Politics Against Convergence Current Trends in ... - Wiley Online Library

© (1996) Swiss Political Science Review 2(3): 1-xxx

Politics Against Convergence? Current Trends in European Social Policy

Giuliano BONOLI, Vic GEORGE and Peter TAYLOR-GOOBY

Abstract In this article, we look at the issue of convergence in social policymaking within EU member-states. Proponents of the convergence thesis argue that since European welfare states are facing similar socioeconomic challenges, they are likely to develop common response strategies. Our analysis, based on a survey carried out among policymakers, looks at political debates on the future of social protection in four EU member states. The main focus is on the different institutional structures which characterise European welfare states, and on their impact on political debates in terms of coalition formation. The analysis shows that the influence of socio-economic change on political debates is important, but that it is significantly mediated by the different institutional arrangements which distinguish European welfare states. We conclude that the presence of common socio-economic problems does not necessarily imply a convergence in policy-making.

Introduction Much of the discussion surrounding the future of European welfare states has focused on the notion of convergence.1 It has been argued that since Western European countries are facing the same sort of challenges, it is likely that they will adopt similar responses which will gradually lead to a convergence of social policies. The convergence thesis is supported by a number of theoretical arguments. On a general world level, it is argued that current trends of intensification of economic and cultural relations between countries (globalization) 1

We gratefully acknowledge the financial support of the ESRC (grant no R000234.711), the Directorate General V of the European Commission (grant no Soc 94 103019), the Anglo-German Foundation (grant no 1037), COSZ (Netherlands) and the University of Kent for the project "Squaring the Welfare Circle in Europe".

2

GIULIANO BONOLI, VIC GEORGE AND PETER TAYLOR-GOOBY

are leading towards a more homogeneous world, in which national differences tend to disappear (Lash and Urry 1987). In relation to social policy, the impact of globalization is expected to entail a downward trend in social protection standards, in order to preserve the competitive edge of national industry in the face of the emergence of new zones of production, such as South-East Asian countries (Euzéby 1992). On the European scale, convergence is obviously related to the ongoing integration process into a European Union.2 There are a number of ways in which European integration can foster convergence in social policies. First, the existence of a single market without restrictions on the movements of goods and services, people and capitals is likely to put pressure on memberstates to harmonise their levels of social protection, in order to avoid a possible redirection of investment towards member-states with lower labour costs resulting from lower standards of social protection. Second, the European Commission is actively encouraging member-states to develop compatible, if not similar, social policies, in an effort aimed at harmonisation (Kosonen 1994; Hantrais 1995). Third, the criteria for the monetary union set out by the Maastricht treaty are putting pressure on governments to reduce expenditure. The combined outcome of these pressures is expected to lead towards a residual welfare state, in which only basic needs will be covered by the state, while the rest will be catered for by occupational arrangements (Ferrera 1993b) or by private provision (Abrahamson 1991). These accounts concentrate mainly on the quantitative dimension of convergence, i.e. on the level of social protection. However, another important dimension of the debate on convergence relates to the structure of European welfare states.3 Broadly speaking, two different models can be identified in Western Europe.4 On the one hand, there are systems based on social insurance of Bismarckian inspiration, in which entitlement is based on each individual's contribution record and which grant earnings-related benefits (France, Germany). On the other hand, there are the universalist welfare states, which refer to the Anglo-Scandinavian tradition. In this case the whole population is covered by a single plan, cash benefits are usually flat-rate, and provision is predominantly tax-financed. These two models have characterised the origin of European welfare states. In more recent years there has been some convergence, so that ini2

For an account of recent social policy developments in the European Union, see George and Taylor-Gooby (1996). 3 On the distinction between different dimensions in social policy research, see Ferrera (1993a: 11 and 318-320). 4 Typologies and classifications of welfare states are legion. In general, however, they all distingush between the two models mentioned here, ie. continental social insurance and AngloScandinavian universalism (Korpi 1983; Esping-Andersen 1990; Ferrera 1993a).

POLITICS AGAINST CONVERGENCE?

3

tially Bismarckian welfare states have developed new schemes aimed at covering those who are not protected by social insurance. Conversely, in Britain and in Scandinavian countries, contributory earnings-related schemes have been introduced as a supplement to universal provision, especially in the area of pensions (Baldwin 1990; Ferrera 1993a). The process of convergence, however, is far from complete, so that if one looks at the current structure of European welfare states, one will find that the initial choices (in terms of universalism-social insurance) still affect the shape of today's welfare states (Merrien 1990b) and on the direction of current reform (Pierson 1994; Myles and Quadagno 1996). The two models of social protection have different implications in terms of their effectiveness in guaranteeing sufficient resources to individuals and in terms of their impact on economic competitiveness and employment. In particular, the adequacy of social insurance as the major means to provide state welfare has been challenged by a number of commentators. This has been the case in France, where two major weaknesses of the social insurance model are frequently stressed (Rosanvallon 1995; Castel 1995a and 1995b). Social insurance, it is argued, is a non-efficient method to guarantee citizen's welfare, because it contributes to unemployment and to social exclusion. It contributes to unemployment because contributions represent a direct increase in the cost of labour. It contributes to social exclusion because the weakest individuals in society, who are in most need of social protection, have often been unable to built up sufficient contribution records so as to receive adequate benefits. The debate has recently gained prominence also in Switzerland. For instance, Rossi and Sartoris (1995) suggest a radical shift in the structure of the Swiss welfare state towards a Scandinavian model. Their argument is based on the observation that social insurance is increasingly inadequate to deal with current problems of social exclusion. Since entitlements to benefits need to be earned by performing paid work, there is a risk to leave those with inadequate connections with the labour market unprotected. Flückiger and Cordero (1995) have also questioned the adequacy of a predominantly social insurance-based welfare state, by emphasising the detrimental impact of contribution financing on employment. In this article we look at the debate on the relative merits of social insurance and its alternative, universal tax-financed provision, in the wider European context. The analysis is based on interviews with representatives of influential organisations (political parties, trade unions, businesses, voluntary organisations and civil servants) carried out in 1994. The material gathered gives an insight into the pattern of arguments in welfare debates and the views of the different actors that can be used to test the prediction of theo-

4


ries about welfare state development. The study covered France and Germany, two countries which reflect many features of the social insurance model; Denmark, which in contrast has a welfare system based to a larger extent on tax-financing and on universal provision; and Britain, which combines elements of both models. By looking at the preferences expressed by different actors in the four countries, we will try to assess whether or not there is a tendency towards further convergence in social protection policies.

France: Towards "Fiscalisation"5 The French debate is characterised by the presence of two different and fairly incompatible normative conceptions of social protection. In general, a shift towards universal-tax-financed provision is advocated by the government, with the support of the Socialist opposition and the civil service. Such a shift, however, is fiercely opposed by the trade unions and to some extent by employers (CNPF).6 The government's argument is mainly based on the need to reduce the cost of labour in order to deal with the problem of unemployment. It is argued that high contributions discourage the creation of new jobs and thus it is preferable to replace them with taxation (fiscalisation). A second facet of the government's argument refers to a notion of coherence between the financing method of a programme and its coverage. It is relatively widely accepted that universal schemes (solidarité nationale) should be taxfinanced, whereas those which cover only employees (or other groups defined in occupational terms) are to be financed through contributions (Rosanvallon 1995: 82). The debate mainly concerns two schemes: health insurance and family benefits. This is the case arguably because these two programmes, although they originated as pure social insurance schemes, have been subsequently reformed so that they are currently virtually universal in their coverage. Family benefits are already partly financed through taxation (CSG).7 In addition, the family benefit scheme (CNAF) grants means-tested benefits 5

Organisations approached in France included: members of parliament of the RPR, the UDF, the Socialists and the PCF; representatives of the main trade unions, ie. CGT, CGT-FO, CFDT, CFTC, CFE-CGC; the federation of French employers (CNPF) and officials at the Ministry of Social Affairs and at the INSEE. 6 For an in depth account of the French debate on social insurance, see Palier and Bonoli (1995). 7 The CSG (Contribution Sociale Généralisée) was introduced in 1990 by the (Socialist) Rocard government as a new form of tax. Unlike insurance contributions, it is levied on all kind of incomes (not only salaries), including capital revenues and welfare benefits. Unlike income tax, it is proportional, it is ear-marked for non-contributory welfare programmes, it is levied also on low incomes. Despite the use of the term "contribution", the CSG is a tax, rather than a social insurance contribution (the French equivalent of contribution is "cotisation").


5

which are seen as belonging to the realm of solidarité nationale and should thereby be financed through taxation. Similarly, in the case of health insurance, the coverage of the scheme is now virtually universal so that 99.7 per cent of the resident population is covered (Chatagner 1993), and that is seen by the government as a reason for shifting part of its financing to taxation. As a senior civil servant put it: “At present, there is a consensus on the fact that health insurance can be financed by a tax based on all revenues, such as the CSG. ... The financing of the health insurance scheme can be done through the CSG. This would also take into account the macro-economic argument in favour of reducing the cost of labour.” (interview INSEE, 1994)

The shift towards more tax-financing in health care is opposed by the unions. Their argument is that health insurance should continue to provide for employees only, while the state should be responsible for the coverage of those who do not have access to social insurance. This view was expressed by a representative of the trade unions in the following terms: “The health insurance scheme for employees tends to be considered by the government as a universal scheme. As a result, the government imposes on the health insurance scheme the financial and political burden of covering that part of the population which is unemployed, lives in poverty, etc. ... we would like to see a clarification between the obligations of the state and the obligations of the health insurance scheme.” (int. CGT-FO, 1994)

There is thus a clear contrast between the government, supported by the Socialist opposition, and the trade unions. While the former are keen to shift part of financing from contributions to taxation, the latter would like to restore the original principle on which social insurance is based, i.e. a strict relationship between contributions and entitlements, with additional meanstested tax-financed provision. As another trade unionist put it, “the financing of contributory benefits ... must be done through contributions based on salaries. In contrast, non-contributory benefits must be financed by the public purse” (int. CFE-CGC, 1994). The debate on the financing method has also a strong implication for the management of the social insurance system and for the power relationship between the government and the trade unions. In France, social insurance is managed jointly by representatives of employers and of the trade unions, whereas universal and means-tested provision (solidarité nationale) is the government's responsibility. As a result, a major shift away from the social insurance model would almost necessarily imply a loss of managerial power by the unions. This aspect, although it is rarely present in the official discourse, arguably represents a key dimension in the confrontation between the

6


government and the unions.8 This emerged quite clearly from our survey. The unions “oppose the tendency towards shifting financing from contributions to taxation” since “the transfer of financial obligations to the state will imply the transfer of decision-making power” (int. CGT-FO, 1994). The government, in contrast, for exactly the same reason, is keen on moving towards more tax-financing. As a senior civil servant put it: “The financing system based on contributions has created a system of discussion between people who believe they are the representatives of employees and employers. Such a system has resulted in much abuse. The CSG can remove the legitimacy, or false legitimacy, from the trade unions, which have not done much for the welfare state, and will allow parliament to examine the social insurance budget.” (int., Ministry of Social Affairs, 1994)

The managerial motive seems to be a key dimension in the confrontation between the government and the unions on social protection financing. The abandonment of the principle of contribution financing for insurance benefits will almost certainly mean the questioning of the managerial role played by the unions, a role which is extremely important for them, since it somewhat compensates for the low and declining rates of unionisation (Rosanvallon 1995).

Germany: Confident Conservatism9 The German debate on the adequacy of social insurance is characterised by the presence of a strong coalition of interests supporting the status quo. These include the vast majority of the CDU-CSU, part of the SPD, trade unions and employers. A major shift away from the social insurance model is not really advocated by anyone. However, there is some support among the SPD for the introduction of a supplementary non-contributory pension scheme which would cover those who do not have a sufficient contribution record. Such a move seems to be opposed by most other relevant actors. Radical departures from the present model, but in the direction of a minimal, or residual, welfare state, are suggested by some FDP and the DIHT (medium sized business), but again, these proposals are not met with approval by other relevant actors. 8

Interestingly, the reform plan of the social security system presented by Alain Juppe in November 1995 takes up this issue by announcing the tabeling of a constitutional amendment which would bring the social insurance system under the control of parliament, instead of the social partners (Le Monde 16/11/95). 9 Organisations approached in Germany included: MPs of the CDU-CSU, SPD, and FDP; officials of the employers organisations BDA and DHIT (Deutscher Industrie- und Handelstag); representatives of the IG-Metall and the DAG; and officials of the Ministry of Labour and Social Affairs.


7

There are three main elements in the debate on social insurance. First, as it is the case in France, there is a concern for the impact of contribution financing on the cost of labour and thus on employment. The issue is picked up by the SPD, which “is in favour of shifting the emphasis more to taxation; both for equity reasons but also as a way of reducing labour costs” (int. SPD, 1994). This "French style" solution to the problem of the impact of contribution financing on the cost of labour seems to be favoured only by the SPD. The alternative, which seems to receive more support, consists of introducing more user charges (in order to replace or to avoid increases in contributions); or of shifting the balance between employers and employees contributions in the direction of the latter. These proposals refer to the health insurance scheme and are supported by employers (BDA, DIHT), and sections of the FDP and the CDU-CSU. According to an FDP MP “in future individuals will have to pay more for their health care, either through insurance contributions or direct patient charges” (int. FDP, 1995). These suggestions have not met with unanimous approval by the Christian Democrats: “... there are pressures to reduce employer contributions. I don't agree with this ... but there are different opinions on this in my party” (int. CSU, 1994). A second aspect of the confrontation over the social insurance model, refers to its exclusion-encouraging dimension, especially in the area of pensions. Those who do not have access to sufficient insurance coverage have to rely on social assistance (Sozialhilfe), which is administered at the local level and is regarded as highly stigmatising. The Social Democrats are considering reform in this area: “the key question is how to achieve a more decent minimum standard for all old people. ... at the moment the SPD is moving towards the idea of a basic social minimum that would be means-tested, but provided at the federal level” (int. SPD, 1994). Support from other relevant actors, however, is absent. In contrast, it seems the SPD proposals are not being met with approval by the Christian-Democrats: “we have a social assistance system that provides a de facto national minimum for those with low pensions. That's enough” (int. CDU, 1994). The third and most often mentioned dimension of the debate shows the strong support enjoyed by social insurance in Germany. The social insurance system has contributed to the financing of the German unification.10 Decisions with regard to entitlements in the East have been made without sticking to the contributory principle. Nonetheless, there is a widespread perception that social insurance should strictly comply with the contributory 10

In 1993, the financial transfer from West Germany to the new Länder through the social insurance system, amounted to 47.1 bn D-Mark, equal to 4.4 per cent of social expenditure or to 1.5 per cent of GDP (Statistisches Budesamt 1995).

8


principle and thus the pressure for removing non-contributory benefits (Fremdleistungen) from social insurance. As a Christian Democrat respondent put it “these trends have distorted our social policy system and it is a task for the future to remove these Fremdleistungen and get back to the principles of social insurance” (int. CDU, 1994). The restoration of strict correspondence between benefits and contributions, has been stressed by the majority of the relevant actors. According to a trade unionist: “We would like to see a determined attempt to take these unification elements out of the social insurance system and have them financed directly by the state. If these Fremdleistungen go, then things are not so bad. The health and pension systems are quite viable. Without the unification element pension contributions could be reduced to 15 %.11” (int. IG-Metall 1994)

The position of the German trade unions, thus, is not very dissimilar from that of their French counterparts. Both strongly support the social insurance model and argue that it would be perfectly workable if the government was to take responsibility for non-contributory benefits. Both French and German trade unions participate to the management of social insurance (jointly with employers) and that may explain why they are so attached to the social insurance model. However, in contrast to France, German trade unions can count not only on their own stronger resources (in terms of membership, but also with regard to organisational and financial resources) but also on the support from the Christian-Democrats (see also Clasen and Gould 1995). It seems, thus, that although the French and the German social insurance systems share many common points, including the problems of social exclusion and of labour cost inflation, the way these issues are translated into the political debate is substantially different. In particular the composition of the two camps, those in favour of maintaining the social insurance system and those in favour of a shift towards an Anglo-Scandinavian taxfinanced welfare state, is significantly different in the two countries. While the SPD is alone in suggesting a (moderate) rethinking of social insurance, the same proposals in France can count on a much stronger coalition.

11

The current contribution rate is 19.2 per cent (MISSOC 1995).


9

Denmark: Social Democracy at Stake12 In relation to the mix between social insurance and universal provision, Denmark is the exact opposite of France and Germany. While in France 80 per cent of social expenditure is financed through contributions, the same figure for Denmark is only 10.8 per cent (Eurostat 1995). The fact that Denmark relies predominantly on taxation for financing social protection, could suggest the absence of any debate on social insurance in that country. Some Danish policy-makers, however, seem to be concerned with what they regard as an excessive reliance on the state budget for social policy. The main areas concerned are pensions and unemployment benefits. With regard to pensions, Denmark is a rather unusual case in Europe, since it does not have a compulsory earnings-related pension scheme (Petersen 1991). In this respect, the process of convergence which took place after WWII, did not affect Denmark as significantly as other Anglo-Scandinavian welfare states (Baldwin 1990). As a result, occupational and private pensions developed significantly over the last few decades on a non-compulsory basis, de facto creating substantial inequalities between those who have access to supplementary provision and those who have not. In general, there seems to be a convergence of interests between the different actors on an expansion of occupational pensions. The left, keen on tackling the issue of inequality among pensioners, agrees with the extension of occupational pensions (int. Socialdemocratiet 1994). The right, in contrast, sees the expansion of occupational pensions as a way to meet the projected rising costs due to population ageing (int. Venstre 1994). The coalition supporting the expansion of occupational pensions is thus very strong and includes other political parties (Socialists, Radical liberals, Conservatives) as well as the white collar union FTF. Moves in that direction are contrasted by the SID (a blue collar union), since “we have to pay for it anyway, whether it is a private or a public pension, but the publictax financed pension is the most solidaristic solution” (int. SID 1994). A similar debate is going on in relation to unemployment insurance. The present system is financed mainly through taxation, although the contribution financed part has been increased in 1994 (MISSOC 1995). Nevertheless, there are strong pressures to increase further the part of contribution financing in the area of unemployment compensation. According to a liberal MP “the government is paying too much. I think that the government, employers and employees should contribute equally to unemployment bene12

Organisations approached in Denmark included: the Socialist Party (Socialistik Folkeparti), the Social Democratic Party (Socialdemocratiet), the Radical Liberal Party (Det Radicale Venstre) the Conservative Party (Det Konservative Folkeparti) the Liberal Party (Venstre), the associacion of industrial employers (Dansk Industri), the trade unions FTF (white collars), SID (blue collars) and KAD (blue collars).

10


fits” (int. Venstre, 1994). Similarly, a conservative MP thinks that “the economic responsibility of employers and employees should be expanded” (int. Konservative Folkeparti, 1994). The suggestion that contribution financing should play a bigger role in unemployment benefits is supported mainly by the right, with the approval of employers who would like the “responsibility of the labour market to expand” (int. Dansk Industri, 1994). The left is not opposing a shift towards contribution financing, “as long as we do not change the principles: the benefits must not become dependent on the level of contributions or on the unemployment rate of a given branch” (int. Socialdemocratiet, 1994). Similarly, a trade unionist argued in favour of contribution financing, but stressed that “it is important to make sure that contribution financing is collective and solidaristic” (int. FTF 1994). The Danish left does not seem to be particularly enthusiastic about Bismarckian social insurance, as it is applied in France or Germany. In these countries, the strict relationship between contributions and benefits is a key feature of the system. In a way this is easy to understand, since social insurance runs against left-wing notions of solidarity and redistribution, which are strongly enshrined in Danish social policy. Right-wing support for contribution financing can be understood as a way to reduce comparatively high rates of direct taxation. The development of French or German-like social insurance schemes seems quite unlikely, since only the financing side of it is attracting attention among Danish policy makers, at least as far as the left (in government) is concerned.13 An interesting aspect of this comparison is the difference in attitudes of French and German trade unions on one hand, and their Danish counterparts on the other. While the former are strenuously defending the social insurance system as a major social achievement, the latter regard social insurance as essentially a less solidaristic system compared to what currently exists in Denmark. In this respect, existing arrangements seem to have an impact on the structure of the debate.

13

The current Danish political situation is characterised by a minority government comprising the Social Democrats and the Radical Liberals. In order to get laws passed the government needs the support of at least the Conservative Party. This position of weakness of the current centre-left government, might contribute to explain its acceptance of some moderate departures from the traditional principles of social democracy.


11

Britain: Welfare Away From the State14 One striking feature in the British debate one social protection is the absence of almost any concern on social insurance.15 The British welfare state does nevertheless include a substantial social insurance component. If measured in terms of the proportion of social expenditure financed through contribution, it amounts to 41.9 per cent (Eurostat 1995). In this respect, the British welfare state can be considered as a middle case between continental social insurance and Scandinavian universalism. However, in the case of Britain, it might be misleading to measure the extent of its social insurance component solely in terms of the proportion of social expenditure financed through contributions. The fact is that British social insurance does not follow the same principles which are applied in continental Europe. In particular, there are no separate budgets for the various social insurance schemes. Contributions are treated as taxation and used to finance the general state budget, which is responsible for the payment of benefits. In addition, social insurance cash benefits are flat-rate so that there is no relationship to the income they replace. Finally, the difference in amounts between means-tested and contributory benefits is often negligible, which makes the distinction between the two systems even more difficult to appreciate. The use of unemployment insurance benefits is indeed rather limited, since over 75 per cent of registered unemployed receive assistance benefits. For these reasons, the distinction between social insurance contributions and taxation is not regarded as relevant as it is the case in France or in Germany. The two fiscal instruments are basically seen as interchangeable. As a result, the British debate is characterised not so much by controversy over the part of social insurance in the welfare state but rather by whether social insurance should be maintained at all. In our survey, we were able to identify three main positions on this issue. First, some left-wing respondents view contributions as a less solidaristic means to finance welfare than taxation. Unlike taxation, contributions in Britain are progressive only up to a point in a person's earnings (int. Labour Party, 1994). The abolition of social insurance contributions is advocated also by some right-wing respondents, albeit for different reasons. As a Conservative MP put it “I want simplicity in the tax system, so I can't see any particular 14

Organisations approached in Britain included: the Consevrative Party, the Labour Party, the Liberal Democratic Party, the Confederation of British Industry (CBI) the Trade Union Congress (TUC) and the Transport and General Workers Union (TGWU). 15 For an in depth analysis of current debates on social policy in Britain, see George and Miller (1994).

12


reason to continue having separate National Insurance contributions. It's all going into the same pot” (int. Conservative Party, 1994). Another problem identified in relation to social insurance by some respondents of the business community relates to the fact that contribution financing, since it is related to entitlement, represents an obstacle to further means-testing of cash benefits. According to a representative of the CBI, “it is quite difficult to target [unemployment benefits] because of the structure of National Insurance. It might be easier if you did away with the National Insurance principle and put the unemployed person on Income Support [social assistance]” (int. CBI, 1994). Although in Britain, as noted above, compliance with social insurance principles is much weaker than in other European countries, entitlement to a benefit is guaranteed by the payment of social insurance contributions. In this sense contribution financing is seen as an obstacle to additional targeting. The third position, and by far the one which receives the strongest support, argues for maintaining National Insurance contributions as part of social protection financing. This is the official position of the two leading parties (int. Labour Party 1994). The basic idea is that National Insurance contributions are a less painful way to raise revenue for social policy. The merging of contributions and taxes “would mean putting up the marginal rate of tax, which wouldn't be popular” (int. Labour Party, 1994). Similarly, a Conservative MP, argued in favour of the "insurance link", since “it is good that the individual is taking responsibility” (int. Conservative Party, 1994). On the whole, however, the debate on social insurance is a rather marginal one in Britain, obscured but what are seen as more pressing issues. The balance between public and private provision is arguably one of these.16 In some respects, it seems that the role played by social insurance, i.e., maintenance of one's living standard when unable to work, is fulfilled by private provision. This is increasingly the case with pensions, since after 1986 it is possible to opt out of state and occupational second tier pensions (II Pillar) and buy a personal private pension. It is also the case with unemployment compensation, since the replacement rate is 23 per cent for an average net wage in manufacturing (European Commission 1993), and given the flat-rate nature of the benefit, it is much lower for higher salaries. As a result, private mortgage relief insurance policies in case of unemployment have become rather common in recent years. Typically, such policies guarantee mortgage repayments for people who find themselves unemployed.

16

For an account of political debates and social policy developments in Britain during the 1980s, see Taylor-Gooby (1991).


13

Politics Against Convergence Most analyses of the convergence thesis concentrate either on the socioeconomic pressures which are supposed to bring about convergence (Lash and Urry 1987), or on policy outcomes (Kosonen 1994). In contrast, in this article, we have tried to look at the processes which, in the four countries, can be expected to translate socio-economic pressures into policy outcomes. The political debate is the starting point of policy-making, and by looking at it we can appreciate the range of possible policy options. However, what we are most interested in, are the patterns of coalition formation in relation to a specific issue. Contrary to what one might expect, these patterns seem to differ in the four countries. In France, the trade unions are alone in defending the current social insurance system. The main political parties and the administration agree on the fact that a relatively important shift towards tax-financed universalism is needed. In Germany it is also the unions who seem to be the staunchest supporters of the existing social insurance arrangement. Nevertheless, they can count on a substantial support from important sections of the two main parties, the SPD and the Christian-Democrats. In Denmark it is the right-wing and employers who would like a more insurance based social security system, a move which receives lukewarm support from the Social-Democrats and is not welcomed by the unions. In Britain, finally, there is no actor which can be seen as having a stake in the defence of social insurance, nor in expanding it. Within both major parties there are divisions in this respect. This possibly reflects the fact that the issue is not perceived as a particularly important one if compared to other questions such as the balance between public and private welfare provision. This is a striking result. In four countries experiencing the same sort of socio-economic pressures, the same issue produces four different patterns of coalition formation. These differences can be accounted for only by focusing on the institutional structures which are peculiar to each of the four welfare states and on the long term relationship between the state and organised interests. In France and in Germany the trade unions are involved in the management of the social insurance system. This gives them a powerful reason to support the current arrangement. This is particularly relevant in the case of France, where the lack of legitimacy due to the low level of unionisation is to some extent compensated by the unions' role in social insurance management (Rosanvallon 1995). While France and Germany share a similar welfare state structure, they have a very different history of relationship between the state and the unions. In Germany, in fact, organised interests are typically included in policy-making, while France has traditionally adopted a

14


more static approach, in which policy change is often imposed by the government, paying relatively little attention to the requests of organised interests (as in the case of the Juppé Plan in November 1995). This different pattern of relationship between the state and the unions might explain the different place occupied by political parties in the debate in the two countries. Since government-unions negotiations are more common in Germany than they are in France, in the former, the government, but more in general political actors, are more inclined to take unions' preferences into account. In Britain and in Denmark, in contrast, the trade unions do not play a substantial role in the management of social security, which possibly explains why, unlike their French and German counterparts, Danish and British unions are not keen supporters of the social insurance model. In addition, in Denmark, the left regards social insurance as a less redistributive and less solidaristic arrangement than the current system, hence their lack of enthusiasm for that approach. These variations in political debates are likely to have an impact on actual policy outcomes. It thus seems appropriate to conclude that the impact of convergence oriented socio-economic pressures is significantly mediated by national institutions,17 which produce different debates and different coalitions in the political arena. In addition, with regard to future developments, the impact of socio-economic factors on policy is likely to be significantly mediated by the power relationship between the different coalitions. If convergence has occurred and is likely to occur in the future, it is more because of the combined prevalence of convergence-oriented coalitions in the different countries, rather than because of the impact of uniform socio-economic change. Interestingly, this argument reflects the observation by Montanari (quoted in Kosonen 1994: 106) who has found that between 1960 and 1990 there has been more convergence in social policy among EFTA countries than among EC member-states. In his view, it is the “relatively similar patterns of interplay between political actors that explains convergence, rather than functional requirements” (ibid.). This analysis suggests that the different structure of the debates and the different balance of political forces in the four countries should be understood in terms of the history of the welfare state and of state-unions relation. As Merrien put it “the theoretical challenge is now to find out the historical constraints which are affecting practical developments of policies and States” (1990b: 45). While this task is clearly beyond the scope of this article, the study of interview data gathered from those playing an active role in 17

The hypothesis that institutional factors have an impact on (social) policy outcomes has been defended by a number of authors, from the neo-institutionalist school. See for instance Skocpol and Amenta (1986), Merrien (1990a and 1990b), Immergut (1992) and Pierson (1994).


15

welfare policy debates, offers an insight into the relationship between institutions and political debates. Three main points emerge from our analysis. First, in relation to welfare states, the balance between social insurance and universal provision seems to affect the debate in a significant way. In France and in Germany there is concern for high contributions and social exclusion; in Britain and in Denmark the concern is for high rates of taxation. Second, the place occupied by trade unions in the management of social insurance seems to be related to their support (or lack of support) for the social insurance model. Finally, the existence or the absence of a corporatist tradition in policy-making influences the different patterns on coalition formation observed in France and in Germany. For these reasons, the pressures towards convergence in European social policy are likely to encounter powerful resistance. References ABRAHAMSON, Peter (1991). “Welfare and Poverty in the Europe of the 1990s: Social Progress or Social Dumping?”, International Journal of Health Services 21(2): 237-264. BALDWIN, Peter (1990). The politics of Social Solidarity. Cambridge: Cambridge University Press. CASTEL, Robert (1995a). Les métamorphoses de la question sociale. Paris: Fayard. CASTEL, Robert (1995b). “Elargir l'assiette”, Projet 242: 9-16. CHATAGNER, François (1993). La protection sociale. Paris: Editions Le Monde. CLASEN, Jochen and Arthur GOULD (1995). “Stability and Change in Welfare States: Germany and Sweden in the 1990s”, Policy and Politics 23(3): 189-201. ESPING-ANDERSEN, Gøsta (1990). The Three Worlds of Welfare Capitalism. Cambridge: Polity. EUROPEAN COMMISSION (1993). Social Protection in Europe. Brussels. EUROSTAT (1995). Basic Statistics of the European Communities 1994. Brussels. EUZEBY, Alain (1992). “La protection sociale en Europe: tendences et defis”, Futuribles 12: 59-87. FERRERA, Maurizio (1993a). Modelli di solidarietà. Bologna: Il Mulino. FERRERA, Maurizio (1993b). “Dinamiche di globalizzazione e stato sociale: un introduzione”, in Maurizio FERRERA (ed.). Stato sociale e mercato mondiale. Torino: Fondazione Agnelli, pp. 1 - 21. FLÜCKIGER, Yves et Javier CORDERO (1995). Analyse économique des différentes propositions de réforme du financement des assurances sociales. Genève: Université de Genève, Laboratoire d'Economie Appliquée, No 7. GEORGE, Vic, and Stuart MILLER (eds.) (1994). Social Policy Towards 2000. London: Routledge. GEORGE, Vic and Peter TAYLOR-GOOBY (1996). European Welfare Policy: Squaring the Welfare Circle. London: Macmillan.

16


HANTRAIS, Linda (1995). Social Policy in the European Union. London: Macmillan. IMMERGUT, Ellen (1992). Health Politics. Interests and Institutions in Western Europe. Cambridge: Cambridge University Press. KORPI, Walter (1983). The Democratic Class Struggle. London: Routledge & Kegan. KOSONEN, Pekka (1994). European Integration: A Welfare State Perspective. Helsinki, University of Helsinki. LASH, Scott and John URRY (1987). The End of Organized Capitalism. Cambridge: Polity. LEIBFRIED, Stephan (1992). “Towards a European Welfare State? On Integrating Poverty Regimes into the European Community”, in Zsuzsa FERGE and Jon Eivind KOLBERG (eds.). Social Policy in a Changing Europe. Frankfurt: Campus, pp.245-279. MERRIEN, François-Xavier (1990a). “Etats et politiques sociales: contribution à une théorie néo-institutionnaliste”, Sociologie du travail 3: 267-294. MERRIEN, François-Xavier (1990b). “Etats-providence: l'empreinte des origines”, Revue Française des Affaires Sociales 44(3): 43-56. MISSOC (1995). Social Protection in the Member States of the European Union. Brussels, European Commission, DG V. MYLES, John and Jill QUADAGNO (1996). Recent Trends in Current Pension Reform: A Comparative View. Paper presented at the Conference on Reform of the Retirement Income System, Queen's University, February 1-2. PALIER, Bruno et Giuliano BONOLI (1995). “Entre Bismarck et Beveridge: crises de la sécurité sociales et politique(s)”, Revue Française de Science Politique (45)4: 668-699. PETERSEN, Jon (1991). “Problems of Pension Policy. American, British and German ideas”, Social Policy and Administration 25(3): 249-260. PIERSON, Paul (1994). Dismantling the Welfare State? Reagan, Thatcher and the Politics of Retrenchment. Cambridge: Cambridge University Press. ROSANVALLON, Pierre (1995). La nouvelle question sociale. Repenser l'Etat-providence. Paris: Seuil. ROSSI, Martino and Elena SARTORIS (1995). Ripensare la solidarietà. Locarno: Armando Dadò. SKOCPOL, Theda and Edwin AMENTA (1986). “States and Social Policies”, Annual Review of Sociology 12: 131-157. TAYLOR-GOOBY, Peter (1991). Social Change, Social Welfare and Social Science. Hemel Hempstead: Harvester Wheatsheaf

Politik gegen Konvergenz? Aktuelle Trends in der Sozialpolitik in Europa Die Autoren beschäftigen sich in diesem Artikel mit der Konvergenz der Massnahmen in der Sozialpolitik in EU-Mitgliedstaaten. Verfechter der Konvergenzthese argumentieren, dass die Wohlfahrtsstaaten in Europa, infolge der ähnlichen sozio-ökonomischen Herausforderungen, wahr-


scheinlich auch gemeinsame Antwortstrategien erarbeiten. Ihre Analyse basiert auf einer Studie über Entscheidungsträger und befasst sich näher mit den politischen Debatten über die Zukunft der sozialen Deckung in vier EU - Mitgliedstaaten. Hauptsächlich betrachtet werden die verschiedenen institutionellen Strukturen, welche die Wohlfahrtsstaaten in Europa kennzeichnen und ihr Einfluss auf Koalitionsbildungen bei politischen Debatten. Die Analyse zeigt, dass der Einfluss von sozioökonomischen Veränderungen auf die politischen Debatten wichtig ist, aber dass dieser auf signifikante Weise durch die verschiedenen institutionellen Arrangements, welche europäische Wohlfahrtsstaaten charakterisieren, gemindert wird. Die Autoren kommen zum Schluss, dass gemeinsame sozio-ökonomische Probleme nicht zwingendermassen eine Konvergenz in politischen Massnahmen zur Folge haben.

La politique va-t-elle à l'encontre de la convergence? Tendances actuelles de la politique sociale européenne Dans cet article, les auteurs abordent le problème de la convergence des politiques sociales adoptées par les Etats membres de l'Union européenne. Les tenants de la thèse de la convergence prétendent que les Etats-providence européens étant confrontés à des défis socio-économiques similaires, il y a de fortes chances qu'ils développent des stratégies qui offrent une réponse commune à leurs problèmes. L'analyse en question, basée sur une étude menée auprès de décideurs, porte sur les débats politiques liés à l'avenir de la protection sociale dans quatre Etats membres de l'Union européenne. L'accent principal est mis sur les différences qui existent au niveau des structures institutionnelles des Etats-providence européens et sur leur impact touchant la formation de coalitions au cours des débats politiques. L'étude montre que si les débats politiques sont, certes, fortement influencés par les changements socio-économiques, leurs effets sont toutefois considérablement atténués par les différences qui caractérisent les arrangements institutionnels des Etats-providence européens. Les auteurs arrivent à la conclusion que la présence de problèmes socio-économiques communs n'implique pas forcément une convergence des politiques menées dans les différents pays.

Giuliano BONOLI, Assistant, Institut des Sciences Sociales et Pédagogiques, BFSH 2, Université de Lausanne, 1015 Lausanne; E-mail: [email protected].

17

18


Vic GEORGE, Professor of Social Policy, Department of Social and Public Policy, Darwin College, University of Kent, Canterbury CT2 7NY, United Kingdom; E-mail: [email protected]. Peter TAYLOR-GOOBY, Professor of Social Policy, Department of Social and Public Policy, Darwin College, University of Kent, Canterbury CT2 7NY, United Kingdom; E-mail: [email protected]. Paper submitted 28 November 1995; accepted for publication 4 April 1996.