therefore, it seems to be wise to apply concepts and methods, which have been developed for the Western welfare states to non-Western countries. The question is whether or not this is possible. It depends on one the hand to what extent emerging welfare states are comparable with Western welfare state models and on the other hand to what extent existing explanation models and approaches are able to explain certain developments, which differ from developments we know from Western countries. These basic questions lead to the problem, which this thesis is dealing with.

1.1 Formulation of the Research Questions

There are several regions and countries, which are considered to be emerging welfare states. In East Asia several countries are relatively advanced in their efforts to create a comprehensive welfare system. Therefore, countries and territories such as South Korea, Hong Kong, Taiwan, and Singapore are interesting cases to analyze the welfare state development in non-Western regions. Besides, the information on the development of welfare systems in East Asia compared to what is available in the Western welfare states is insufficient and unsatisfying (Aspalter 2002a: 1). The following questions regarding social policy and welfare state in East Asia come up:

- Which structures can be found there,
- how do they differ from Western patterns,
- and what are the causes for the special development?

Since the late 1990s scientists turned towards this region to get a deeper insight in the development regarding the welfare state system. Either they attempted to use existing explanation models, that is to say such, which have been developed for Western welfare states (e.g. Esping-Andersen, 1990) or they brought up new yet controversial approaches (Jones, 1993; Aspalter, 2002b; Holliday and Wilding, 2003). This thesis picks up these three main questions above. In doing so I will overcome the limited character of Western-based approaches. While drawing on the very popular and dominating classification of welfare states of Esping-Andersen, the Three worlds of welfare capitalism (1990), this thesis proposes to overcome its shortcomings by expanding these three worlds by two more: the East Asian and the South European regime type. Esping-Andersen’s Three worlds of welfare capitalism (1990) are still an obvious reverence point, not only because others have turned to it in examining East Asian social policy but also because it provides a powerful framework for comparative analysis. This case study focuses on East Asia, containing the four countries Hong Kong, South Korea, Taiwan, and Singapore.

In order to find out more about the precise nature of East Asian special position, an interregional comparison with the South European countries is carried out. In a context-oriented way I will identify the factors that make up the welfare state in East Asia. It offers the advantages of holistic, rich descriptions and full, chronological case account. I expand my comparison approach to different fields of welfare phenomena including development contexts, qualitative characteristics, quantitative expenditures, and historical development through different stages. In that respect my theoretical approach follows to some extend a classic comparative research tradition. The comparative analysis with South European countries is meant to show the similarities between both regions yet particularly the distinctiveness of the East Asian countries. Comparisons reveal the particularities of the cases, which are not single countries but regional entities.

In the final part I will discuss the causes of East Asia’s special position by slipping in the results of the comparison. They will help to assess to flaws and virtues of existing explanation models. I will focus on three different approaches,

- on a conflict-theoretical approach, which claims that democratization is the

driving force behind welfare state development,

- an cultural approach, which draws on Confucian values because of its emphasis

on family obligations, education, paternalism, and social harmony, to explain

East Asia’s special position, and

- the development state approach, which is usually defined as a state where elite

policy makers set economic growth as the fundamental goal and pursue a coherent strategy to achieve it.

1.2 Content

It is in six parts. In Chapter 2 I will make some methodical considerations. I will comment on my terminology, my choice of cases, and on the selection of the theoretical approach.

Chapter 3 summarizes traditional welfare explanation models. This theoretical knowledge gives the reader, who might be less familiar with theories of welfare state development, the opportunity to know more about it and to apply these approaches to welfare state development in East Asia. The last part of this Chapter discusses Gøsta Esping-Andersen’s The Three Worlds of Welfare Capitalism from 1990, which sought to place Western welfare systems in three categories. It has been a basis for much modern welfare state research and I will demonstrate its inapplicability for East Asian countries.

Chapter 4 discusses the East Asian welfare state. The chapter explores the historical development of social welfare in the four societies, and then analyses their development in the social policy sectors of social security and health.

Chapter 5 will turn to the empirical and analytical work. It is a comparative study of the emerging welfare states in East Asia. I will compare the East Asian welfare states with those in Southern Europe. Although there are some basic similarities in the development, differences between both regions emerge in history, politics and economical concerns. Here I will try to find out how the East Asian welfare states differ from those in Southern Europe.

Chapter 6 concludes the comparative analyses by seeking for explanations for the particular development in our four societies. I will apply approaches, which were presented in Chapter 3, to East Asia. We will see that the theoretical interpretation of East Asian welfare state development by traditional welfare state approaches is a difficult task, because they were originally developed for Western regions. Hence, I will present new concepts, which focus especially on this region. I will argue that the explanatory capacity of the developmental approach, which emphasis the subordination of welfare to economic considerations in East Asia due to the overriding commitment to economic growth and full employment, in connection with the cultural values of the Confucianism, is the most convincing explanation for East Asia’s special position.

I have chosen to focus on what Hort and Kuhnle (2000) call Asia’s tiger economies Hong Kong, Singapore, South Korea, and Taiwan.^[1] In order to give an answer to how the social welfare systems of those four countries differ from Western patterns they are compared with those in Southern Europe, which includes Italy, Spain, Portugal, and Greece. Ferrara (2000) differentiate these countries as an own welfare regime type, besides the liberal, social democratic and continental/corporatist regimes types, which were set up by Esping-Andersen (1990). Both regions share a few of similar development characteristics. Both are latecomers in terms of installing social security systems and democratization, so East Asia and Southern Europe share a development backwardness compared to Western European or North American countries. Furthermore there are similarities in the strong role of the family in providing social security and a general similarity in living standard and income.

This comparison is a historical, qualitative, and a quantitative one, where measurable welfare outcomes will be compared. It is not an intraregional comparison of the Asian countries but an interregional comparison between countries from two different regions.

2. Methodological Part

In this part I will make some methodical considerations. The chapter is divided into five sections. The first section is about terminology understanding. In the second section I will comment on the used data sources. The third section is a broad introduction to the advantages and the general problems of the comparative approach, whereas the fourth section takes a look at the comparative welfare state literature. In the fifth section I explain my choice of cases.

2.1 Comments on Terminology and Definitions

There is a remaining a problem of terminology. The term ‘welfare state’ in the Asian context frequently denotes ‘state hand-outs’ or charity. It implies something like a state as a welfare provider, which must not necessarily be like that. Especially in East Asia, the role of the welfare state is much more a welfare regulator (Aspalter, 2001: 3). Perhaps social policy regime or human development regime would be a preferable term as suggested by Gough (2000: 4).

Based on ideologies, for conservative-oriented people the welfare state is something negative, i.e. a system of high taxation and high redistribution from the more rich to the less rich, or from the majority to a poorer minority. For Conservatives, who draw on Christian social thoughts e.g. by the teachings of influential pioneers of the Christian social movement (such as Bishop Ketteler), the welfare state is a means for achieving a greater societal well-being, which supports an occupationally divided societal structure. For representatives of the left political spectrum, the welfare state is often seen as the response of social democracy, socialism, or communism to capitalism. For most of them, a welfare state system connected with redistribution is a very positive thing, for, which it is worth to bear the burden of high taxes (Aspalter, 2002b: 10).

For Esping-Andersen, “the welfare state [is] a principal institution in the construction of different models of post-war capitalism” (Esping-Andersen, 1990: 5). I will go on using the term in the meaning of a ‘welfare state system’, as Aspalter (2002b: 5) suggests it, with the understanding that it refers to a wide definition of social programs encompassing broad social and developmental goals. Welfare state systems are according to Aspalter defined as the sum of governmental regulations and provisions that aspire an enhancement of people’s living conditions (Aspalter, 2002b: 11). The use of the term system implies a broader view. The system include laws, regulations and planning of all governmental institutions in the fields of employment, taxation, social insurance and social assistance, health care education and housing. The concept of welfare state systems avoids the notion of a strong financial engagement of the state in provision of social welfare (Aspalter 2001: 3).

2.2 Data Sources

There is an increasing amount of useful data from international databases, such as the one used for quantitative comparison issues in this paper. The most important and accessible sources are perhaps the Organization for Economic Cooperation and Development (OECD), the United Nations Development Programme (UNDP), the World Factbook published by the Central Intelligent Agency (CIA), and the World Bank. Such organizations have reworked national government statistics and put them on a comparable basis. One is of course left to analyze and theorize the data, though obviously it contains its own bias shaped by the kinds of statistics, which governments collect. In addition, agencies like OECD or the World Bank are not primarily concerned with the comparative social policies. Their interest in the field is in examining the relationship between social policies and the regulation of public expenditure and labor markets. Such data tend to concentrate on direct public services and benefits, so that fiscal, occupational, and private welfare forms are sometimes neglected, not least because governments often do not collect good data on these.

2.3 The Comparative Method in General

The comparative method is as old as political science itself. In this respect, it can be said that because of the general impossibility of experiments in political science, we are forced to compare, e.g. between different time periods.

Skocpol and Somers (1980) differentiate three distinct approaches. The first type of comparative history is the parallel demonstration. The point of the comparison is to assert a similarity among the cases. It demonstrates that the theoretical argument, which is posed, is applicable to multiple cases. Cases are selected to cover all possibilities, or to represent a range of sub-types or points on continua (Skocpol/Somers 1980: 176).

The second approach, the one I will concentrate on, is the contrast of contexts. It concentrates on what is special for your case. Important is to bring out the unique features of each particular case included in the discussion and to show how these unique features affect the working-out of social processes. In the first line, contrasts are drawn between or among individual cases (Skocpol/Somers 1980: 178).

A third type of approach, the macro-causal analysis, contains two analytical designs. On the one hand, macro-analysts try to establish that several cases having the phenomenon in common to be explained also have the hypothesized causal factors, even though the cases vary in other ways that might have seemed causally relevant. This approach is labeled the “Method of Agreement”. On the other hand, macro-analysts can contrast cases in, which the phenomenon to be explained and the hypothesized causes are compared to cases in, which the phenomenon and the causes are both absent, although they are as similar as possible. This procedure is called the “Method of Difference” (Skocpol/Somers 1980: 181-185).

There are a large number of methodical difficulties in the comparative approach. A fundamental problem is the presence of too many variables and too few cases. Statistically speaking, the number of variables is infinite and the number of cases is limited. Furthermore, many of the statistical data cannot be directly compared, which creates even more problems. At last, also the meaning of apparently the same events can be different in different countries (Marsh et al. 1997 p.180-183).

2.4 Quantitative and Qualitative Comparisons of Welfare States

A form of comparison develops quantitative indices and concepts for comparing the origins and performance of welfare states. It is therefore perhaps the most deserving of the title ‘comparative’ in the eyes of social scientists. Here analysts use data from national government and international agencies such as those mentioned above to develop comparative hypotheses about the development and impact of welfare states. Wilensky (1975) compared social security spending as a proportion of GNP in sixty-four countries for 1966. The most prominent example of this genre is Esping-Andersen’s Three Worlds of Welfare Capitalism (1990).

Beyond the quantitative literature, qualitative comparative analysis of social policy is dominated by a ‘structure diversity’ approach, which emphasizes the diversity or even the uniqueness of each welfare state in its national social and historical context (Ginsburg, 1992: 23). Domestic and political processes, cultural values, economic forces, demographic factors, or whatever may structure the diversity. Such texts sometimes confining themselves to single policy areas and/or few states. Notable examples, which all offer their own structural emphases, are Rimlinger (1971), Flora/Heidenheimer (1981), Ashford (1986), Skocpol (1992) and Pierson (1994) and recently on East Asia Tang (2000). Whether this kind of literature is comparative is a matter of debate, since methodical and/or theoretical comparison is often either underdeveloped or absent (Ginsburg, 1992: 23).

Therefore, comprehensive framework including qualitative, quantitative, and historical context-oriented aspects seem to be the best procedures for exploring welfare phenomena. Through the context-oriented way I will identify the factors that make up the welfare state in East Asia. The advantage a of holistic chronological case account gives the opportunity to expand my comparison approach to different fields of welfare phenomena including various development contexts.

2.5 Selection of Cases

Probably the most important dimension in comparative method is the number of countries analyzed. Aim of the thesis is to find out more about the peculiarities of the emerging welfare states in East Asia. This region has been a star performer in terms of economic growth, industrialization, and poverty reduction over before the economical crisis in the late 90s (World Bank, 1993). Especially been Hong Kong, Korea, Singapore and Taiwan were singled out initially as economic miracles.^[2] The period of high and sustained growth in the miracle economies was accompanied with an implementation of welfare services. Besides the economic and social development, democratization took place as well. In Taiwan, South Korea, Japan, and Hong Kong developed strong democracies within only one and a half decades (Aspalter 2002b: 2).

As the selection of cases, I will choose as the four leading economies (South Korea, Taiwan, Hong Kong, and Singapore) behind Japan in this region. Japan is not an emerging welfare state anymore since social welfare has been introduced there much earlier.

According to Esping-Andersen (1990) Europe comprises three distinct welfare regimes. These are reflected in different patterns of social policies, different welfare outcomes, and different stratification effects and paths of development (Esping-Andersen, 1996: 26). In recent times, besides the East Asian countries distinct model of Southern European welfare has been contented (Ferrera 1996, 2000; Moreno 2000). The discussion revolves around the contention whether the Mediterranean type of welfare constitutes a regime of its own or is simply lagging behind those of the ‘continental’ model of social insurance to, which they might belong.^[3]

The South European countries will be used as the counterpart to the developing East Asian group for the comparison. By ‘Southern Europe’ I will refer to Greece, Italy, Portugal, and Spain. These four countries share analogies regarding historical backgrounds, value-systems and institutional peculiarities. They all had past experiences of authoritarian and dictatorial rule (for longer periods in the case of Portugal and Spain), and have suffered from economic and industrial ‘delays’ in the processes of modernization. The religious factor has had a structuring role in all four countries. Furthermore, the membership in the European Union brought about incentives to economic convergence with Northern and Central Europe (Economic and Monetary Union). In broad terms, similar social-demographic trends and macroeconomic constraints can be observed in all four countries. Nevertheless, a much richer and heterogeneous reality is acknowledged (Moreno, 2003: 147).

South Europe seems to be suitable for a comparison with East Asia: Esping-Andersen (1997) characterized the Japanese welfare regime as a hybrid one, with similarities to that of Southern Europe. Above all the family role is similar in Southern Europe and East Asia. The families are an important source of assistance (in the form of work, cash and care-giving) because they have weak formal safety nets. In Italy, Spain, Portugal and Greece around three fifths of young people are employed by family members. Similarly, they exhibit low rates of female employment (except Portugal), together with segmental interest representation, clientelist party relationships, and Catholic charitable provision primarily for the aged and disabled. This cluster of factors explains the absence of national safety nets in Southern European countries. Despite these elements in common between Southern Europe and East Asia, there are differences, notably in labor market regulation and flexibility (Gough, 2000a: 18).

Because of shared similarities South Europe seems to be the appropriate candidate for a comparison with East Asia. By selecting East Asia and the South Europe as cases, I believe I have chosen a dynamic middle course. The eight countries have a large number of socio-economic similarities. Both have reached the same quality of life according to the Human Development Report and a similar GDP per inhabitant. They also share analogical features regarding the welfare state: modest safety nets and the importance role of the family. Yet both country groups derive from a completely different welfare state tradition. The analogies make it more possible to isolate distinct welfare state characteristics and to reveal the causes of the East Asia’s special position.

3. Theoretical Part

In this part of the essay I will create the theoretical and analytical framework that will be the basis for the analysis. The research question ‘why state welfare?’ has intrigued many researchers in Western democracies, and there is no lack of explanations and theoretical frameworks for this phenomenon (Tang, 2000: 19). To get an overview of the amazing number of welfare state explanation models, I will give in this Chapter an overview containing the main theoretical explanations and distinguishing between different levels of analysis. There are many different ways to comprehend the phenomenon of welfare state development. In general, these concepts and methods has been developed to explain Western welfare state development and have been not applied to East Asia. Therefore, in the second part, I will briefly discuss the approach of Esping-Andersen, the Three worlds of welfare capitalism (1990) and try to overcome its shortcomings by expanding these three worlds by the East Asian one.

Normally, the authors did not try to transform their thoughts into empirical verifiable statements but in their works one can find several claims, which have hypothesis character or, which are transformable into hypothesis. Nevertheless, one has to avoid the term ‘welfare state theory’ as it uses for instance Aspalter (2002a).

3.1 Functionalist Explanation Models: The Logic of Industrialism

The functionalist approach explains welfare state extension principally with an increase in the need for social welfare, such as industrialization, or the advance of capitalism. The extension of welfare needs is only a necessity not a sufficient condition for the establishment and the extension of welfare state systems. The functionalist position is the thesis of convergence, which suggests that industrialization and economic growth encourage convergent welfare state forms, despite differences in political ideology.

In the functionalist tradition there are two streams of theories, the Marxist, and the Non-Marxist stream. Marxist functionalist approaches stress the role of collective protest and ‘Klassenkampf’, whereas non-Marxist pluralistic approaches stress the right of organization and the democratization of suffrage (Alber, 1987: 74).

Marxist theorists promote the idea that social policy is a response of non-capitalist states that try to legitimize their unjust, capitalist rule with more or less generous welfare policies. The welfare state is understood as a logical development of capitalism to a stage where the state on the one hand must offer concessions to the oppressed majority and on the other hand to secure the capitalist mode of production, especially the accumulation of capital. In this stream, the hypothesis of legitimization crises of capitalism was put forward, which influenced a great deal of welfare state research from the 1970s onwards (Aspalter, 2002a: 11).

Non-Marxist functionalist explanation models brought up the idea that the degree of industrialization of a certain country is the main cause of the amount of welfare state expenditures. Economic growth, demographic and bureaucratic development were made responsible for the establishment and the extension of welfare state institutions. Hence, not only that the industrial society increases the need for vast social policy but also the growths of massive bureaucratic organizations that provide the administrative power to engage in large-scale social security and redistribution programs. At the same time, the ability of the family to take of the elderly and children is weakened. Therefore the industrial society is demanded to satisfy growing welfare needs. Since Non-Marxist functionalist approaches identify industrialization as the main cause of welfare state development, they predict a common trend of convergence of welfare state systems but fail to explain the persisting divergence of these systems (Aspalter, 2002a: 11-12).

Aspalter argues that none of the functionalist groups of welfare state approaches is sufficiently capable of explaining welfare state development (Aspalter 2002a: 20). Therefore, functionalist comparison has become unfashionable in the wake of the sociological critics of functionalism (Ginsburg 1992: 20).

3.2 Conflict-Theoretical Explanation Models

The conflict school does not see the welfare state primarily as a necessity of social economical change but as a result of growing sociopolitical demands, which emerge alongside sociopolitical mobilization processes (Alber, 1987: 74). The conflict school provides a solution and overcomes the shortcomings of the functionalist tradition: e.g. the involvement of political parties and social movements in welfare politics.

In Marxist traditions of conflict-theoretical explanation models are three major approaches:

1. the classical labor-centered approach, which follows the concept of the working-class mobilization thesis,
2. the new class/group-centered approach, which incorporates the important role of social groups and social classes,
3. the policy-centered approach, which recognizes the importance of ruling political parties, the logic of elections and corporatism (Aspalter, 2002a: 13-16).

The point here is that political forces of agency, particularly party politics, have had a predominant influence over welfare state and therefore comparative political differences account in large measure for differences in welfare expenditures (Ginsburg, 1992: 20).

The classical labor-centered approach explains the different progress in the development of welfare states with the strength of left-wing parties and labor unions. However, this approach neglects to determine the role of other, non-leftist political parties, and labor unions in welfare politics. The new class/group centered approach overcomes some difficulties of the classical working-class mobilization concept, e.g. it acknowledges the importance of farmers’ and middle classes, beside the working class and social groups that influence politics as well (Aspalter, 2002a: 13-14).

The policy-centered approach advanced a greater inclusion of the policymaking processes in the development of welfare states. The policy-centered approach is not an exclusive Marxist approach, since there are also a great number of Non-Marxist theorists who developed this approach by side with the Marxists. The policy-centered approach analyses the impact of democratic participation, electoral and party competition, election and party systems, strength and participation in government of all political parties, government coalition formation, distributional coalitions and so on. The studies of the policy-centered approach share a common focus on political determinants that shape social policy, especially political parties and democratic elections (Aspalter, 2002a: 14-16).

The Non-Marxist tradition of conflict-theoretical explanation models developed also three major approaches:

1. the policy-centered approach, which recognizes the importance of ruling political parties, the logic of elections and corporatism,
2. the state-centered/institutional approach, which concentrates on the impact of political systems and institutions on welfare politics, and
3. the elite-centered approach, which investigates the role of the governing elite in welfare politics (Aspalter, 2002a: 16-20).

The policy-centered approach analyses the impact of democratic participation, electoral and party competition, election and party systems, and the like (Aspalter, 2002a: 17). The study on political parties and democratic elections recently became the key factor for an understanding of political and social forces behind welfare state development and the political processes that are directly responsible for changes in welfare arrangements. Several studies on East Asian countries (Ramesh and Asher 2000; Aspalter 2002; Croissant 2004b; Tang et al. 2004) have focused on the impact of democratization on welfare state systems. In East Asia, the relationship between welfare and democratization has been under scrutiny since democratization was introduced abruptly in countries like Taiwan and South Korea. There are some indications that governments in some Asian Tiger economies like Taiwan and South Korea have made welfare a prime subject of electoral promises, while the more authoritarian government of Singapore has also relied on welfare measures to win public support (Tang et al., 2004: 2-3).

The state-centered/institutional approach focuses on the nature of political systems, power structures, governmental institutions, constitution making, and involvement in wars, national development formation processes, and types of political economies such as corporatist bargaining systems. The state-centered/institutional approach was mainly developed in the United States and helped a great deal in explaining the failure and the relative absence of social policies in America, compared to its European counterparts. The essential findings of this new approach are that differences in state structures matter, and that the lack of independent power of the state may explain the lack and the failure of social policy programs. Furthermore, national mobilization for war efforts, national development formation processes, and strong governmental and corporatist institutions may also explain a large-scale extension of social policies (Aspalter, 2002a: 15-16).

The elite-centered approaches stress that a governing elite, those who hold positions of command in society controls the mass of people. What this approach distinguishes is that it describes a particular group of e.g. governmental, economic, and military elites that try to legitimate their power by means of welfare policies. It does not ascribe the power in social politics to social groups, social classes, or political parties but to governing elites (Aspalter, 2002a: 16).

In the next section I will present in detail the important contribution to the literature: Esping-Andersen’s Three Worlds of Welfare Capitalism. His framework will serve as my theoretical foundation in exploring the East Asian welfare state.

3.3 The Three Worlds of Welfare Capitalism by Esping-Andersen

Recently Gøsta Esping-Andersen has brought about a major step forward in the evolution of welfare state theory. His famous book, the Three Worlds of Welfare Capitalism (1990) is a representative of the policy-centered approach in Marxist conflict theory (Aspalter, 2002b: 8). Esping-Andersen has avoided two opposing traps:

- A functionalist approach, which over-emphasizes convergence between national social systems.
- A detailed writing of histories that over-emphasizes national uniqueness and divergence, and specific social programs (Gough, 2000b: 2).

The task Esping-Andersen sets himself is to examine “a novel phenomenon in the history of capitalist societies” (Esping-Andersen, 1990: 1), which is the transformation into welfare states.

He developed a comparative approach, which could identify countries with clusters of features in common, and which could combine institutional persistence and group conflict. It is “qualitatively different welfare-state logics” (Esping-Andersen, 1990: 4) that divide the 18 states he looks at into conservative, liberal and social democratic regimes marked by distinct developmental trajectories. The welfare state regime includes a certain welfare state arrangements and a certain labor market regime. For him the types of institutional organization, political reasons of welfare state measures and distribution criteria are important as well as the financial amount of welfare state (re-)distribution. Two aspects as dimensions of his typology are prior: The extent of decommodification, which means, to what extent individuals are made independent by the welfare state to the forces of the capitalist market (Esping-Andersen, 1990: 21-22). Moreover, as a second aspect, the way in which the welfare state influences the social structure and the way it (re-) structure social disparity (Esping-Andersen, 1990: 23).

Within the OECD world, Esping-Andersen distinguishes three welfare regimes - the liberal (exemplar country: the US: the UK comes closest in Europe), conservative (exemplar countries: Germany, Italy) and social-democratic (the Scandinavic countries) (Esping-Andersen, 1990: 26-27). By demonstrating the way welfare regimes shape interests and ideas in different societies, he claims that they follow a different path of development. The liberal regime generates a division between the poor and the rich, the conservative regime divides people along status lines but commits all to a state welfare and family-based politics, and the social democratic generates a universal support for the welfare state (Esping-Andersen, 1996: 29-32).

Esping-Andersen’s approach has not been unchallenged. One point is that in concentrating on class analysis, it ignores other sources of stratification such as religion, ethnicity, and gender. Then, in my opinion, the effects of the gendered division of labor and household forms are ignored at all three levels (social programs, welfare outcome and stratification effect). Holliday (2000) criticizes the criteria for inclusion. There is no good reason to exclude such states, which subordinate social policy to other policy objectives. Therefore, major regions of the world are left out. Furthermore, on critic is that he solely relied on selected quantitative data. For Aspalter (2002a), the major weakness in Esping-Andersen’s approach is the unsystematic identification of the causes of the particular development. While referring to the dominance of the social democratic party in Scandinavia since the 1960s, he refers to the historical corporatist-statist legacy of Germany, France, Austria, and Italy and to traditional liberal work ethnic norms in the case of the liberal regime type (Aspalter, 2002a: 15).

Aspalter argues that Esping-Andersen confuses Christian democratic social policy in postwar Germany with that of archconservative Chancellor Otto von Bismarck at the end of the 19th century. Aspalter argues that he completely neglects any new development in Germany after 1945. Thirdly, Esping-Andersen does not distinguish between social policies and welfare systems in Japan, the United States, Australia, and the United Kingdom (Aspalter, 2002b: 8).

Other critical commentaries on his typology are that it gets some countries wrong: Ferrera (2000) on Italy and Jones (1993) on Japan. They propose that those two countries make up an own brand of welfare regime: the South European and the East Asian one. In the next part of the chapter I will briefly examine, which difficulties appear when Esping-Andersen’s Three Worlds of Welfare Capitalism is applied to East Asia.

3.4 Applicability of Esping-Andersen’s Approach to East Asia

Esping-Andersen left the East Asian countries out of his typology. The question arises, whether or not this approach can be applied to the countries of East Asia, in particular South Korea, Hong Kong, Taiwan, and Singapore. There are several reasons for doubting this. Table 1 shows that the role of the East Asian state in welfare is tiny, even by the standards of the developing world:

Table 1: Welfare spending in Percent of the GDP

illustration not visible in this excerpt

Source: Data for the Asian countries: CIA 2005; Health, Welfare and Food Bureau,

Hong Kong 2005; ILO 2005; Ku, 2003: 150f; Mok, 2003: 61; OECD 2005a; Social Welfare

Department, Hong Kong 2005; UNESCO 2005; World Bank 2005.

Data for the other countries: OECD 2005a.

Notes: Classification of countries to the three models according to Esping-Andersen (1996).

Values are means.

The amount of welfare spending in the four tiger economies is, compared to each of the three regime clusters, extremely low. The average spending of liberal countries is two times higher, and for conservative and social-democratic countries it amounts to the threefold.

The degree of decommodification is also insignificant due to the low welfare spending. A means-tested system dominates the structure of welfare policy in the sense of an orientation towards a social security as a ‘last resort’. Social security is restrictive, fragmented and thereby distributing the risks; exceptions are the education system, the housing sector, and partly the health insurance. Pension insurance is not available or oriented at a minimum output (Rieger and Leibfried, 1999: 425-426). A means-tested poor relief offers a safety net of last resort but if the benefits are low and associated with social stigma, it will compel all except for the most desperate to participate in the market.

A third point for the inapplicability of East Asia to one of these three worlds is the small role of formal welfare institution in shaping stratification outcomes and interest group formation:

Table 2: Income Disparity (Gini-Index)

illustration not visible in this excerpt

Source: UNDP 2005; for Taiwan: Tang, 2000: 149; data 1985-1996.

Notes: Classification of countries to the three models according to Esping-Andersen (1996).

Values are means.

The Gini-index is a measure for income disparity. Values can be between 0 (absolute equality) and 1 (maximum inequality). Table 2 shows the higher value of the East Asian countries compared to the other regime types. This means that income disparity is averagely higher in East Asia than in liberal, conservative, social democratic countries. It may be likely that the input factor social spending and the output factor income disparity are connected with each other: the higher the amount of social expenditures the higher income disparity.

According to Gough (2000a), further factors can undermine the inapplicability of the welfare regime approach of Esping-Andersen to East Asia. These are the questionable validity of the concept in societies with a significant agricultural population, the small role of formal welfare institution in shaping stratification outcomes and interest group formation, and the impact of external factors like the global economy and supra-national institutions is correspondingly larger than in the OECD world (Gough, 2000a: 3). In the same way Holliday argues that we need not to restrict analysis to internal matters: we need to focus on two levels of analysis: unit and system (Holliday, 2000: 717).

[...]

---

^[1] Formally, Hong Kong is known as the Hong Kong Special Administrative Region of the People’s Republic of China, Singapore as the Republic of Singapore, and South Korea as the Republic of Korea. Taiwan has an ambiguous international status, and limited recognition. It is sometimes known as the Republic of China.

^[2] Hort and Kuhnle labeled them the four “small tigers or dragons” (Hort and Kuhnle, 2000: 163).

^[3] Italy, the only country of the four Southern European countries included in Esping-Andersen’s framework of the Three worlds of welfare capitalism, was clustered as a conservative regime type (Esping-Andersen, 1996: 74).