Religion, Class Coalitions and Welfare State Regimes

3 downloads 117 Views 1MB Size Report
Capoccia, Giovanni. ...... and in anticipation of a major suffrage expansion in 1912, Giolitti proposed in 1910 .... 5 See Giolitti's Discorsi parlamentari, Vol IV, p.
Religion, Class Coalitions and Welfare State Regimes Kees van Kersbergen and Philip Manow (Editors)

2

1

Religion and the Western Welfare State – The Theoretical Context Philip Manow and Kees van Kersbergen 1.1 1.2 1.3 1.4 1.5

Introduction Protestantism, Secularization and the Welfare State The Party Political Correlates of the Countermovement Contributions to the Volume Outlook

4 4 8 18 29 33

2

Western European Party Systems and the Religious Cleavage Thomas Ertman

38

3

The Religious Foundations of Work-Family Policies in Western Europe Kimberly J. Morgan

56

3.1 3.2 3.3

58 64 77

4

Italy: A Christian Democratic or Clientelist Welfare State? Julia Lynch 4.1 4.2 4.3 4.4

5

7

The Christian Democratic Isomorphism of the Italian Welfare State Clientelism and the Maintenance of Occupationalism after World War II Post-War Policy Drift and the Creation of a Familialist Welfare State Conclusion

90 93 103 111 114

Religion and the Welfare State in the Netherlands Kees van Kersbergen

119

5.1 5.2 5.3

119 121

5.4 6

Gendered Welfare Regimes The Religious Origins of Gendered Welfare Regimes Consequences during the Expansion of the Welfare State

Introduction Cleavage Structure, Pillarization and Late Social Policy Development Cross-cleavage and Cross-pillar Coalitions and the Postwar Expansion of the Welfare State Conclusion

129 142

A Conservative Welfare State Regime without Christian Democracy? The French Etat-providence, 1880–1960 Philip Manow and Bruno Palier

147

6.1 6.2 6.3 6.4

147 150 159 171

Introduction Consolidating the Nation through the ‘Laïcisation de la Protection Sociale’ Christian Democracy and the French Post-War Welfare State Conclusion

Religion and the Consolidation of the Swiss Welfare State, 1848–1945 Herbert Obinger

176

7.1 7.2 7.3

176 178

7.4 7.5

Introduction Religion and State-Building Institutional Repercussions: Actor Constellations, Power Resources, and Coalition Building Welfare State Consolidation, 1874–1945 Conclusion

183 193 205

3

8

The Church as Nation? The Role of Religion in the Development of the Swedish Welfare State Karen M. Anderson 8.1 8.2 8.3 8.4 8.5 8.6 8.7

9

10

Introduction Politics and Religion during the Formative Period of the Swedish Welfare State Social Democrats, Liberals and Agrarian Competition in Social Policy: 1900–1932 Secular Ideological Influences on Early Swedish Social Policy Liberal Intellectuals Social Democrats, the Church of Sweden, and Early Social Policy Ideas Discussion and Conclusion

210 210 214 224 225 225 227 228

Religious Conservatives in U.S. Welfare State Politics Jill Quadagno and Deana Rohlinger

234

9.1 9.2 9.3 9.4 9.5

236 239 241 242 258

American Religious Exceptionalism Religion and Social Welfare The Rise of the Christian Right Religious Conservatives in US Welfare Politics Conclusion

Religion as a Cultural Force: Social Doctrines and Poor Relief Traditions Sigrun Kahl

263

10.1 10.1.1 10.1.2 10.1.3 10.2 10.3 10.4

264 266 269 273 279 281 285

Three Traditions of Work, Poverty, and Charity Catholic Poor Relief: Feed the Poor, Get Saved Lutheranism: Bread First, Work Second Reformed Protestantism: Work for Your Own Bread Timing and Principles of Modern Social Assistance Programs Social Doctrines and Church Interests within the Welfare State Conclusion

4

1

Religion and the Western Welfare State – The Theoretical Context*

Philip Manow and Kees van Kersbergen

1.1

Introduction

Most comparativists who study welfare state development agree today that religion has played a role in the development of modern social protection systems. The early protagonists of the power resources approach, however, had only stressed the causal impact of Socialist working class mobilization on modern social policy (see EspingAndersen and Van Kersbergen 1992). In their view it was the working class and its socialist organizations that had been the driving force behind the ‘social democratization’ of capitalism via the welfare state. To them it came as a surprise that not only Social Democracy but also (Social) Catholicism promoted welfare state development. John D. Stephens (1979: 100), one of the leading spokesmen of this approach, put it in prudent terms when he argued that ‘it seemed possible that anti-capitalist aspects of catholic ideology – such as notions of fair wage or prohibitions of usury – as well as the generally positive attitude of the catholic church towards welfare for the poor might encourage government welfare spending’. Similarly, Schmidt (1980, 1982) asserted that Social Democracy and Christian Democracy were functionally equivalent for welfare state expansion, at least during periods of economic prosperity. Wilensky (1981) argued that the two movements overlapped considerably in ideological terms and that Catholicism indeed constituted an even more important determinant of welfare statism than left power did. Catholic social doctrine called for a correction of the most abhorrent societal effects of the capitalist order. The Catholic principle of subsidiarity, moreover, posited that in the last instance the (nation) state had a duty to intervene to correct for morally unacceptable market outcomes. At the centre of the doctrine was not the type of workers’ social rights and emancipation argument that one finds in Social Democratic ideology, but rather the conviction that people have the Christian obligation to help the poor and that social policy can help protect a stable and fair social order.

* We are grateful for extremely helpful comments by Thomas Ertman, Bo Kaspersen, David Leege, Kimberly Morgan and Kenneth Wald.

5

However, it was not only the moral obligations defined in social doctrine and the preoccupation with the problem of social order that determined the pro-welfare stance of religious political parties. Stephens also suggested that there were more straightforward political reasons why Christian Democratic parties were supporters of the welfare state. These parties operated in the political centre, were seeking the working class vote and hoped to cooperate with the Catholic unions. Social policies promised to secure the support of the Catholic working class. Admitting the possibility that other political movements could be attractive to the working class, however, implied that one of the constitutional assumptions of the power resources model had to be relaxed, namely that the political identity attached to wage labour in capitalism is inherently and of necessity Social Democratic. But apparently workers could be mobilized and organized as Catholics, too. Of course, much of the apparent contempt for the continental European welfare state in the comparative literature stems from Marxian notions of a ‘false consciousness’ attached to all forms of political mobilization that do not follow class-lines. Through an elaboration of the power resources approach in Esping-Andersen’s (1990) regime approach and a specification of the association between Christian Democracy and the welfare state (Van Kersbergen 1995), the welfare state literature posited that it was the combination of Christian Democracy and Catholic social doctrine that explained why Christian Democratic welfare states were as generous in terms of social spending as the Scandinavian ones, but were not designed to counter market pressures (to decommodify labour) to the same extent as the Social Democratic welfare states. Christian Democracy-cum-Social Catholicism rather produced and preserved a traditional, patriarchic, status-oriented model of society. It is this reading of the history of the western welfare state that owes much to the power resources and regime approach that is challenged in this book, because it seems at best incomplete. First, an exclusive focus on the labour question and on worker mobilization ignores other highly contentious issues, particularly whether state or society should be responsible for protecting workers, but also mothers and families, against the vagaries of life. And here ‘society’ often had to be read: the church. Church and modern nation state also waged bitter conflicts over who should be the central agent of socialization, as the Dutch ‘schoolstrijd’ or the similar conflicts over education in France exemplify. These conflicts over education exerted a pro-

6

found impact on early welfare state building as well, as the contributions to this volume demonstrate. Second, taking into account not only the capital – labour conflict due to the industrial revolution but also the state – church conflict over education and social policy due to the ‘national revolution’ (Stein Rokkan) is key to our understanding of modern welfare state development. Only in those countries where in the last quarter of the 19h century bitter state – church conflicts were waged did parties of religious defence form. These parties later became so decisive as political actors that mobilized workers and the middle class not along class lines, but along cross-cutting lines of denominational belonging. To explain why parties of religious defence formed in continental Europe but not in Scandinavia or England, then, is – as we argue in this volume – a precondition for a thorough understanding of why the continental welfare state developed in such a different way than the Nordic or Anglo-Saxon welfare regimes. Third, it is not easy to reconcile the historical facts with the power resources and regime narrative about European welfare state development. For instance, it was Liberalism and anti-clericalism rather than Catholicism or Christian Democracy that prevailed in the formative period of the Italian and French welfare states. In fact, in these countries much of the early social legislation had an obvious anti-clerical momentum, since the aim was to establish central national state responsibilities in a domain for which the church had always claimed exclusive competency. Yet, despite their obvious Liberal and anti-clerical pedigree today countries like Italy, Belgium or France are regularly classified as belonging to the Conservative – Catholic welfare state regime type. Therefore, the dominant reading in the literature, which explains the specific features of the continental welfare as a manifestation of Catholic social doctrine, is historically inadequate and blurs the decisive causes for the institutional variance among West-European welfare states. Fourth, an exclusive focus on Catholic social teaching and Christian Democracy also neglects the influence of Protestantism. For instance, a review of historically oriented studies of the social and political role of Protestantism and Protestant political parties led one of us (Van Kersbergen 1995: 254, footnote 1) to reject the idea that Protestantism has had any positive contribution to either Christian Democracy or the welfare state. But this conclusion is only warranted in a limited context, that is when one in-

7

deed focuses on the direct impact of political parties on the emergence and development of social policies and at the same time disregards the differentiation between Lutheran and reformed Protestantism. We have come to the conclusion that a universal statement on the irrelevance of Protestantism for welfare state development is not justified. Historical evidence suggests two things. First, reformed Protestantism substantially delayed and restricted the introduction of modern social policy and therefore had a negative impact on welfare state development. Second, the Lutheran state church in Germany or in the Scandinavian countries held no major reservations against the state playing a dominant role in social protection, at least mounted no substantial resistance against the nation state taking over this new responsibility and even often supported and welcomed this development. Lutheran state churches therefore positively contributed to welfare state development. In our view, the differences between Catholicism and Protestantism and between the major variants of Protestantism are very important for an accurate understanding of the different directions nations went in their social policy development. On the one hand, the Protestant free churches and other reformed currents (Dissenters, Calvinists, Baptists, et cetera) held a strongly anti-étatist position, whereas, on the other hand, Lutheran state churches never questioned the prerogative of the central state in social policy and education. In contrast, when workers in Southern Europe fought for their political and social rights and when Liberals in these countries tried to found modern nation states, they always had to fight against the Catholic clergy as well, which had been closely attached to the ruling elite of the ancien régime. Bitter conflicts between the church and the Liberal elite in the new republican nation states of Southern Europe were the consequence. It is for this reason that Liberal parties in these countries often introduced new social legislation with explicitly anti-clerical motives. This clearly speaks against any unqualified statement that Catholic social doctrine was dominant in the development of the Southern or Continental welfare state. Christian Democratic parties, which to a large extent were the unintended offspring of the church’s political fight against Liberalism, did play an important role, but only much later. Moreover, these parties did not always exactly play the prowelfare state role that the literature imagines, as in the case of the Italian Democrazia Cristiana that used the welfare state primarily as a clientelist resource in its effort to mobilize voters and as a means of becoming more independent from the official church hierarchy.

8

In sum, according to the power resources and regime analysis Protestantism has played no significant role in modern welfare state development and Catholicism did so only insofar as Christian Democracy possessed a Catholic social doctrine and was successful in organizing and mobilizing Catholic workers. In this volume we argue that the impact of political Catholicism and of reformed Protestantism on welfare state development in the western world was quite different from what the literature so far has suggested. The role of religion in the development of the western welfare state is far more than just a variation of the dominant ‘strength of the worker movement’ theme or a question of doctrine influencing policies. In this introductory chapter we start by reviewing how the impact of religion on modern welfare state development had been conceived first in modernization theory – both in its ‘bourgeois’ and its Marxist variants (Section 2). We then continue by positing our own argument of the role of religion in modern welfare state development within the broader political economy literature, which explains the welfare state as the outcome of different political class coalitions (Section 3). After having sketched our own argument, we briefly summarize the contributions to this volume and put them into broader perspective (Section 4). We end by pointing to further implications of our argument for the comparative welfare state literature and for the renewed interest in the role of religion in modern welfare state development (Section 5).

1.2

Protestantism, Secularization and the Welfare State

The issue of the relationship between religion and the development of social policy came up in the context of the theory that pointed out modernization as the root cause of the welfare state. Generally speaking, the origin of the welfare state and its development were largely interpreted as effects of modernization, which encompasses industrialization and democratization. The ‘question’ to which the welfare state was an answer concerned the increasing demands for social and economic equality, that is the demands that completed the Marshallian trias of civil, political and social rights. In catering for such demands, the scope of state intervention was increased tremendously and the nature of the state was transformed:

9

With the structural transformation of the state, the basis of its legitimacy and its functions also change. The objectives of external strength or security, internal economic freedom, and equality before the law are increasingly replaced by a new raison d’être: the provision of secure social services and transfer payments in a standard and routinized way that is not restricted to emergency assistance. (Flora and Heidenheimer 1981: 23)

The increased demand for socioeconomic security came from a system of industrial capitalism that dislodged masses of people and made them dependent on the whims of the labour market, thus rapidly destroying traditional forms of social protection. Welfare state development was related to the problem of social disorder and disintegration that was created by the increasing structural-functional differentiation of modern societies. Such differentiation involves a loosening of ascriptive bonds and a growing mobility of men, goods, and ideas. It leads to the development of extensive networks of exchange and greater disposable resources. As differentiation advances and breaks down traditional forms of social organization, it changes and exacerbates the problem of integration (…). (Flora and Alber 1981: 38)

Modernization caused social disintegration and reinforced the functional requirement of intervention by social organizations and the state. Modernization involved rapidly changing working conditions, the emergence of the free labour contract, the loss of income security among weak groups in the market and through unemployment. The market did not provide the collective goods needed to cope with these problems. At the same time, large parts of the population were mobilized and organized as a consequence of the increasing concentration of people in factories and cities and the extended means of communication. Mobilization was expressed in public protest and violence or in social and political organizations, thus making the spectre of disorder and disintegration directly visible and perceptible for the state elites. In addition, there emerged a pressure generated by the power of organization itself, especially the organization of workers. The causal link between industrialization (or modernization more generally) and welfare state development was not always elaborated well theoretically. The theory that modernization (especially industrialization and its correlates) is the root cause of the welfare state is a functionalist theory that understands the growth of the welfare state in developed nations by and large as the response of the state to the growing needs of its citizens. This theory stresses that industrialization generates demands for

10

social security that can only be met by rational means of state intervention. Similar problems demand similar rational solutions. Consequently, social welfare was seen as a function of industrialization, which created the preconditions for welfare state development in the sense of generating both the need and resources for intervention. It was technical rationality rather than political conflict that governed this response to the transformation of societies. We think that modernization theory is right in constructing a causal link between industrialization and the growing need for social policy, with two crucial modifications: 1) the association must be understood in a somewhat different manner and 2) it is not the only link of causal importance. First, it is not industrialism per se, but the establishment of a full-fledged labour market which links the modern market economy to the welfare state. Second, not only the industrial revolution, but also the national revolution has been of extreme importance for the advent of modern systems of social protection. Here we focus on the first aspect, the labour market – welfare state nexus and inquire into the national revolution as a root cause of modern welfare state formation in section 3. According to the original theory a ‘normal’ development would be that modern social policy, usually conceptualized as social insurance, originated as an effect of the dislocation that industrial capitalism caused. The theory pointed to societal problems that emerge from the wide disruption created by the industrial revolution and the advent of capitalism. However, theoretically the reference to ‘industrialism’ was not always well thought-out. In our view, the root cause of the demand for modern social policy does not lie in industry but in the advance of a full-fledged and self-regulating market on which labour could be bought and sold as a commodity. This, of course, is the single most important characteristic of industrial capitalism as Karl Polanyi (1944 [1957]: 40–41) understood it: (…) once elaborate machines and plant were used for production in a commercial society, the idea of a self-regulating market was bound to take shape (…). Since elaborate machines are expensive, they do not pay unless large amounts of goods are produced. They can be worked without a loss only if the vent of the goods is reasonably assured and if production need not be interrupted for want of the primary goods necessary to feed the machines. For the merchant, this means that all factors involved must be on sale, that is, they must be available in the needed quantities to anybody who is prepared to pay for them. Unless this condition is fulfilled production with the help of specialized machines is too risky to be undertaken both from the point of view of the

11

merchant who stakes his money and of the community as a whole which comes to depend upon continuous production for incomes, employment, and provisions.

Polanyi’s point was that such conditions did not exist in agricultural society, but had to be brought about. The transformation needed was fundamental and required that the economic market logic took over all other social motives – gain rather than subsistence – and all other social institutions. A market economy could only function in a market society. Highly regulated markets as places for trade and barter, of course, existed everywhere and since long, but a self-regulating market system was a new and unique phenomenon. Self-regulating implies that all production is for sale on the market and that all incomes derive from such sales. Accordingly, there are markets for all elements of industry, not only for goods (always including services), but also for labor, land, and money (…). (Polanyi 1944 [1957]: 69)

This implied that land, money and labour were assumed to be produced for sale, that they could be bought and sold, that is, that they functioned as commodities. Polanyi stressed that labour was obviously not produced for the sole purpose of selling. It was a fictitious commodity; it was forced to function as if it were a commodity. It was therefore also subject to the forces of supply and demand, embodied in the price mechanism, and nothing (and especially not state intervention) ought to prohibit its functioning as a commodity. But labour could simply not function as a commodity, because if it really did so on an unrestrained market, it would destroy society. For the alleged commodity ‘labor power’ cannot be shoved about, used indiscriminately, or even left unused, without also affecting the human individual who happens to be the bearer of this peculiar commodity. In disposing of a man’s labor power the system would, incidentally, dispose of the physical, psychological, and moral entity ‘man’ attached to that tag. Robbed of the protective covering of cultural institutions, human beings would perish from the effects of social exposure; they would die as the victims of acute social dislocation through vice, perversion, crime, and starvation’ (Polanyi 1944 [1957]: 73). Labour had to be adapted to the demands of the market system. And this implied a complete reorganization of society itself, as a result of which human society became

12

an ‘accessory of the economic system’ (Polanyi 1944 [1957]: 75). The result were disastrous and ‘human society would have been annihilated but for protective countermoves which blunted the action of this self-destructive mechanism’ (idem). Although Polanyi nowhere used the words, his thesis is, of course, that the commodification of labour was necessarily followed by its decommodification. This, then, is the ‘double movement’ that Polanyi identified as being so characteristic for the social history of the nineteenth century: ‘society protected itself against the perils inherent in a selfregulating market system (…)’ (Polanyi 1944 [1957]: 76). Following Polanyi, it is the timing of the industrial transformation and the development of a self-regulating market for industrial labour (the first movement) which determines whether the countermovement in the form of social protection arises in a country. We can understand nineteenth century developments in social legislation in terms of the creation of a self-regulating market, the social dislocation this caused and the counteraction in the form of social protection this provoked. This is the common experience of all countries in which the self-regulating market did its destructive work. This is Polanyi’s brilliant analysis of the double movement of commodification, also of labour power, under capitalist markets and of decommodification as the inevitable response to this. In fact, Polanyi (1957 [1944]: 147) made a historical-comparative statement to this effect: Victorian England and the Prussia of Bismarck were poles apart and both were very much unlike the France of the Third Republic or the Empire of the Habsburgs. Yet each of them passed through a period of free trade and laissez-faire, followed by a period of antiliberal legislation in regard to public health, factory conditions, municipal trading, social insurance, shipping subsidies, public utilities, trade associations, and so on. It would be easy to produce a regular calendar setting out the years in which analogous changes occurred in the various years (…).

The supporting forces were in some cases violently reactionary and antisocialist as in Vienna, at other times ‘radical imperialist’ as in Birmingham, or of the purest liberal hue as with the Frenchman, Edouard Herriot, Mayor of Lyons. In Protestant England, Conservative and Liberal cabinets labored intermittently at the completion of factory legislation. In Germany, Roman Catholics and Social Democrats took part in its achievement; in Austria, the Church and its most militant supporters; in France, enemies of the Church and ardent anticlericals were responsible for the enactment of almost identical laws. Thus under the most varied slogans, with very different moti-

13

vations a multitude of parties and social strata put into effect almost exactly the same measures in a series of countries in respect to a large number of complicated subjects. Polanyi’s theory of the double movement explains why all nations, irrespective of regime-type or the political-ideological leanings of the ruling elites, developed social protection against the social disruption caused by the creation of the capitalist labour market. However, neither the theory of industrialism nor Polanyi’s theory of the double movement appear satisfactory when it comes to explaining the varying forms the welfare state countermovement took. Their incapability to explain why the state responses to the challenges posed by the full ‘marketization’ of moderns societies varied so much is due to their lack of a micro-foundation, since they never detail how exactly pressures of social misery and dislocation were actually translated into some social policy response or ‘the’ welfare state. In both the modernization approach and the Polanyian explanation it remains obscure how needs and demands can create their own fulfilment. There is little or no account of the causal mechanisms or the societal actors producing the political responses. This is why – implicitly or explicitly – they are also theories of convergence, arguing that societies were increasingly becoming alike as they approached a certain level of industrial development and as they developed a self-regulating labour market. Therefore both theories have little to say empirically about cross-national variation among developed nations, why the responses to the social disruptions caused by modern capitalism were so different from each other with such long lasting socio-economic effects for these societies. In our view, it mattered a great deal who exactly executed, so the speak, the countermovement. It is here that political actors like parties as the organized expression of social cleavage structures gain relevance. Among those cleavages that have structured the party systems of Western Europe, the religious one has been of particular importance. This cleavage has been the result of state – church conflicts in the wake of the national revolution when state building elites challenged the position of the church in domains perceived crucial for the creation of modern nation states, particularly education, but also social protection. Of course, the 1970s generation of welfare state researchers, inspired by the (functionalist) theory of modernization and masterly represented by Peter Flora, already claimed that religion influenced modern

14

welfare state development, but again they rather emphasized the role of religion as a structural factor and in the longue durée. Modernization theory cherished the claim that secularization (as a correlate of industrialization and urbanization or as a phenomenon of modernization in general) and Protestantism were a source of welfare state development. It was the decline of religion, the impact of Protestantism and the rise of the secular nation state – as a consequence of the ‘surrender’ of the church to the state or as a result of the retreat of the church into the ‘private’ realm – that governed the development of the welfare state. Secularization was taken to refer to the decline of the categorical impact of religion on human conduct (Chadwick 1975) and pointed to the increasing powerlessness of organized religion in temporal affairs, coupled with the decreasing plausibility of the religious interpretation of the world (Martin 1978). Or in Bruce’s (2002: 3) summary of the secularization paradigm: a) the declining importance of religion for the operation of non-religious roles and institutions such as those of the state and the economy; b) a decline in the social standing of religious roles and institutions; and c) a decline in the extent to which people engage in religious practices, display beliefs of a religious kind, and conduct other aspects of their lives in a manner informed by such beliefs. Religious institutions lost their dominance over society and culture and Protestantism ‘served as the historically decisive prelude to secularization’ (Berger 1990: 113). Christianity was thought to be related to welfare capitalism only so far as Protestantism involved a first step in the process of secularization and individualization, because this belief caused ‘an immense shrinkage in the scope of the sacred in reality’ (Berger 1990: 111). Protestantism qualitatively changed church – state relationships which, in turn, facilitated the construction of the welfare state. Protestantism and secularization influenced the transformation of traditional societies into mass democracies and this process affected the institutional arrangements of modern welfare states. Thus Flora’s generalization of the Rokkan macro-model of European history asserted that

15

the protestant nationalization of the territorial culture in the North favored the mobilization of voice ‘from below’: the early development of literacy encouraged the mobilization of lower strata into mass politics, and the incorporation of the church into the state apparatus reduced one potential source of conflict and produced a clear-cut focus for the opposition of the dominated population. By contrast, the supra-territorial influence of the catholic church favored a mobilization ‘from above’: the late development of literacy retarded spontaneous mass mobilization and the conflicts over the control over the educational system led to efforts by the church to mobilize against the state. (Flora 1983: 22)

In nations in which the reformation had a lasting impact and in which state – church relations gradually developed, the conditions for collective welfare services were argued to be most favourable, the more so as the decline of religion was believed to facilitate the growing political salience of class. In contrast to the argument that the power of (Calvinist) Protestantism delayed welfare state development (cf. Manow 2004), the modernization scholars were convinced that in those nations where Catholicism continued to shape culture and politics, the conflict between state and church inhibited or at least retarded the emergence of a welfare state. This contrast between the Protestant and the Catholic nations was then taken to explain the difference in timing and variation in the quality of the welfare states. These latter differences concerned the degree of ‘stateness’ (the level of centralization; the level of state – church integration; the degree of state intervention in the economy) and the degree of institutional coherence (universalism versus fragmentation). The ideal type of the welfare state that Flora and others at the time had in mind referred to a historical combination of universalism and stateness and was taken to comprise characteristics such as political centralization, nationalization of the church, cultural homogeneity, advanced agriculture, lack of, or at least, limited absolutism, smooth democratization and a limited division of state and society. This ideal type of welfare state development was found in Europe’s periphery, that is to say in Scandinavia, where the physical distance from Rome was greatest. In other words, what Peter Flora discovered as the differentia specifica of the ‘Protestant welfare state’, was later rediscovered by Walter Korpi and his followers (notably Gøsta EspingAndersen) as the ‘Social Democratic welfare state’. In an important article, Heidenheimer (1983) focused on the relationship between religion and secularization patterns on the one hand and what he called the ‘westward spread’ of the welfare state on the other. His article was in the form of two

16

imaginary dialogues, the first between Max Weber and Ernst Troeltsch, taking place in 1904, and the second between Ernst Reweb and Max Schroeltt, the modern impersonations of the two great sociologists. The birth of the welfare state was dated 1883, the year of the introduction of the first workers’ insurance in Bismarckian Germany. The question to Weber and Troeltsch was ‘whether the spread of social insurance is at all related to the religious ethos prevalent in different countries and if so, how do the different branches of christianity compare in the degree to which they have welcomed or opposed this trend?’ (Heidenheimer 1983: 6). In the Weberian perspective it was expected that the adoption of social insurance occurred in an early stage in Protestant countries because both doctrine and the intimate relationship between state and church were favourable to paternalist types of social policies. Catholic countries were expected to be laggards because Catholicism inhibited economic development. In the account of Troeltsch, on the other hand, there was a crucial difference between Calvinist and Lutheran countries, the latter probably more willing to accept social insurance as a tolerable intervention, the former, because of the association with liberal capitalism, probably even slower in their embracing of the welfare state than the Catholic nations. Secularization patterns were assumed to affect national experiences by altering the velocity with which social insurance schemes were introduced (cf. Manow 2004). In the Weberian perspective secularization was assumed to accelerate the development of the welfare state because it was viewed as a concluding phase of western rationalization. According to the Troeltschian interpretation, however, secularization was a multi-dimensional process by which religion lost its influence entirely in some domains, whereas in other domains traditional values were transplanted into secular structures and processes (Heidenheimer 1983: 9). The link between religion and the welfare state that Van Kersbergen (1995) advanced consisted of four theses: 1) Christian Democracy is the heir of the (social) Catholic parties that mobilized roughly between 1870 and 1914; 2) Christian Democracy fostered a distinctive welfare state regime labelled ‘social capitalism’; 3) this regime was both the medium and outcome of Christian Democratic power mobilization through the ‘politics of mediation’; and 4) the specific configuration of interests (including denominational interests) representation and accommodation within Christian De-

17

mocratic parties explained different outcomes in terms of social policy performance, that is within regime variations. The explanatory problem did not concern the possible association between religion(s) and welfare state emergence, timing or development. Rather the question was to what extent an institutionally distinct welfare state regime could be explained in terms of the impact of Christian Democratic power mobilization and to what extent the successful founding of ‘social capitalism’ implied increasing returns for Christian Democratic power mobilization. In the wake of this argument, the thesis of the irrelevance of Protestantism came to refer to the issue of the distinctive regime and to the minimal contribution of Protestants to Christian Democracy. The main argument was based on the conviction that it was in the social motivation or concern that we find not only the origins of the distinctiveness of Christian Democracy (especially vis-à-vis conservatism), but also the root cause of the movements’ success in terms of electoral competition and policy performance. The religious inspiration of Christian Democratic parties distinguished them from conservative or secular centre parties, but also from Liberal and Social Democratic parties (Van Kersbergen 1994; 1995; 1999). The social concern of Christian Democratic parties was related directly to its main historical ancestor, Social Catholicism. Even where the Protestant influence was discernible (Germany, the Netherlands), Catholic social teaching was a prominent property of the political ideology of Christian Democratic parties and through them impinged on the welfare state. To sum up: in contrast to Polanyi we argue that it did matter who politically executed ‘the countermovement’. And following Flora, Heidenheimer and others we emphasize the importance of religious cleavages when it comes to the question which kind of class or party coalitions were able to formulate the welfare state response to the Great Transformation. In the following paragraph we outline two possible narratives of how political class coalitions have shaped the welfare state response to the challenges posed by the industrial revolution. We then continue by developing the outlines of our own cleavage theory of political class coalitions, emphasizing the importance of the religious cleavage in continental Europe as well as the importance of the absence of a religious cleavage in Scandinavia for the diverging paths of welfare state formation and development in these two regions. In this introductory chapter, we do not spell out in detail which kind of institutional consequences the different party

18

political coalitions behind the welfare state in continental Europe and in Scandinavia had. This is, among other things, the task of the various contributions to this volume, which will be briefly introduced at the end of this chapter.

1.3

The Party Political Correlates of the Countermovement

It is important to account for the kind of political coalitions that were behind the national formulations of the Polanyian countermovement. For this reason we look at national party systems and how they have led to different strengths of the left and the right. The power resources approach in the comparative welfare state literature argues that the welfare state was a project of the left, of Social Democratic parties and unions. Where the left was strong, the welfare state became generous and encompassing. Where the left was weak, the welfare state remained residual. Yet, Social Democracy nowhere was able to achieve an electoral majority on its own (Przeworski and Sprague 1986). The left always remained dependent on coalition partners who would join them in their struggle for more social justice and equality, for workers’ better living conditions, and for the decommodification of labour. Therefore, ‘the history of political class coalitions [is] the most decisive cause of welfare state variations’ (Esping-Andersen 1990: 1). In the first chapter of his seminal The Three Worlds of Welfare Capitalism EspingAndersen (1990) himself sets out to provide us with such a stylized history of political class coalitions. He highlights three elements of such an account: the ‘nature of class mobilization (especially of the working class); class political coalition structures; and the historical legacy of regime institutionalization’ (Esping-Andersen 1990: 29). With respect to the mobilization of the working class, he is quick to add that working class strength itself does not help explaining much of the history of welfare state development: ‘It is a historical fact that welfare state construction has depended on political coalition-building. The structure of class-coalitions is much more decisive than are the power resources of any single class’ (p. 30). He then goes on to stress the importance of the pro-welfare state support coalition between Social Democracy and agrarian parties in the Nordic countries, a coalition which could be expanded after World War II to include the middle class. Esping-Andersen explains the specific suc-

19

cess of Swedish Social Democracy in building a generous welfare state with the fact that it succeeded to broaden the political support for a new kind of welfare state that ‘provided benefits tailored to the tastes and expectations of the middle classes’ (31). Without pointing out possible causes, Esping-Andersen goes on to state that in Anglo-Saxon countries ‘the new middle classes were not wooed from the market to the state’ and therefore these countries ‘retained the residual welfare state model’ (31). ‘In class terms, the consequence is dualism. The welfare state caters essentially to the working class and the poor. Private insurance and occupational fringe benefits cater to the middle classes’ (31). The continental welfare states, finally, also depended on the support of the middle classes, but out of ‘historical’ (ibid.) – again not particularly well explained – reasons the outcome was different. ‘Developed by conservative political forces, these regimes institutionalized a middle-class loyalty to the preservation of both occupationally segregated social-insurance programs and, ultimately, to the political forces that brought them into being’ (Esping-Andersen 1990: 31–32). As valuable as the distinction of the three regimes has proven for the comparative analysis of welfare states and as critical as the emphasis on the importance of political class coalitions is, Esping-Andersen does not provide us with an explanation why these groups of countries followed such different institutional trajectories of development. His account rather ‘comes across as a post-hoc description’ (Iversen 2006: 609), in particular since Esping-Andersen gives no systematic reasons why some welfare states were able to include the middle class while others were not. In a recent paper Torben Iversen and David Soskice (2006) have proposed a different version of a political class coalition theory of welfare state formation and growth. They start from the basic observation that in multi-party systems the left is in government more often whereas the right more often governs in two-party systems. Why is this so? At the risk of oversimplification their argument may be summarized as follows. Their model assumes three classes – the lower, middle and upper classes – and a system of (nonregressive) taxation and redistribution. Their model’s key variable is the electoral system. With majoritarian electoral rules, a two-party system emerges with a centre-left and a centre-right party, with proportional representation a multi-party system emerges. In a two-party system the middle class can either vote for a centre-left party or a centre-right party. If the left governs, the middle class has to fear that the left government will tax both the upper and the middle class for the

20

exclusive benefit of the lower class. If a right party governs, the middle and upper class will not be taxed and redistribution will be marginal. Therefore, in a two-party system the middle class has the choice of either to receive no benefits but to be taxed, if the left governs, or to receive no benefits but then also not to be taxed, in case that the right should govern. Obviously, it would then prefer not to be taxed. In a multiparty system, however, the middle class’s choice is different. If the middle class party enters into a coalition with the left party, the lower and middle class can tax the rich and divide the revenue. From this simple and highly stylized account it is clear that the middle class will more often vote together with the lower class in multi-party systems. To be more precise: middle class parties will more often enter into coalitions with lower class parties in multi-party systems than in two-party systems, the left will be in government more often, redistribution will be higher, the welfare state more generous.1 As in Esping-Andersen’s sketch of different welfare state coalitions, Iversen and Soskice also develop a class-coalitional approach, but they provide us with a clearer mechanism that explains the formation of different class coalitions. The authors stress electoral rules as the most important mechanism, as these either lead to multi-party systems-cum-generous welfare state or two-party systems-cum-residual welfare state. The critical analytical distinction therefore is the distinction between plurality and proportional representation (PR). In essence, Iversen and Soskice provide us with a very elegant explanation for the well-known fact that the left more often governs in countries with PR systems (and it is here the welfare state tends to be bigger and more redistributive), whereas the right more often governs in countries with majoritarian electoral rules – countries in which the welfare state tends to be less generous and more residual (see Table 1; Iversen and Soskice 2006: 166).

1 The crucial assumption in the Iversen/Soskice model is that centre-left parties in a two party system cannot credibly commit themselves to a political agenda tailored to median voter interests. Problems of credible commitment are more easily solved in coalition governments. There is ample evidence that indeed a ‘coalition of parties’ functions differently than ‘coalition parties’ (Bawn and Rosenbluth 2003).

21

Table 1 Electoral Systems and the Number of Years with Left and Right Governments 2 (1945–1998) Government Partisanship

Electoral System

Proportion of Right Governments

Left

Right

Proportional

342 (8)

120 (1)

0.26

Majoritarian

86 (0)

256 (8)

0.75

Number in parenthesis: number of countries that have an overweight (more than 50 %) of centre-left or centre-right governments during the 1945–1998 period.

But ever since the contribution of Esping-Andersen we know that the level of spending per se is not what should interest us most, but rather the profound differences in the institutional setup of European welfare states with their substantially differing socio-economic effects. However, is there a way to combine the class-coalition model of Iversen and Soskice with the three regimes approach of Esping-Andersen? Yes, there is such a link or nexus, but it is one which becomes apparent only once we give due credit to the importance of societal cleavage structures and especially the religious cleavage.3 Our argument runs as follows. We agree with Duverger and Iversen and Soskice that majoritarian electoral rules lead to a two-party system and that here the middle class more often votes for conservative parties. In such a two-party system mainly one societal cleavage is present, namely the one dominant in all advanced industrial countries, the left – right or labour – capital cleavage. All other cleavages are ab-

2 Whether ‘years in government’ is a good measurement for the effects of electoral rules remains an open question. Since one of the effects of PR in Scandinavia is the frequent occurrence of minority governments, where parties supporting the government nonetheless can have substantial policy influence, the measure might underestimate the importance of informal party coalitions between left and centre parties. Therefore, also vote shares can be considered when assessing the potential for policy influence of different parties, which, however, does not alter the picture substantially (see Manow 2006). 3 Korpi (2006) recognizes the importance of the PR electoral system for the strength of the left, but does not reach the conclusion we present here on the translation of cleavage structures and societal interest representation and their impact on welfare state development. He replicates the power resources thesis that ‘European confessional parties were organized as explicitly cross-class parties intended to undercut the class base of socialist parties and compete for workers’ votes by strengthening the role of religious cleavages. In association with the Catholic Church, they supported the state-corporatist institutional model of welfare states’ (Korpi 2006: 194).

22

sorbed, latent or ‘incorporated’ in this basic cleavage.4 A good example would be the fierce conflict between the Anglican high church and the Protestant dissent in Britain in the last quarter of the 19th century. This was a virulent conflict line between the Tories and the Liberal Party (Parry 1986), but quickly receded into the background once the Labour Party crowded out the Liberal Party in the last decade of the 19th century and the first two decades of the 20th century. The religious dissent then lost its strong own political representation in the party system, the liberal party split and subsequently Nonconformism became influential both within the Tories and within the Labour party, where it strongly influenced Labour’s social policy program (see Pelling 1965; Caterall 1993). It is here where the basic mechanism described by Iversen and Soskice applies: with only two parties it’s the economic cleavage which is predominantly represented in the party system, and within this setting the middle class more often votes for Conservative parties. The welfare state remains residual. In PR systems, in contrast, a larger (effective) number of parties allow for the representation of more than the one dominant cleavage dimension in advanced industrialized countries, that is more than the conventional labour – capital cleavage (Neto and Cox 1997; Clark and Golder 2006). Which kind of additional cleavages are represented in the party system depends on the cleavage structure of the country in question. Here the distinction between the Nordic and the Continental countries and their welfare states achieves particular relevance. In the north of Europe a religious cleavage did not become politicized and ‘particized’ (see Stoll 2005) because neither were these societies religiously heterogeneous nor did the ‘national revolution’ (Rokkan) lead to strong state – church conflicts. ‘All the Nordic countries belong to (and, indeed, collectively constitute) Europe’s sole mono-confessional Protestant region’ (Madeley 2000: 29). The northern Protestant churches as Lutheran state churches, in contrast to the Catholic church in southern Europe, did not feel fundamentally challenged when the new nation state started to take over responsibilities which previously had fallen under the responsibility of the church. Anti-clericalism never became a strong political current in the Scandinavian countries. A cleavage that did

4 The analysis of the impact of religion does not become irrelevant in this context, however, since it is of interest how the religious cleavage has played out itself within the dominant left – right divide. The role of Christian Socialism and Nonconformism within the British Labour party and the substantial influence of these currents on the social policy program of Labour is an example.

23

become politicized and particized was the cleavage between agrarian and industrial interests, since the agrarian sector was still very strong at the moment of mass democratization in the late industrializing Scandinavian countries. It is in Europe’s north where strong parties of agrarian defence emerged and where they received a substantial share of the votes over the entire post-war period. The Finnish Agrarian Union (Malaisliitto), renamed Centre Party in 1965, won between 21 and 24 percent of the vote in all elections between 1945 and 1970. Even in the 1970s and 1980s the Centre Party never received less than 17 percent of the vote. In the general elections of 1991 the agrarians even became the biggest party with a vote share of almost 25 percent, three percent more than the Social Democrats (see Caramani 2000: 275–289). Electorally less successful, but still with an impressive electoral record, was the Swedish agrarian party, the Bondeförbunet, renamed Centerpartiet in 1957. It gained between 12 and 16 percent of all votes in the 1950s and 1960s, and then even increased its vote share substantially in the 1970s, gaining up to 25 percent thereby becoming the second largest party behind the Social Democrats. In the 1980s and 1990s, the agrarian party then again lost much of its former strength and had a vote share of 15 percent in the early 1980s and went down to 5 percent in the late 1990s. The Norwegian Bondepartiet (since 1961 Senterpartiet), smaller than its Finnish or Swedish counterparts, received around 9 percent on average throughout the 1950s and 1960s, had slightly more than 10 percent of the vote in the 1970s, and then fell back a bit and remained between 6 and 8 percent in subsequent elections (with the exception of the 1993 election where the agrarian party got 16.7 percent and was the third largest party in the Storting) (see Caramani 2000: 762–775). In Denmark the agrarian vote was first very much concentrated in the Liberal party, the Venstre or ‘Agrarian Liberals’ (Johansen 1986: 351), established in 1870 as ‘a derivation of the Bondevennerne, the peasants’ friends’ (Caramani 2000: 204). However, it was not the Venstre, but a split from the Venstre/Liberals, det Radicale Venstre established in 1905, which became part of the Danish red-green pro-welfare coalition. Whereas the Liberals represented the interest of large farms and were in strong opposition to the Social Democrats, the radical liberals defended small farm interests (Huber and Stephens 2001: 141) created by a land reform in 1920. Over the entire postwar period, the radical liberals gained on average 7.3 percent of the vote, with particular electoral

24

success in the late 1960s and early 1970s (with almost 15 percent in 1968). Over time this party lost its agrarian affiliation, since small farms disappeared due to their lack of competitiveness, and the party looked for a new constituency. In sum, the distinguishing feature of the Scandinavian party systems is the strong role that agrarian parties play in them. Over the entire post-war period from 1945 to 1999 agrarian parties in Finland, Norway, Sweden, and Denmark gained on average 20.6, 13.9, 8.9 and 7.31 percent of the vote.5 No comparable figures can be found in any other European party system except in Switzerland, where the Schweizerische Bauern-, Gewerbe- und Bürgerpartei (since 1971 Schweizerische Volkspartei) gained more than 10 percent of the votes in each election in the post-war period (cf. Caramani 2004: 181).6 In all other European countries ‘the urban – rural cleavage was incorporated into other party alignments – state – church and left – right in particular – and did not give rise to specific political parties’ (Caramani 2004: 184). Given the strong position of the agrarian parties, it comes as no surprise that almost all accounts of the historical development of the Nordic welfare state stress the importance of red – green coalitions for the formation and the subsequent expansion of the welfare state (see Olson 1986: 5, 75; Esping-Andersen 1990: 30) or see agrarian parties even as the driving force of early welfare state development (Baldwin 1990: 55–94). The influence of the agrarian or centre parties was due to their pivotal position within the Scandinavian party systems. When not itself a part of the government coalition, centre parties tolerated the minority governments often led by Social Democrats, especially in Norway and Sweden (see Narud and Strom 2000; Bergman 2000). Social legislation depended on their consent and therefore was tailored to the agrarian needs and interests. One could say that the place occupied by agrarian parties in the north is occupied by Christian Democratic parties on the continent. The German CDU, the Dutch CDA, the Austrian ÖVP, the Belgian CVP/PSC (Parti Social-Chrétien or Christelijke Volkspartij), the Italian Democrazia Cristiana or the Swiss Christlich-Demokratische

5 The Venstre party gained on average 18.5 percent over the entire postwar period (see Caramani 2000). 6 The Bauern-, Gewerbe- und Bürgerpartei, however, is much more regionally concentrated and much less nationally dispersed than the Nordic agrarian parties.

25

Volkspartei are parties with their roots in political Catholicism. They are the offspring of the fierce state – church conflicts in the last quarter of the 19th and the first quarter of the 20th century – in the Belgian case offspring of the national independence movement of the Catholic southern provinces against the Protestant northern provinces of the Low Countries. In other words, also in these countries (which all introduced PR no later than 1919) we have a relatively high (effective) number of parties and, subsequently, more than one cleavage is represented in the party system. However, instead of the urban – rural cleavage, which was prominent in the religiously homogenous Protestant north, here in the religiously mixed or homogenously Catholic countries (like Belgium and Austria) the state – church conflict is prominently represented in the party system in addition to the left – right cleavage. Both with respect to vote shares and with respect to time in government, the Christian Democratic parties have been dominant (if not hegemonic like the Italian Democrazia Cristiana) in the continental countries. Since Christian Democratic parties combined the religious and large parts of the bourgeois vote, their electoral fate was better than that of the Nordic agrarian parties. Only counting the vote shares of the Catholic parties like the Österreichische Volkspartei, the Christlich-Demokratische Union, the Democrazia Cristiana, the Christen Demokratisch Appèl (and its former member parties like the Katholieke Volkspartij)7 and the Parti Social-Chrétien or Christelijke Volkspartij of Belgium (and ignoring the Protestant parties in these countries for a moment), it becomes evident that Christian Democracy was much more successful than the agrarian parties in Scandinavia. On average, the Belgian PSC or CVP received 34.9 percent of the vote in all elections that took place between 1945 and 1999, the German Christlich-Demokratische Union gained on average of 44.2 percent of the vote during this period, the Katholieke Volkspartij (and later the Appèl) gained on average 28.6 percent of the vote in the same period, the Austrian Volkspartei received 41.5 percent, and the Italian DC received 33.8 percent of the vote. And if we do not count the elections after the breakdown of the first Italian republic – that is if we discard the elections after 1992 – than this share rises even to

7 Since part of the joined Christen-Democratisch Appèl were the Anti-Revolutionary Party and the Christlijk-Historische Unie, since 1977/1980 we cannot speak of a purely Catholic party. Numbers from before 1977 however refer to ‘purely’ Catholic parties, mainly the Katholieke Volkspartij.

26

37.9 percent. The Swiss Christlich-Demokratische Volkspartei won around 20.9 percent of the vote. At the same time, parties of agrarian defence remained largely absent in continental Europe. The urban – rural cleavage dimension remained latent and was not politicized and particized in the continental welfare states. Put pointedly, according to our history of political class coalitions we find liberal welfare states in countries with a majoritarian electoral system in which only one political cleavage dimension is present (exemplary case: the UK). The Social Democratic generous welfare states, which we find in the Nordic countries, however, have been the result of a coalition between Social Democratic parties and parties of agrarian defence (red-green coalition). One important precondition for this coalition has been the absence of a strong religious cleavage in the Scandinavian countries. On Europe’s continent, in turn, we find welfare states that are the product of a coalition between Social and Christian Democracy (red-black coalition). This is due to the fact that the second cleavage represented in the party systems of continental Europe, besides the dominant left – right or labour – capital cleavage, has been the religious cleavage, a cleavage inherited from the state – church conflicts in the wake of the national revolution, in which Liberal state elites challenged the church in its former domains like education or poor relief. 8 What we propose here, in other words, is a Rokkanian complement to the Iversen and Soskice model of welfare state class coalitions. In our view Iversen and Soskice are perfectly right in stressing the importance of a class coalition between lower and middle class, but once we look at the party political coalitions behind the Nordic and the continental welfare states, we are also able to identify which type of middle class has entered into a coalition with Social Democracy. This insight will also allow us to explain the type of welfare state to which these party political class coalitions have led. Similarly, the variation in Christian Democracy on

8 Within this broader picture France is an outlier case. It shares many features with the social insurance states of other continental European nations. However, in France a Christian Democratic party did not emerge in the critical period of mass democratization (Kalyvas 1996), and then, after its late foundation (MRP in the year 1944), proved to be only very short-lived. If one considers the continental or conservative welfare states to be a product of a coalition between left Social Democrats and centre-right Christian Democrats, then the French case obviously does not fit particularly well into this broader picture (see for an explanation of the French case, Manow and Palier, this volume). The other obvious outlier – Switzerland – is explained by its strong federalism, which itself was a way to pacify a fierce religious conflict, see Obinger’s contribution to this volume. We therefore put the Swiss case aside for the moment.

27

the European continent – sometimes being hegemonic like in the case of the Italian DC, sometimes having vanished over the course of the post-war years like the French Mouvement Républicain Populaire (MRP) – allows us to address systematically the question of the within-type variation in the case of the Conservative, Christian Democratic welfare state predominant on the continent. To be clear, there have been and still are religious parties in the Nordic countries, such as the most important and electorally successful Norwegian Kristelig Folkeparti founded in 1933,9 and there have been agrarian parties in continental Europe. But both remained marginal, without political influence and impact, especially when it comes to welfare state formation and growth. We also would like to emphasize that we do not claim that the PR electoral systems can explain the formation of either agrarian parties or parties of religious defence. In numerous cases these parties were founded before the introduction of PR, which in most countries occurred only after World War I. However, in contrast to the simple majority system in force in Britain, the two-ballot or two-round majority system in place before the Great War in almost all other European countries has the non-Duvergian tendency to sustain a higher effective number of parties, despite its ‘Duvergian drive’ towards the formation of two blocks or party camps. This was a crucial difference between Britain and the rest of Western Europe in the period of suffrage extension. In this introductory chapter we do not deal with the many institutional consequences which the different welfare state coalitions in Europe’s north and on its continent had. We restrict ourselves to simply mentioning some of the more important ones, whereas the chapters that follow will provide the reader with much richer empirical accounts of the institutional variations between the Nordic and continental regime types. One important difference, however, is evident from the outset: agrarian parties in the north voiced resistance against the income differentiated social benefits that Social Democrats favoured (cf. Olson 1986; Johansen 1986). Instead they preferred universalist, flat rate benefits since many small landholders had no long histories of steady income and therefore feared that they would be actually unable to benefit

9 The Finnish Christian League, Suomen Kristillinen Liitto, was founded in 1958; the Swedish Kristen Demokratisk Samling was founded 1964; the Danish Kristeligt Folkeparti was founded in 1970.

28

from welfare entitlements which were contribution financed with contribution related benefit levels (cf. Baldwin 1990: 55–94). Christian Democratic parties, on the other hand, which mobilized workers as did their Social Democratic counterparts, had far less reasons to object to differentiated, wage-based contributions and entitlements. Social insurance contributions promised to ease the party-internal conflict over social policy between Catholic workers and the middle-class, since contributions ‘naturally’ seemed to limit the extent of welfare state redistribution. Moreover, during the severe economic crisis of the late 1920s and early 1930s, which had particularly affected Scandinavian agriculture, many farmers became less interested in unemployment payments, which would bridge the spells without work. Rather, they demanded active labour market policies to ease the transition to the second and third sector. The problem was perceived as one of sectoral change, in which the loss of employment in the first sector had to be compensated through employment growth in industry and services. Other important differences were the integration of the churches in the continental welfare states in the provision of social services (hospitals, old-age homes, kindergartens et cetera) as compared to state provision of these services in the Nordic countries. This had long-term consequences for female labor force participation, but also for the relative ease with which child rearing and dependent employment could be combined. Agrarian parties were also strongly in favour of financing the welfare state through indirect taxes because this promised to shift ‘the expense of meeting risk from the most progressively assessed levies of the day – the direct land taxes they (the agrarians) paid to underwrite the poor-relief system – to the consumption habits of their urban political opponents’ (Baldwin 1990: 64). It is therefore in the most crucial dimensions of the welfare state – the mode of financing, the benefit structure, the provision of social services – that we see the varying impact of the different party political coalitions in the North and on the continent. In the following section we briefly outline how the various contributions to this volume address the issues raised in our broad theoretical framework.

29

1.4

Contributions to the Volume

Thomas Ertman’s contribution, with which this volume starts, provides us with a thorough, concentrated analysis of party system formation in Western Europe since the mid-19th century and emphasizes the centrality of state – church conflicts within this process. Ertman details in particular how Liberal national state elites challenged the traditional social role of the church in education and poverty relief in the last quarter of the 19th century and how the ensuing state – church conflicts led to the formation of parties of religious defence in a number of European countries. Most crucial for the subsequent analyses of this volume, he shows why in Scandinavia, in spite of at times intense conflicts between state, Lutheran state churches and revivalist movements, no parties of religious defence emerged in the important period of mass democratization. Kimberly Morgan’s comparative investigation into the origins of family policy in Western Europe offers an explanation for the unexpected – and within the conventional regime approach unexplained – pattern that a number of countries usually categorized as typical Christian Democratic welfare states like Belgium, France, and Italy nonetheless established a strong role of the state in child and family policy. The strong state role in child and family policy led to a progressive, completely ‘un-Catholic’ policy profile of these countries. This is, for instance, reflected in a relatively high labour force participation rate of mothers in these three countries, a rate that in some cases even surpasses that in the presumably most ‘women-friendly’ welfare states of Scandinavia. Had we followed the mainstream account, we would have expected that these allegedly Conservative welfare states assign a strong role to the male breadwinner model and support traditional familialistic models of the division of labour between the sexes. Morgan elegantly explains the surprisingly un-catholic policy profile as the outcome of the strong political conflicts between the Liberal nation state and the Catholic church around the turn of the century, in which the nation state gained the upper hand. Julia Lynch in her provocative treatment of the Italian case, often conceived of as the prototypical Catholic welfare state, unveils a party political logic behind Italian welfare state development rather than any straightforward influence of Catholic social doctrine. She observes that the formative period of welfare state development was a period of Liberal political domination with a program of social legislation that had an explicitly anti-clerical momentum. Where early social legislation was in line with

30

church interests, Julia Lynch argues that it was nevertheless not this agreement that caused the legislation. She shows that many of the policy outcomes of the Italian model, in particular it being a pensioners’ state, as well as the high degree of program fragmentation, are largely unintended and unanticipated outcomes of a clientelistic policy mode for which the Democrazia Cristiana became notorious. Her analysis highlights political Catholicism’s influence on Italian welfare state development, but at the same time demonstrates that one cannot assume a direct impact of Catholic social doctrine on policy outcomes. One rather needs to investigate closely the party political channel through which this influence became pertinent. Kees van Kersbergen focuses on two puzzling features of the Dutch case. First, given the level of economic development, social policy was late to emerge and remained underdeveloped for a long time. Second, after World War II, the Dutch welfare state rapidly expanded to become one of the most comprehensive and generous of its sort. If anywhere, it was in the Netherlands that the mobilization of the socio-economic cleavage via Socialist or Social Democratic parties was suppressed by the pervasive presence of the religious cleavage and religious fragmentation. It brought orthodox Protestants to power who were uninterested if not openly reluctant to develop modern social policy. Liberal anti-interventionism lined up with the Protestants’ mistrust of the state and the still vulnerable Catholics and weak Socialists were unable to gather enough support for social policy. The explanation for the speedy expansion of the welfare state after World War II is the mirror image of late development. Because Calvinist and Liberal politics were marginalized in the immediate period after the war, the way was paved for a pro-welfare Roman – red coalition between Social Catholics and Social Democrats. In the 1960s, politics mattered less under favourable economic conditions and mechanisms of ‘automatic adjustment’ contributed to the further expansion of the Dutch welfare state. Similarly, Philip Manow and Bruno Palier in their chapter on the French welfare state trajectory are confronted with two puzzles. The first puzzle is: why is there no Christian Democracy in France given that the state – church conflict had been so virulent in France around the turn of the century? Second, why does the French welfare state largely resemble the Bismarckian blue-print if political Catholicism could not have been the driving force? By focusing on the two formative moments of French welfare state formation, the decade between 1890 and 1920 and the early years after World War II, Manow and Palier show that it

31

was not social insurance, but social assistance where the French state began to define its social policy responsibilities and to crowd out the church around the turn of the century. This allowed voluntary self-help organizations, caisses and mutuelles to enter and occupy the field of accident, health, unemployment and old age insurance. After World War II, these groups, especially where they were linked to the Catholic milieu, could have been a natural electoral resource for French Christian Democracy, but the MRP was never able to tap this resource systematically and to establish stable electoral ties to the Catholic unions, the Catholic farmers’ organizations and the plethora of cooperatives and funds surrounding them. Manow and Palier point to the feebleness of French parties in the two ballot majority electoral system, the precarious legitimacy of a Catholic party and the Gaullist strategy to woo away these organizations from the MRP with a strong corporatist party platform as explanations both for the MRP’s lack of electoral success and of the corporatist design of the French welfare state (with étatist elements interspersed). In his treatment of the complex Swiss case, Herbert Obinger shows that the fierce religious conflicts in Switzerland, culminating in the Sonderbund war between Protestant and Catholic cantons in 1847, could be pacified with the devolution of political power in Swiss federalism. It is then via this federalist character of the Swiss state that many of the specific features of Switzerland’s welfare state and of its evolution can be explained. This comprises also the impact of the referendum on social policy development, since the referendum developed into an instrument that protected the Catholic minority against the Liberal Radicals, which for more than 40 years dominated the Swiss Bundesrat and challenged the Catholic church with secularization. Obinger shows how the super-majoritarian features of Swiss federalism had longterm consequences for welfare state development in this country, with a referendum that tends to block initiatives that aim at expanding the welfare state as well as those that aim at cutting it back. Karen Anderson in her investigation into Swedish social policy underlines the importance of an absence. She shows why it is so important to take the lack of a visible state – church conflict in the Scandinavian countries into account when attempting to understand the developmental path of the Nordic welfare state, here demonstrated with the exemplary case of Sweden. Her reconstruction of the Swedish political debates in the first decades of the 20th century, especially looking at the position of the Swedish state church on social policy, shows that the state

32

church is sometimes opposed to public social policy, and often simply disinterested. However, the state intrusion into the domain of social protection never became an issue of high priority for the Lutheran Swedish church. When the anti-clericalism of the Swedish Social Democrats finally was perceived to be endangering the church’s strong role in society, it was too late for the church to protect its former domains of exclusive competency. The state had already become the largely uncontested provider of social protection for all Swedish citizens. Quadagno and Rohlinger take the analysis to the contemporary period. In their investigation into the role that the Christian right played in US welfare reform of the last 30 years they address another important research question generated by the general explanatory approach of this volume: how can the religious cleavage re(-)gain political importance in a two-party system like the US-American? That the religious cleavage has regained relevance is shown in much detail in Quadagno’s and Rohlinger’s thorough account of the social policy debates of the recent past. In their explanation of the surprising resurgence of moral and religious controversies that have come to dominate US-American politics, the authors point to the shift in electoral competition from inter-party to intra-party competition. Very high incumbency return rates due to gerrymandering and to the ‘personal vote’ have shifted party political conflicts increasingly from being conflicts between Democrats and Republicans towards conflicts between middle-of-the-road Republicans and much more extremely positioned candidates and voter groups of the Christian right. Finally, Sigrun Kahl complements our approach that restricts itself to the analysis of party coalitions by focussing on religion as a cultural force that comes clearly to the forefront in such highly morally grounded issues as care for the poor and destitute. Since here the basic moral economy of a society is affected, the role of religious doctrines cannot be overestimated. Her comparative account of poor relief policies across Western Europe shows the different impact of three broad, religiously grounded social doctrines, namely those of reformed Protestantism or the Protestant sects, those of the Lutheran state churches and finally of the social doctrine of the Catholic church. Her comparison of the different treatment of the poor in countries with either more Protestant-Liberal, Lutheran-étatist or Catholic-corporatist traditions highlights how important it is to differentiate among the Protestant family of nations in order to account for the fundamental differences in Lutheran and reformed Protestant social doctrines.

33

1.5

Outlook

The evidence presented in the contributions to this volume challenge the conventional wisdom of the comparative welfare state literature. In it, the role of religion and religious cleavages, of parties of religious defence and of the legacies of fierce state – church conflicts tended to be neglected. Where addressed at all, the influence of religion was perceived as largely restricted to political Catholicism, and here most of the emphasis was put on the influence of Catholic social doctrine. The contributions to this volume acknowledge the importance of social doctrines, but stress the role of parties of religious defence more generally. The following chapters focus on the role that parties have played as those central political actors that translated religious concerns into the realm of modern democratic politics, and – even more importantly – as those actors that represent different societal interests and therefore backed different types of cross-class compromises embodied in different redistributive regimes. With this we do not mean to say that we see the impact of religion on modern welfare state development restricted to work through this party-political and electoral channel. As one important ‘transmitting’ channel parties were never simple porteparoles of religious doctrines (or other ideologies), but above all interested in maximizing votes, seats or office. Parties usually need to attract specific electoral groups and have to satisfy specific societal interests if they want to be elected. Welfare state regimes can then be explained as formulas of political compromise between different electoral and societal groups, a compromise between farmers’ and workers’ interests in Scandinavia and both an inter-party and intra-party compromise between workers and the Catholic middle-class on Europe’s continent. Yet, in order to understand which kind of political class-compromises were struck in the different European countries, we need to analyze systematically the presence or absence of different societal cleavage lines. This perspective directs attention to the different logics of redistributive politics in different party-system settings – something we hope to analyse in some more detail in future work. Our re-assessment of the impact of religion on western welfare state development is an invitation for a renewed debate on the causal sequences behind the different institutional setups of contemporary welfare states. Contributions to the volume show

34

that the threefold categorization between Social Democratic, Conservative and Liberal welfare states hides rather than elucidates the causal factors in the development of the various welfare state regimes as we know them today. Taking into account the role of religion in welfare state development allows us to better understand some of the important features of various welfare states that in mainstream analyses are either treated as anomalies, remain only poorly explained or are simply ignored. Among these features are, for instance, the ‘women friendliness’ of the French and Belgian welfare states (Kimberly Morgan’s contribution), the liberal character of the Swiss welfare state (Herbert Obinger’s contribution), the ‘belated’ generosity of the Dutch welfare state (Van Kersbergen’s contribution) or the strong role of voluntary organizations in the USA (Quadagno and Rohlinger’s contribution) versus a highly organized church run third sector in continental Europe. While we focus on the impact of religion on the welfare state, the findings of our book may also contribute to the debate on the reverse causal arrow that focuses on the impact of the welfare state on religiosity. The central message of this evolving literature is that public policies affect religiosity to the extent that a large welfare state has a negative impact on religious activity and overall religiosity. Particularly intriguing is the finding that ‘religious social mobilization and political involvement are more likely in countries with less extensive welfare systems and, conversely, that the expansion of state-sponsored social welfare will diminish, though not eliminate, the role religion will play in politics (Gill and Lundsgaarde 2004: 401). We also accept this literature’s rejection of the crude modernization perspective on secularization, just as we have amended the modernization account of welfare state emergence and development. Although we cannot elaborate on this here, we feel that further research on this issue will benefit from the position we defend here by better specifying both the independent and the dependent variable. Operationalizing the welfare state as the independent variable in terms of social spending probably fails to specify its impact very well. Taking more seriously the well-established criticism of the spending variable and including the qualitative regime specification are likely to improve the explanation of the cross-national variation in the welfare state’s impact on religiosity. Similarly (as Norris and Inglehart 2004 for instance do), distinguishing between Catholicism and Protestantism and between Protestantism’s variants along the lines we propose in this book, would specify the dependent variable much better.

35

Our understanding of the link between the welfare state and religiosity would benefit from applying the insights we develop here for the reverse causal story.

To conclude, the main problem that gave occasion to this book is that comparative research on religion and the welfare state has been incomplete because it has fairly exclusively, but mistakenly, focused on the role of political Catholicism in the development of social protection systems, wrongly interpreted or simply ignored the role of Protestantism, failed to differentiate between different strands of Protestantism and put an undue emphasis on the impact of religious ideas on welfare state institutions. This volume corrects this view, fills a gap in the literature on religion and the welfare state, and introduces an adapted model of political class coalitions that takes into account societal cleavage structures to show how contrasting church – state constellations and conflicts in the north, centre and south of Europe, variation in the party-political representation of those cleavages and differences in the social and political teachings of Catholicism, Lutheranism and reformed Protestantism have led to different coalitions between lower and middle classes, which became manifest in the distinct institutional paths of welfare state development in the West.

References Alexander, Gerard (2004), France: Reform-mongering between Majority Runoff and Proportionality, in: Josep M. Colomer (ed.), Handbook of Electoral System Choice, Houndmills, Basingstoke: Palgrave MacMillan, 209–221. Baldwin, Peter (1990), The Politics of Social Solidarity. Class Bases of the European Welfare State 1875–1975, Cambridge: Cambridge University Press. Berger, Peter L. (1990), The Sacred Canopy. Elements of a Sociological Theory of Religion, New York: Doubleday. Bergman, Torbjörn (2000), Sweden. When Minority Governments are the Rule and Majority Coalitions the Exception, in: Wolfgang C. Müller and Kaare Strom (eds), Coalition Governments in Western Europe, Oxford: Oxford University Press, 192–230. Bruce, Steve (2002), God is Dead. Secularization in the West, Oxford: Blackwell. Caramani, Daniele (2000), Elections in Western Europe since 1815. Electoral Results by Constituencies. The Societies of Europe, Basingstoke, Oxford: Macmillan. Caramani, Daniele (2004), The Nationalization of Politics. The Formation of National Electorates and Party Systems in Western Europe, New York: Cambridge University Press.

36

Chadwick, Owen (1975), The Secularization of the European Mind in the Nineteenth Century, Cambridge: Cambridge University Press. Clark, William Roberts and Matt Golder (2006), Rehabilitating Duverger’s Theory. Testing the Mechanical and Strategic Modifying Effects of Electoral Laws, Comparative Political Studies, 39 (6): 679–708. Elgie, Robert (2005), France: Stacking the Deck, in: Michael Gallagher and Paul Mitchell (eds), The Politics of Electoral Systems. Oxford, New York: Oxford University Press. Esping-Andersen, Gøsta (1990), The Three Worlds of Welfare Capitalism. Cambridge: Polity Press. Flora, Peter (1983), State, Economy and Society in Western Europe, 1815–1975, Chicago: St. James, 2 vols. Flora, Peter and Jens Alber (1981), Modernization, Democratization, and the Development of the Welfare State, in: Peter Flora and Arnold J. Heidenheimer (eds), The Development of Welfare States in Europe and America, New Brunswick and London: Transaction Books, 37– 80. Flora, Peter and Arnold J. Heidenheimer (1981), Introduction, in: idem (eds), The Development of Welfare States in Europe and America, New Brunswick and London: Transaction Books, 17–34. Gill, Anthony and Erik Lundsgaarde (2004), State Welfare Spending and Religiosity. A Crossnational Analysis. Rationality and Society, 16 (4): 399–436. Heidenheimer, Arnold J. (1983), Secularization Patterns and the Westward Spread of the Welfare State, 1883–1983. Two Dialogues about How and Why Britain, the Netherlands, and the United States Have Differed’, in R.F. Tomasson (ed.), The Welfare State, 1883–1983. Comparative Social Research 6, Greenwich and London: Jai Press. Iversen, Torben (2006), Democracy and Capitalism, in: Donald Wittman and Barry Weingast (eds), Oxford Handbook of Political Economy, Oxford: Oxford University Press, 2006, 601–623. Iversen, Torben and David Soskice (2006), Electoral Institutions and the Politics of Coalitions: Why Some Democracies Redistribute More than Others, American Political Science Review 100, 165-181. Johansen, Lars Norby (1986), Denmark, in: Peter Flora (ed.) Growth to Limits. The Western European Welfare States since World War II. Berlin, New York: De Gruyter, 293–381. Kalyvas, Stathis (1996), The Rise of Christian Democracy in Europe, Ithaca, NY: Cornell University Press. Korpi. Walter (2006), Power Resources and Employer-centered Approaches in Explanations of Welfare States and Varieties of Capitalism. Protagonists, Consenters, and Antagonists, World Politics, 58 (2): 167–206. Madeley, John (2000), Reading the Runes: The Religious Factor in Scandinavian Electoral Politics, in: David Broughton and Hans-Martien ten Napel (eds), Religion and Mass Electoral Behaviour in Europe, London: Routledge, 28–43. Manow, Philip (2004), The ‘Good, the Bad and the Ugly’. Esping-Andersen’s Welfare State Typology and the Religious Roots of the Western Welfare State. MPIfG working paper 04/3, Max Planck Institut für Gesellschaftsforschung, Cologne.

37

Manow, Philip (2006), Electoral Rules, Political Class Coalitions, Worlds of Welfare and Electoral Change: The ‘De-industrialization’-Thesis Revisited. Paper presented at the Center for European Studies, Harvard University, 16th of May 2006. Martin, David (1978), A General Theory of Secularization, Oxford: Basil Blackwell. Morgan, Kimberly J. (2002), Forging the Frontiers between State, Church and Family: Religious Cleavages and the Origins of Early Childhood Education in France, Sweden, and Germany, Politics & Society, 30 (1): 113–148. Morgan, Kimberly J. (2003), The Politics of Mothers’ Employment, World Politics, 55 (2): 259– 289. Morgan, Kimberly J. (2004), The Religious Origins of the Gendered Welfare State. The Example of Work-Family Policies in Western Europe. Paper presented at conference on Religion and the Welfare State, Cologne, 30/04–01/05. Narud, Hanne Marthe and Kaare Strom (2000), Norway. A Fragile Coalitional Order, in: Wolfgang C. Müller and Kaare Strom (ed.), Coalition Governments in Western Europe, Oxford: Oxford University Press, 158–191. Neto, Octavio Amorim and Gary W. Cox (1997), Electoral Institutions, Cleavage Structures and the Number of Parties, American Political Science Review, 41(1): 149–174. Norris, P. and Inglehart, R. (2004), Sacred and Secular: Religion and Politics Worldwide, Cambridge: Cambridge University Press. Olson, Sven (1986), Sweden, in: Peter Flora (ed.) Growth to Limits. The Western European Welfare States since World War II. Volume 1, Sweden, Norway, Finland, Denmark. Berlin, New York: de Gruyter, 1–116. Palier, Bruno (2004), The Politics of the Reforms in Bismarckian Welfare Systems, Paper presented at the Center for European Studies, Harvard University, 16th of May 2006. Parry, J. P. (1986), Democracy and Religion. Gladstone and the Liberal Party, 1867–1875. Cambridge: Cambridge University Press. Polanyi, Karl (1944 [1957]), The Great Transformation. The Political and Economic Origins of our Time, Boston: Beacon Press. Przeworski, Adam and John Sprague (1986), Paper Stones. A History of Electoral Socialism. Chicago London: The University of Chicago Press. Rodden, Jonathon (2005), Red States, Blue States, and the Welfare State: Political Geography, Representation and Government Policy around the World. Manuscript, MIT. Boston, Mass. Rogowski, Ronald (1987), Trade and the Variety of Democratic Institutions, International Organization, 41 (2): 203–223. Van Kersbergen, K. (1994), The Distinctiveness of Christian Democracy’, in David Hanley (ed.), Christian Democracy in Europe. A Comparative Perspective, London and New York: Pinter. Stephens, John D. (1979), The Transition form Capitalism to Socialism, London: Macmillan. Van Kersbergen, Kees (1995), Social Capitalism. A Study of Christian Democracy and the Welfare State, London and New York: Routledge. Van Kersbergen, Kees (1999), Contemporary Christian democracy and the Demise of the Politics of Mediation, in: H. Kitschelt, G. Marks, P. Lange and J.D. Stephens (eds), Continuity and Change in Contemporary Capitalism, Cambridge: Cambridge University Press, 346–370.

38

2

Western European Party Systems and the Religious Cleavage

Thomas Ertman

Karl Marx has long cast his shadow over the study of European political development. Beginning with the appearance of Barrington Moore’s classic The Social Origins of Dictatorship and Democracy leading theorists of long-term political change in Western Europe have centered their analyses around class actors and coalitions. More recently, however, newer explorations of comparative party formation and consolidation have begun to take into account the central role played by religious conflict across the continent during the 19th and 20th centuries. While such conflicts led to the creation of large and powerful “parties of religious defense” in Germany, Austria, Switzerland, the Netherlands, Belgium and Italy, no such parties emerged in Britain, France, the Iberian peninsula or Scandinavia despite high levels of religious tension there. Instead, in these latter cases religious cleavages were absorbed into existing, often bipolar, party landscapes. As Kees van Kersbergen and Philip Manow have argued above, such differences in party configuration with respect to religion, when combined with the effects of contrasting electoral regimes, go a far way towards explaining variations in welfare states found later in the 20th century. Building on the writings of Stein Rokkan and the more recent studies of Hans Righart, Stathis Kalyvas and Andrew Gould, this contribution will attempt to account for the divergent ways in which religion as well as class came to shape European party systems between the Congress of Vienna and the Second World War. Despite the significance of religious issues to most western Europeans throughout the 19th and early 20th centuries, such issues played little or no role in the pioneering theoretical works that sought to make sense of the continent’s rocky road to modern liberal democracy and the party landscape that arose along the way. Both Barrington Moore’s (1966) attempt to account for divergent political outcomes in England, France, Germany and Italy and Rueschemeyer, Stephens and Stephens’ (1992) extension of Moore’s project to eight additional European states rely on a class coalitional framework in which socio-economic interests alone determine political behavior (Ertman 1998). Hence it is hardly surprising that the only reference to religion in Moore’s sections on Europe are a few sentences on the sale of church lands during

39

the French Revolution (1966, 99). Rueschemeyer, Stephens and Stephens mention religion as a possible explanatory factor in their theoretical introduction (1992, 50, 67) and in their discussion of the European cases point to possible differences in patterns of peasant mobilization between Protestant and Catholic countries, but in the end argue that these differences are less significant than those between weak and strong agrarian elites (101–102). Michael Mann, in the second volume of his The Sources of Social Power (1993), sees Europe’s long nineteenth century (1760–1914) as marked by the rise of not only classes, but also nations, yet once again religion figures little in these processes. This picture becomes more complicated when we turn to Gregory Luebbert’s ambitious, complex Liberalism, Fascism or Social Democracy (1991). Like all of the authors mentioned above, he views classes as fundamental political actors, but he sees them as operating principally through parties. For Luebbert, a key determinant of interwar party constellations and the political outcomes influenced by them is the strength of liberal movements prior to 1914. Such movements were strong, as in Britain, France and Switzerland, where “their most natural potential constituency, the middle classes, were not politically divided by antagonisms rooted in religious, regional, linguistic and urban-rural differences.” (1991, 7) In these cases, he argues, religion tended in fact to reinforce middle class liberal cohesion vis-a-vis conservatives (108). In Scandinavia, however, religious as well as territorial and cultural conflicts undermined this cohesion, thereby opening the door to social democratic victory in the 1930’s; in Germany, Italy and Spain, by contrast, middle class liberal weakness paved the way for fascism, though for reasons rooted entirely in class coalition politics (66, 303–305; see also Ertman 1998). Both Mann’s attempt to broaden a class-based analytic framework through a complementary focus on nationalism and Luebbert’s effort to do the same through a consideration of parties and of non-class (including religious) cleavages owe much to the work of Stein Rokkan, one of the few classical theorists of European politics to have taken religion seriously. Indeed, at the heart of his “geoeconomic-geopolitical model” of European development (Rokkan 1981) stands the impact of two revolutions – the industrial and the national, but his understanding of what the latter entailed is much more far-reaching than that of Mann or Luebbert. Rokkan defines the national revolution as “the conflict between the central nation-building culture and the increasing

40

resistance of the ethnically, linguistically, or religiously distinct subject populations in the provinces and peripheries …” (1970, 102; italics in original; also Lipset and Rokkan 1967, 14). Furthermore, he sees this revolution as independent, rather than merely an epiphenomenon, of the industrial revolution and claims that it “produced the deepest and bitterest oppositions. The decisive battle came to stand between the aspirations of the mobilizing nation-state and the corporate claims of the churches. This was far more than a matter of economics” (1970, 103; italics in original; also Lipset and Rokkan 1967, 15). Rokkan goes on to illustrate what he means by describing the emergence of Calvinist and Catholic “parties of religious defense” in the Netherlands in response to state attempts to gain control of education. The result was the emergence of a “pillarized” society in that country made up of self-contained politico-religious communities built around networks of exclusive professional and leisure-time associations. He then suggests that the pattern of social and political organization that arose in 19th century Scandinavia and Britain in the wake of nationbuilding differed from that found in the Netherlands (1970, 103–105; Lipset and Rokkan 1967, 15–18). This point receives further elaboration in an earlier article (1967) by Rokkan on his native Norway. He shows there how an opposition movement to the central government emerged beginning in the 1830’s that united urban professionals outside of the state administration with a rural population seeking to defend its language and customs from state attempts at cultural centralization. An important element in this movement, which in the 1860’s organized itself into a party known as the Venstre (Left), was opposition to the established Lutheran Church by those in both town and country inspired by Haugist revivalism (1967, 370–375). Hence in this case mobilization against state cultural policy did not lead to the emergence of a “party of religious defense;” rather, opposing religious groups were part of broader, rival political coalitions (the “Left” and “Right” [Hoeyre]). As we have seen, Rokkan’s concern with nationbuilding and the central role of religious conflict within it was not shared by subsequent theorists of European political development. This might have been otherwise had his “geoeconomic-geopolitical model” actually succeeded in accounting in a parsimonious way for the pattern of democratic and non-democratic outcomes in interwar Western Europe. This, however, it could not do. As Rokkan himself was forced to admit (1981, 88): “One conclu-

41

sion is clear; our five cases [of democratic failure, i.e. Italy, Germany, Austria, Spain and Portugal] fall into several distinctive cells on the [conceptual] map; they do not form one single cluster.” This was even more true of the other western European states that did not succumb to dictatorship. Fortunately, the thread of Rokkan’s argument has been taken up once again in monographs by Hans Righart, Stathis Kalyvas and Andrew Gould, all of whom explore in greater detail aspects of the relationship between nationalbuilding and religious conflict only briefly touched upon by the Norwegian political scientist. As such, they represent an indispensable starting point for any attempt to bring religion into the understanding of divergent European party landscapes. In 1986, historian Hans Righart published his comparative study De katholieke zuil in Europa [The Catholic Pillar in Europe], which has not received the attention it deserves because it appeared only in Dutch. In this work, Righart shows how, beginning in the 1880’s, members of the lower clergy and lay activists in Austria, Switzerland, Belgium and the Netherlands slowly built up a network of Catholic social, economic, leisure time and interest organizations as an act of defense against what was perceived as the relentless forward march of secularization brought about by social change—and above all by industrialization. In so doing, they took as their model the organizations of their political and ideological rivals, the liberals and the socialists. This process faced extensive though ultimately futile resistance from both the ecclesiastical hierarchy and established, often patrician or aristocratic, Catholic elites fearful of its democratic potential. It culminated in the formation, first in Austria (1890/1907), then in Switzerland (1912), Belgium (1921) and finally the Netherlands (1926) of precisely those “parties of religious defense” mentioned by Rokkan above (1986, 28, 274). In his prize-winning book The Rise of Christian Democracy in Europe (1996), political scientist Stathis Kalyvas goes beyond the work of Righart by including the cases of Germany, Italy and France in addition to Austria, Belgium and the Netherlands, though he does exclude Switzerland. If, for Kalyvas, the liberal, anti-clerical policies of the 1860’s, 1870’s and 1880’s feature more prominently as a direct cause of Catholic mobilization than they do for Righart, this is because his primary focus is on the process of party formation rather than the broader social activism and interest organization building (“pillarization”) studied by his Dutch counterpart. Kalyvas

42

makes a number of points about this process that are relevant to our concerns. First, he illustrates the extent to which political conflict in much of Western Europe from the 1860’s until World War I was dominated by questions of church-state relations, especially as they affected education. Second, he shows that the events surrounding the formation of confessional parties were extremely complex, involving autonomous actions and calculations by many groups of Catholic actors – the lower clergy, the national church hierarchy, the Papacy, and political and social activists – as well as by their liberal opponents. Third, once in existence confessional parties both helped redefine what it meant to be Catholic and altered the dynamics of the entire party landscape surrounding them. Finally, he stresses that the mere presence of Catholics did not dictate the emergence and consolidation of a Catholic party. While such parties did indeed appear in Germany, Austria, Switzerland, Belgium, the Netherlands and Italy, they did not do so in equally Catholic France (or in Spain or Portugal, not included in the study). Kalyvas explains the French “puzzle of nonformation” by the existence there of a “regime cleavage” that coincided with the clerical/anticlerical divide. As a result of this cleavage, the origins of which go back to the Revolution and beyond, supporters of the Church tended to be monarchists and their opponents partisans of a republic. Hence opposition to anticlerical measures during the crucial period 1876–1889 was spearheaded by already-organized monarchist conservatives. Given the fact that these forces had won the elections of 1871 and could possibly return to power again, it seemed more reasonable for the Church, according to Kalyvas, to entrust its defense to this political grouping rather than embark on the risky and costly strategy of building a new confessional party. By the time the French hierarchy realized that the monarchists had no future, the opportune moment to create such a party had passed (1996, 137–141). Andrew Gould agrees that it was the “regime cleavage,” first between neo-absolutist and constitutional monarchy and then between monarchy and republic, that dominated French politics throughout the 19th century. He also traces in much greater detail than Kalyvas the way in which the clerical/anti-clerical divide was at the heart of that cleavage: views on the Church and its proper place in society served to bind together otherwise economically, geographically and ideologically diverse conservative/monarchist/nationalist and liberal/republican camps until as late as Vichy

43

(1999, 56–64). Gould further argues that the unifying focus provided by anticlericalism made possible the long periods of Liberal dominance in Belgium (1847– 1884) and Switzerland (1848–1914 – not studied by Kalyvas), and that opposition to the privileged position of the Church of England served a similar function for the British liberal movement in its heyday under Gladstone (1999, 31–32, 88, 109, 120– 123). What is more, it was the failure of the (overwhelmingly Protestant) German liberals to win over the established Protestant churches to their cause, despite a common antagonism to Catholicism, that substantially weakened them when compared with their Swiss counterparts (1999, 77–78, 83). The picture of 19th and early 20th century European politics painted recently by Righart, Kalyvas and Gould is quite different than that found among previous theorists with the exception of Rokkan. It is one in which the struggle between Protestants and Catholics, between nationbuilding states and a reinvigorated Papacy, between dissenting and established Protestant churches, and between secular and religiously inspired worldviews were for most of the period as salient, if not more so, than the class struggle. What, then, might an alternative framework for understanding political – and especially party – development in western Europe look like, one that, by building on the work of these three authors, Rokkan and countless historians, would give the role of religion its due? I suggest we start with the distinction mentioned above between countries where “parties of religious defense” emerged between 1870 and 1926 – Belgium, the Netherlands, Switzerland, Austria, Italy and Germany – and those where they did not – Britain, France, the Scandinavian kingdoms, Spain and Portugal. This distinction is significant because it correlates almost exactly (the Italian case is ambiguous) to differences between those European states that after 1945 developed “consociational” or “concordance” forms of democracy characterized by “oversized” coalitions of left and right, negotiated settlement of disputes and a proportional distribution of offices; and those whose democracies could better be classified as “pluralist” or “competitive,” where majority rule has tended to dominate (Lehmbruch 1974; Lijphart 1975; Schmidt 1997, 231–240). Furthermore, it has, I would argue, deeper historical roots, because Europe’s “consociational zone” largely overlaps with the core area of the continent once occupied by the medieval and early modern Holy Roman Empire which from the 11th century through (de jura) 1648 stretched from the Low Countries

44

through Germany, Switzerland and Austria to the Pope’s domains in central Italy (and under Frederick II further south still). By contrast, the competitive democracies can be found to the west and north of this core on the lands of the western Carolingian Empire and on Western Europe’s “periphery” (Otto Hintze) – the British Isles, Iberia and Scandinavia. Why does this geo-historic division matter? It matters because east Carolingian rulers and their successors – unlike their west Carolingian counterparts – were never able to recentralize power in a durable manner after the political collapse of the 9th and 10th centuries. As a result, the Holy Roman Empire developed as an area of political fragmentation and, after the Reformation, religious pluralism. Whereas rulers in France and in the periphery were engaged in state- and precocious nationbuilding for centuries prior to the 19th century, it was only between 1830 (collapse of United Kingdom of the Netherlands and emergence of separate Belgium and Dutch states) and 1874 (creation of federal Switzerland 1848–1874; Italian unification 1861–1870; German unification 1867–1871; Austrian Ausgleich 1867) that modern nation states (Cisleithanian Austria in part excepted) finally appeared in the core zone. The formal creation of these polities was only the beginning, however. Their leaders now faced enormous tasks of statebuilding, nationbuilding and economic integration. Between the 1840’s and 1870’s the most common model chosen for this project was, as Kalyvas and Gould note, the liberal one: nations built around the rule of law, free markets and the political participation of those citizens who were economically independent; but also around a state superior to all other institutions (especially the Church) and outfitted with a non-religious education system reflecting that superiority. After his own traumatic experiences during 1848 – he was driven from Rome by revolutionaries – Pope Pius IX came to see this project as directed above all against the Church and religion more generally. The result was the ultramontane counteroffensive of 1864–1870 (Syllabus Errorum and Declaration of Papal Infallibility) that in turn helped provoke the new anticlerical backlash of the next two decades (Chadwick 1975, 111– 113). In Austria (1868–9), the Netherlands (1878) and Belgium (1879) this backlash mainly took the form of initiatives to secularize the school system, whereas governments in Germany, Switzerland Italy imposed more wide-ranging restrictions on the Church’s freedom of maneuver (“Kulturkampf”).

45

It was in the first instance these liberal attempts to use the ultramontane threat to drive forward their own nationbuilding projects that, as Kalyvas has argued, led to political mobilization and ultimately party formation on the part of Catholics and, in the Netherlands, fundamentalist Protestants (Anti-Revolutionary Party of Abraham Kuyper, 1879) opposed to the supposed liberal goal of a secular, godless state. Though centralized Catholic parties appeared relatively late in Switzerland (1912), Belgium (1921) and the Netherlands (1926), local and regional electoral organizations were active in all of these countries by the 1880’s. Meanwhile, another group opposed to the economic face of the liberal state – social democratic workers – had begun to organize not just parties, but also trade unions, cooperatives, and women’s, youth and leisure-time associations. As Hans Righart (1986) and others have shown, this challenge from a movement even more ungodly than liberalism spurred the religious to create their own, vast associational infrastructures of which the confessional parties became merely the political arm. Hence by the eve of the First World War, all six of successor states of the medieval Holy Roman Empire had come to possess, despite their great differences in size, culture, and level of economic development, political landscapes that shared a striking common feature: the presence of at least one (in the Netherlands: two) politically organized religious subcultures that were in most respects mirror images of the social democratic subcultures with which they were in electoral and ideological competition. Both Catholics/Calvinists and social democrats had succeeded within their respective camps or “pillars” in creating something like self-contained lifeworlds, alternative communities that shielded their members to some extent from the secular, free-market state many of them rejected. As such they had realized an opposition strategy best articulated not by Marx, but by the Dutch fundamentalist Protestant Abraham Kuyper whose battle cry had been sovereiniteit in eigen kring, or sovereignty within one’s own sphere of life (Ertman 2000, 168). It was in that portion of the political spectrum outside of the religious and social democratic camps that these six cases differed among themselves. In Switzerland, Belgium and the Netherlands before 1914, liberal movements offered the only alternatives to confessionalism or socialism, though these movements were fissiparous, most often divided into conservative, moderate and radical wings. Competition from their increasingly well-organized opponents, however, forced some degree of liberal con-

46

solidation: in Switzerland, it led to the founding of the powerful Freisinnigdemokratische Partei der Schweiz (Freethinking-Democratic Party or FDP) in 1894 (Gruner 1977, 54–55, 74–78). In Belgium (1900) and the Netherlands (1912) such competition forced liberal factions to unite around common programs of universal and equal manhood suffrage, free trade and the defense of public schools (Lamberts 1993, 371; Taal 1980, 527–528). While the liberal milieu in all three countries also possessed its own network of associations, newspapers and interest groups, these remained smaller and less dense than those of the confessionals or social democrats (Ertman 2000, 170–173). The contribution of the confessional camp in Switzerland, Belgium and the Netherlands to the democratization of their lands is an ambiguous one. It was the cantons which would later form the bastions of the Swiss Catholic party that opposed the creation of a federal and democratic state promoted by the Radicals, thereby provoking the short civil war of 1847. Thereafter, however, this milieu became a firm supporter of the new Confederation. In Belgium, it was radical liberals and socialists who after 1881 began to demand the transformation of a parliamentary monarchy with limited suffrage into a (male) democracy. The Catholic party was split over this demand, but following a general strike in 1893 a compromise was reached allowing for the introduction of universal manhood suffrage, but with plural voting. To stem the decline of the liberals and prevent the polarization of the country into a (Dutch speaking) Catholic Flanders and a (French speaking) socialist Wallonia, the progressive faction of the dominant Catholic parliamentary group broke ranks in 1899 to permit the introduction of proportional representation. Plural voting was finally removed with the support of all parties in the wake of World War I in 1919, but liberals and socialists blocked Catholic party attempts to extend the vote to women as well, fearful that the latter would vote overwhelmingly for the confessionals. As a result, universal suffrage was delayed until 1948 (Gilissen 1958 121–141). In the Netherlands, as in Belgium, socialists and left liberals campaigned most tirelessly for universal manhood suffrage, while the religious parties were divided on the issue, with the Catholics generally opposed and the Calvinist ARP split. All men finally received the vote in 1917 through an all-party compromise (the Pacificatie) that called for the state to subsidize on an equal footing all three national school systems (public, Catholic and fundamentalist Calvinist). Women were enfranchised three years later (Kossmann 1978, 326–328, 350–361, 555–556).

47

If the record of the confessional parties in promoting full democratization was mixed, their support for it during the troubled interwar years was crucial and unambiguous. The basis of this support was the willingness of former rivals – liberals and Catholics in Switzerland, liberals, Catholics and Calvinists in the Netherlands, and liberals, Catholics and socialists in Belgium – to work closely together in defense of the democratic system. Thus the formerly hegemonic Swiss FDP shared power on the federal executive with the Catholic Conservative People’s Party through this period, and seven of the nine interwar Dutch cabinets were composed either of the confessional parties alone (4) or the latter in coalition with the liberals (3). The degree of crossparty cooperation in Belgium was even more remarkable: half (9 of 18) of the cabinets between 1918 and 1940 contained representatives of all three camps, with six more formed by Catholics and liberals and two others by Catholics and socialists (Gould 1999, 87; Kossmann 1978, 718–733; Capoccia 2005, 108–137). Gerhard Lehmbruch has pointed to the historical precedents for this willingness to share power in many institutional arrangements unique to Europe’s Holy Roman “core”: the periodic meetings of the Swiss cantons (and, one might add, the Estates of the Low Countries) and the interaction among members of the German Confederation (post-1815) and the Empire itself (pre-1806) (Lehmbruch 1974). While cross-class confessional parties clearly played a crucial role in consolidating and defending democracy in Switzerland, the Netherlands and Belgium, they were unable or unwilling to accomplish the same task, despite the traditions of accommodation and concordantia cited by Lehmbruch, in Italy, Germany and Austria. Why was this so? To answer this question, we must look once again at the character of the political forces outside of the Catholic and socialist camps. As we have seen, in Switzerland, the Netherlands and Belgium it was above all alliances between confessional parties and their former liberal enemies that served to protect democracy during the crisis-filled interwar decades. This was so because the socialist camp, even in Belgium, was not yet accepted as a fully “normal” coalition partner, a development that would only occur after World War II. In Germany, Austria and Italy, however, the non-Catholic and non-socialist forces were weak and divided, a result, I would argue, of the “rule-rigging” employed in all three countries before 1914 to aid these forces in the face of the threat posed by highly organized, disciplined confessional and social democratic political forces. In the end, the rejection of parliamentarization (Germany,

48

Prussia, Austria), the use of voting by tax class (Prussia, Austria pre-1907), or of the technique of trasformismo combined with government-sponsored electoral manipulation (Italy) only left liberal and conservative parties less able to face up to the pressures of truly open electoral competition after 1918/19. In both Italy and Germany the consequences were similar: a weak and fragmented liberal and moderate conservative “camp” unable to provide consistent and effective support to those socialists and Catholics attempting to defend democracy. Worse still, it was precisely this growing vacuum on the moderate right that created the political space for charismatic fascist leaders who argued that only they were capable of uniting all those outside of the red and black (clerical) subcultures into a movement that would break the power of these enemies of the nation once and for all by putting an end to all encapsulated subcultures, creating thereby the single national community or Volksgemeinschaft. In Austria, the situation was more complicated. The “third force” there had always been represented by German nationalism, but this camp was weakened after the creation of the Republic both by the endorsement on the part of all parties of unification with Germany and the simultaneous ban on that unification in the Versailles Treaty. The result after 1920 was a dangerous polarization between the socialists and the governing Christian Socials that only became worse in the late 1920’s with the rise of the Austrofascist Heimwehr and national socialism. Faced with perceived threats from both the powerful socialist camp and a victorious Hitler, that faction of the Christian Social party was able to gain the upper hand which favored an alliance with the Heimwehr, thereby leading to the creation in 1933/34 of a “clerico-fascist” dictatorship allied with Italy and to civil war with the left (Simon 1978). Hence in Austria, as in Germany, a Catholic party that had supported the democratic system during the 1920’s turned away from it after 1930, though in this case in the direction of outright dictatorship rather than the presidentialism or mild authoritarianism favored by some in the Zentrum. This is less surprising in light of the legacy of rightwing populism and antisemitism that the Christian Socials’ dominant figure Karl Lueger had bequeathed to the party. In summary, then, the contribution of religious parties to the advent and consolidation of democracy in these three cases was on the whole negative when compared to

49

the role played by their counterparts in Switzerland, the Netherlands and Belgium: in Italy because the long papal ban (until the eve of World War I) on national political activity prevented the Partito Populare from becoming the hegemonic force its Christian Democratic successor would later be; in Germany because the support of the entire Catholic political movement for democracy had never been unequivocal, certainly before 1918 but also thereafter; and in Austria because it was the clerical party itself that, despite its positive role in the 1920’s, ultimately led the country down the road to dictatorship. Yet it is important to remember that confessional parties in the first three cases were also not on the whole advocates of democracy prior to World War I. What transformed them into unwavering supporters of that cause thereafter, I would argue, was the broader political context, one in which they benefited from strong and reliable democratic allies, especially on the center right. And this was precisely what their counterparts in Italy, Germany and Austria lacked. However, after the defeat of 1945, all three countries reintroduced a (far more durable) form of democracy that, in the case of Austria, was clearly consociational and, in Germany and Italy, exhibited strong consociational features. The integration of religious groups and concerns into political life followed a very different pattern outside Western Europe’s historical “core.” Despite the presence of three nations (France, Spain, Portugal) that, like Belgium, were overwhelmingly Catholic; and a fourth (the United Kingdom) that, like the Netherlands, possessed both large Catholic and Protestant dissenting minorities, no confessional parties appeared there before the 1930’s, and if they did emerge thereafter (CEDA in Spain, MRP in France) they were short-lived. Rather, the political landscape tended to be divided throughout most of the 19th century into two broad groupings, Liberals and Conservatives (or Liberal-Conservatives) in Spain (where the terms originated), Portugal and Britain; Liberals, then Republicans and Conservatives in France; and Left and Right in Scandinavia. The root cause of this contrasting binary structure (as opposed to the confessionals-socialists-liberals and/or conservatives model present in the “core”) can be found, I would argue, in the much earlier onset – during the late 17th and 18th centuries – of something like nationbuilding in these countries. It was during this period that some segments of an emerging public sphere began demanding – and some statesmen tried to implement – reforms aimed at improving the quality of state administration, spreading education, modernizing the economy

50

and, in some cases, limiting the prerogatives of the crown. Such reforms inevitably required the curtailing or even elimination of privileges enjoyed by powerful institutions and groups like the monarch, the Church, existing officeholders and noble landowners, and hence they provoked strong resistance. The eventual result was the emergence of a cleavage around supporters and opponents of reform, which often overlapped with that between supporters and opponents of the Church. Yet since the reform agenda at this time was filled with so many items, this cleavage was not driven during the 18th and early 19th centuries primarily or exclusively by religious issues. However, as ever more tasks on the reform agenda – the introduction of constitutional monarchy, the end to “feudal remnants,” the freeing of markets, administrative professionalization – were accomplished, the lingering privileges and influence of the established churches, especially in the area of education, loomed ever larger. This, along with the aggressive (though ultimately defensive) ultramontanism of Pius IX, explains the ever greater role that anti-clericalism came to play from about the 1850’s in holding together the disparate elements of the liberal/republican camp in Britain, France, Iberia and even Scandinavia. Yet the role of religious issues and identities in structuring the political landscape went further than this: it also introduced a crucial “cross-cutting cleavage” that softened class and regional antagonisms and ensured that the opposing liberal/conservatives or left/right blocks would be economically and geographically diverse in character, thereby preventing the kind of polarization that could lead to instability and even civil war as well as rendering the alternation of the two camps in power more palatable. Mention has already been made of the way Republicans in France used anticlericalism to create a broad-based movement uniting peasants and local notables in regions traditionally hostile to the Church with urban democrats in support of the new regime. Historians have also recently come to recognize the absolute centrality of religious issues to 19th century British politics as well (e.g. Parry 1986) . Thus from the 1860’s the most important component of the Liberal Party were dissenting Protestants from the Midlands and the North who furnished the party with its most fervent activists and reliable voters. Liberal leader W.E. Gladstone then skillfully used a number of issues with a religious component – Church rates, Irish disestablishment, state support for confessional schools – to bind this group to the more moderate (and aristocratic) Whig-liberals who shared the nonconformist wish to limit the privileges

51

of the established church. At the same time, however, the image of the Liberals as dominated culturally by “killjoy” teetotalers and puritan moralists drove many in the working class to vote for the Tories, who cleverly positioned themselves as the party willing to defend the English working man’s traditional pleasures of beer and skittles and held many meetings in public houses (Parry 1986, 5–8, 80–81, 199–204; Nossiter 1970, 177–180; Harrison 1996, 181–182). It was the mixed-class support base of both parties, to a significant degree induced by the religious cleavage, which led both groupings to support suffrage extensions in 1867 and 1884–5. Political dynamics in 19th century Scandinavia bear a certain resemblance to those in Britain, and with good reason. From the 1830’s onward the evangelical revival that began in England in the late 1700’s began to arrive there, sometimes brought directly by Methodist preachers. The spread of dissenting “free churches” above all among peasants across rural Norway, Sweden and Denmark led to calls by their supporters both to broaden political participation and to limit the privileges of the state Lutheran Churches. It was these groups that created a solid base for reformist Left (Venstre) parties in all three countries, whereas the supporters of the established political, economic and ecclesiastical order supported the Right (Hoeyre). This identification between the Left and dissent was strengthened with the appearance of strong temperance movements across Scandinavia which, as in Britain, drove anti-temperance voters to the Right. While, again as in Britain, the Venstre parties would increasingly cede their position after the 1890’s to Labor parties, the latter would also continue to draw support from the dissenting/temperance milieu. In all three cases, it was pressure from the liberal/social democratic left (in the broadest sense) with its hard core of religiously motivated members and voters that brought full (male) parliamentary democracy to Norway in 1898, Denmark in 1901 and Sweden in 1917 (Rokkan 1967, 372–379, 391–395, 415–420; Derry 1979, 229–232, 257–268; Lundkvist 1992; Pettersson1996). As Kees van Kersbergen and Philip Manow have implied above, it was the switch from majoritarian electoral systems to proportional representation in Sweden in 1909, Denmark in 1920 and Norway in 1921that pushed these countries away from the dualistic party landscape characteristic of older nationbuilders like Britain towards a four part configuration: conservatives, right and left liberals and social democrats in Denmark and conservatives, liberals, social democracts and agrarian center parties in Norway and Sweden (Caramani 2004, 136–142; Derry 1979, 261).

52

While the positive contribution of dissenting and anti-established church groups to both the advent and consolidation of democracy in Britain, France and Scandinavia is clear, why did a religiously conditioned bi-polar political landscape not protect the republican regime in interwar Spain? The answer lies once again, as with the German, Italian and Austrian cases, with the endemic use of electoral manipulation after 1876 in a way that undermined the potentially beneficial effects of a two party system with a strong clerical/anticlerical cleavage. Such a cleavage was, as in France, central to the struggle between Liberals and Conservatives in Spain between 1815 and 1874 and, even more than in France, this struggle led to numerous revolutions and coups as one side and then the other gained the upper hand. In 1876, however, in the wake of the collapse of the First Republic and the restoration of the monarchy, the conservative politician Antonio Cánovas del Castillo introduced a political system devised to defuse this conflict and bring stability to the country. It involved the systematic alternation of liberals and conservatives in office (the turno), an alternation ensured by widespread electoral manipulation and a reliance on local strongmen (caciques). Both parties in effect converted themselves into patronage machines and put the divisions between them, especially on the issue of the Church, to one side (Linz 1967, 198–205; Callahan 1984, 273–277). This “freezing” of the political landscape with the help of clientelistic methods, a development that spread to Portugal for similar reasons soon afterward (Tavares de Almeida 1991), had the unfortunate effect of preventing the emergence of modern parties of the moderate center and right until well after 1900. As in Italy, Germany and Austria, it was the marked weakness of such parties in 1930’s Spain, and specifically the collapse of the centrist Radicals at the 1936 elections when they lost 100 of their 104 seats in the Cortes, which brought about the extreme political polarization that was the proximate cause of the July 17th military coup against the Republic (Payne 2006). Interestingly, however, Spain after the death of Franco eventually returned to a two party system, but this time an honest one capable of protecting the new democracy. As the social science literature has come increasingly to realize, the religious cleavage stood at the center of political life during the formative period of the western European party landscape. In the older nationbuilders such as Britain and the Scandinavian kingdoms, this cleavage both shaped and was absorbed into a dichotomous division between left and right camps that from the 1890’s onward gave way to a tri-

53

partite (labor/liberal/conservative) party configuration. The introduction of proportional representation in the later polities between 1909 and 1921 then permitted the emergence and consolidation, unlike in majoritarian Britain, of a fourth, agrarianbased political force that, as both Peter Baldwin (1990) and this volume argue, played a crucial role in construction of Esping-Andersen’s social democratic welfare state. In the newer nationbuilders Belgium, the Netherlands, Switzerland, Italy, Germany and Austria, by contrast, confessional conflict led directly to the formation of “parties of religious defense,” to use Rokkan’s term, that right from the start also indelibly marked the social policy regimes of these states.

References Baldwin, Peter. 1990. The Politics of Social Solidarity: Class Bases of the European Welfare State 1875–1975. Cambridge: Cambridge University Press. Callahan, William. 1984. Church, Politics, and Society in Spain, 1750–1874. Cambridge: Harvard University Press. Capoccia, Giovanni. 2005. Defending Democracy: Reactions to Extremism in Interwar Europe. Baltimore: Johns Hopkins University Press. Caramani, Daniele. 2004. The Nationalization of Politics: The Formation of National Electorates and Party Systems in Western Europe. Cambridge: Cambridge University Press. Chadwick, Owen. 1975. The Secularization of the European Mind in the 19th Century. Cambridge: Cambridge University Press. Derry, T.K. 1979. A History of Scandinavia. Minneapolis: University of Minnesota Press. Ertman, Thomas. 1998. “Democracy and Dictatorship in Interwar Western Europe Revisited.” World Politics, vol. 50, no. 3 (April), pp. 475–505. ———. 2000. “Liberalization, Democratization and the Origins of a `Pillarized’ Civil Society in Nineteenth-Century Belgium and the Netherlands.” In Civil Society before Democracy: Lessons from Nineteenth-Century Europe. Eds. Nancy Bermeo and Philip Nord. Lanham: Rowman & Littlefield, pp. 155–178. Gilissen, John. 1958. Le Régime représentatif en Belgique depuis 1790. Brussels: La Renaissance du Livre. Gould, Andrew. 1999. Origins of Liberal Dominance: State, Church, and Party in NineteenthCentury Europe. Ann Arbor: University of Michigan Press. Gruner, Erich. 1977. Die Parteien in der Schweiz. 2nd edition. Bern: Francke Verlag. Harrison, Brian. 1996. The Transformation of British Politics 1860–1995. Oxford: Oxford University Press. Kalyvas, Stathis. 1996. The Rise of Christian Democracy in Europe. Ithaca: Cornell University Press. Kossmann, E.H. 1978. The Low Countries 1780–1940. Oxford: Clarendon.

54

Lamberts, E. 1993. “Belgie sinds 1830.” In Geschiedenis van de Nederlanden. Eds. J.C.H. Blom and E. Lamberts. Rijswijk: Nijgh & Van Ditmar, pp. 341–416. Lehmbruch, Gerhard. 1974. “A Non-Competitive Pattern of Conflict Management in Liberal Democracies: The Case of Switzerland, Austria and Lebanon.” In Consociational Democracy: Political Accommodation in Segmented Societies. Ed. Kenneth McRae. Toronto: McClelland and Stewart, pp. 90–97. Lijphart, Arend. 1975. The Politics of Accommodation. 2nd edition. Berkeley: University of California Press. Linz, Juan. 1967. “The Party System of Spain Past and Future.” In Party Systems and Voter Alignments: Cross-National Perspectives. Eds. Seymour Martin Lipset and Stein Rokkan. New York: Free Press, pp. 1–64. Lipset, Seymour Martin and Stein Rokkan. 1967. “Cleavage Structures, Party Systems, and Voter Alignments: An Introduction.” In Party Systems and Voter Alignments: CrossNational Perspectives. Eds. Seymour Martin Lipset and Stein Rokkan. New York: Free Press, pp. 1–64. Luebbert, Gregory. 1991. Liberalism, Fascism, or Social Democracy. Oxford: Oxford University Press. Lundkvist, Sven. 1992. “Die Volksbewegungen und das `Modell Schweden’.” In Weltgeltung und Regionalitaet: Nordeuropa um 1990. Eds. Robert Bohn and Michael Engelbrecht. Frankfurt/Main: Peter Lang, pp. 261–267. Mann, Michael. 1993. The Sources of Social Power. Vol. II. The Rise of Classes and NationStates, 1760–1914. Cambridge: Cambridge University Press. Moore, Barrington. 1966. The Social Origins of Dictatorship and Democracy. Boston: Beacon Press. Nossiter, T.J. 1970. “Aspects of Electoral Behavior in English Constituencies, 1832–1868.” In Mass Politics: Studies in Political Sociology. Eds. Erik Allardt and Stein Rokkan. New York: Free Press, pp. 160–283. Parry, J.P. 1986. Democracy and Religion: Gladstone and the Liberal Party 1867–1875. Cambridge: Cambridge University Press. Payne, Stanley. 2006. The Collapse of the Spanish Republic 1933–1936: Origins of the Civil War. New Haven: Yale University Press. Pettersson, Lars. 1996. “In Search of Respectability: Popular Movements in Scandinavian Democratization. The Swedish Case.” Democratisation in the Third World. Uppsala: Uppsala University Press, pp. 77–94. Righart, Hans. 1986. De katholieke zuil in Europa. Amsterdam: Boom Meppel. Rokkan, Stein. 1967. “Geography, Religion, and Social Class: Crosscutting Cleavages in Norwegian Politics.” In Party Systems and Voter Alignments: Cross-National Perspectives. Eds. Seymour Martin Lipset and Stein Rokkan. New York: Free Press, pp. 367–444. ———. 1970. “Nation-Building, Cleavage Formation and the Structuring of Mass Politics.” In Stein Rokkan, Citizens, Elections, Parties. New York: David McKay, pp. 72–142. ———. 1981. “Territories, Nations, Parties: Toward a Geoeconomic-Geopolitical Model for the Explanation of Variations within Western Europe.” In From National Development to

55

Global Community. Eds. Richard Merritt and Bruce Russett. London: George Allen and Unwin, pp. 70–95. Rueschemeyer, Dietrich, Evelyne Huber Stephens and John Stephens. 1992. Capitalist Development and Democracy. Cambridge: Polity Press. Schmidt, Manfred. 1997. Demokratietheorien. 2nd edition. Opladen: Leske + Budrich. Simon, Walter. 1978. “Democracy in the Shadow of Imposed Sovereignty: The First Republic of Austria.” In The Breakdown of Democratic Regimes: Europe. Eds. Juan Linz and Alfred Stepan. Baltimore: Johns Hopkins University Press, pp. 80–121. Taal, G. 1980. Liberalen en Radicalen in Nederland, 1872–1901. Den Haag: Martinus Nijhoff. Tavares de Almeida, Pedro. 1991. Eleições e Caciquismo no Portugal Oitocentista (1868–1890). Lisbon: Difel.

56

3

The Religious Foundations of Work-Family Policies in Western Europe

Kimberly J. Morgan

European countries today are increasingly under pressure to improve the lives of working parents, and working mothers in particular. Low fertility rates and the threat of labor shortages when the baby boom generation retires have put the issue of work-family policies onto the political agenda. Even so, the pace of change has been remarkably slow in many of the conservative welfare states. In Germany, for example, change has been stymied by continuing skepticism about mothers working when their children are young and by the fact that local governments and the voluntary sector claim responsibility for social services policy (Evers, Lewis and Riedel 2005). In the Netherlands, governments have sought to increase women’s workforce participation, but this has largely been through part-time work, thereby preserving much maternal care at home. At the same time, other European countries – the Nordic countries, France, and Belgium – have a longer tradition of helping mothers work through parental leave, child care, and working time policies, and they tend to have higher rates of female workforce participation, and higher fertility rates too (Table 1). These policy differences and their apparent stability raise questions about why it is that countries have gone off on diverging policy trajectories and stayed on those paths. This chapter traces the origins of these diverging pathways by exploring the historical roots of policies for working mothers. Temporally, the focus is on two periods: (1) the late 19th and early 20th centuries, a critical time in Western Europe for the formation of political parties and early social programs; and (2) the three decades following the Second World War, when policy legacies and the nature of party systems would influence public policies towards working mothers. In the latter half of the 19th and first decades of the 20th centuries, there were three patterns of religious practice, religious divisions, and related political conflict that would have lasting significance for gendered aspects of the welfare state: (1) a Nordic model, consisting of religious homogeneity, church-state fusion, and resulting lack of religiously-based conflict; (2) a clerical-anticlerical mode of contestation, found in Catholic countries with strong anticlerical forces (see Thomas Ertman’s contribution

57

to this volume); (3) an accomodationist model in countries where religious forces emerged preeminent and would have their interests accommodated. During a formative stage of European political development, these different patterns of church-state relations and religious cleavages shaped the strength of religious forces in both politics and society, affecting political parties across the political spectrum and the strength of a religiously-based voluntary sector. Table 1 Employment of mothers by the age of their youngest child and fertility rates (2002) Percentage of mothers employed with children Under 3

Aged 3–6

Fertility rates

Sweden

72.9

82.5

1.65

Austria

80.1

70.3

1.40

Denmark

71.4

77.5

1.72

Netherlands

74.2

68.2

1.73

Belgium

70.4

67.4

1.62

France

66.2

63.2

1.89

Switzerland

58.2

64.5

1.40

Germany

56.0

58.1

1.31

Italy

54.4

51.7

1.26

Finland

32.2

74.7

1.72

Source: OECD, Society at a Glance 2005. Note: these figures overstate the proportion of women who are actually at work in countries such as Sweden or Austria, where women on a lengthy parental leave are counted as employed, whereas they do not count women on care leaves in France, Germany, or Finland as employed, even though they hold onto their jobs while on leave. Thus, while at first glance it appears that Austrian mothers work in very high numbers, more than half are on parental leave and only a small proportion work full-time. Thus, a recent study showed that while 71.9 percent of Austrian mothers with a youngest child under three were counted as “employed,” 40 percent were on parental leave, 17.2 percent worked part-time, and 14.7 percent worked full time (Babies and Bosses: Reconciling Work and Family Life. Volume 2: Austria, Ireland, Japan (Paris, OECD: 2003), 48).

These patterns also influenced law and early public policies for women, children, and families in a way that was indicative of emerging ideologies about gender roles and the appropriate relationship between the state and the family. In turn, the varying policy approaches laid the foundations for the way states would respond to family change and mothers’ employment during the “golden age” of welfare state expansion. In the Nordic countries, a statist family policy tradition reflected the fusion of church and state, which essentially transferred responsibility for the family from the church to secular political authority. Secularization also spurred the individualization of women and contributed to the decline of patriarchal law. Little surprise, then, that in many Nordic countries the state would take an active role in promoting gender

58

equality and dismantling the male breadwinner model later in the 20th century. In continental Europe, by contrast, patriarchal familialism was more entrenched and state involvement in family affairs was circumscribed by a longstanding practice of leaving family welfare to religious authorities and civil society. The exception was in some of the more strongly anti-clerical countries, such as France, where the state took control over this domain and was more actively involved in policies for children and families. In addition, anti-clericalism made the left more willing to challenge the traditional gender ideologies of conservative parties and actors. By contrast, in countries such as Austria, Germany, and the Netherlands, not only did Christian democrats hew to a conservative vision of the family, but left parties were weak challengers to these views. To develop this argument, this chapter first lays out broad differences in gendered welfare regimes, proceeds to a discussion of the historical roots of these regimes, and then discusses policy change during the 1960s and early-to-mid-1970s, a period when countries were responding to the question of mothers’ employment. The paper wraps up with a brief discussion of why change in work-family policies has been slow in many European countries.

3.1

Gendered Welfare Regimes

Feminist research on the welfare state has highlighted various ways to conceptualize and measure the gendered effects of social policy (Orloff 1993; O’Connor et al. 1999). This chapter focuses on policies for mothers’ employment because they address a period of economic vulnerability in women’s lives – that of child-birth and the years of raising (young) children. Enabling mothers to maintain their attachment to paid employment during that time assures access to the more generous benefits and programs of the welfare state, reduces dependence on familial relations, and diminishes the likelihood of poverty in the event of single parenthood. Given the difficulty of securing these protections in private markets, this is precisely where the welfare state can have a tremendous impact on women’s autonomy and well-being. Countries differ significantly in their policies for working mothers. Scholars often label the Nordic countries “woman-friendly” because of their extensive public child

59

Table 2 Percentage of Children in Publicly-Funded Services 0–2 year olds

3–5 year olds (or until school age)

Austria

11

85

Belgium

31

100

Denmark

52 ► 9 (under 1) ► 78 (1–2)

94

Finland

20 ► 1 (under 1) ► 36 (aged 1–2)

67

France

≈ 38 ►27 (children 4 mo-2–1/2 in child care) ► 32 (2 year-olds in preschool)

100

Germany

9 ► 3 Western Länder ► 37 Eastern Länder

90 (2002) ► 88 Western Länder ► 105 Eastern Länder

Italy

7

98

Netherlands

26

91

Norway

29 ► 2 (under 1) ► 40 (aged 1–2)

82

Sweden

43 ► 0 (under 1) ► 65 (aged 1–2)

91

Figures are from 2002–2004. Sources: European Commission, Employment and Social Affairs DG, Second Version: Indicators for Monitoring the Employment Guidelines, 2004–2005 Compendium (Brussels: January 9, 2006); Janneke Plantenga and Melissa Siegel, Position Paper “Childcare in a Changing World” Part I: European Childcare Strategies (2004); France: Nathalie Blanpain, “Accueil des jeunes enfants et coûts des modes de garde en 2002,” Etudes et Résultats 422 (August 2005) and Repères et Références Statistiques sur les Enseignements, la Formation et la Recherche (Paris: Ministère de l’Education Nationale, 2003); Germany: National Action Plan for Employment Policy, p. 91; Nordic country data from NOSOSCO, Social Protection in the Nordic Countries 2002: Scope, Expenditures, Financing (Copenhagen, NOSOSCO 2004), p. 60–1.

care systems and generous parental leave policies. In addition, individual taxation and entitlement to social benefits remove disincentives to women’s labor force participation. Norway and Finland differ in that their supply of public child care is more limited and they both provide long care leaves for parents with children under three (Leira 2006). While Sweden and Denmark also offer lengthy parental leaves, they have rejected these longer care leaves as detrimental to women’s place in paid work.10

10 The difference between a “care leave” and a “parental leave” is the following: care leaves usually pay flat-rate, relatively low benefits (e.g. $500/month), for two to three years for each child. These leaves aim to encourage mothers to leave paid work for a long period of time. Parental

60

Continental Europe is more complicated. Countries such as Austria, Germany, and the Netherlands long exemplified the male breadwinner model: joint taxation of married couples, the nature of benefits systems, and the lack of full-time child care deterred mothers from paid work while their children were young. Many European countries do provide part-time preschool education for nearly all of the eligible age categories (see Table 2), although these programs are often open for only a few hours a day and cannot serve as a reliable form of child care. In recent years, governments in some of these countries have sought to encourage mothers’ employment through supportive public policies. Nonetheless, the imprint of past policy orientations is still quite visible in the low provision of public day care (Table 2). The tax regime in Germany also has not changed, and both Germany and Austria have had a long, ill-paid care leave that encourages mothers to take long absences from paid employment. While Dutch child care provision seems more extensive, nearly all children attend programs part-time, and so the supply of places is actually half that listed in the table. France, Belgium and Italy fit imperfectly in a third category that overlaps with the Christian democratic cluster. These welfare states have maintained a male breadwinner flavor, which is evident in the continuing reliance on joint taxation and entitlement to benefits through marriage in some cases.11 All three long offered a fairly short parental leave (although Italy extended its leave to 10 months in 1999), and France has developed a long paid care leave much like Germany and Austria, the effect of which has been to reduce mothers’ labor force participation (Bonnet and Labbé 1999). However, one commonality of these countries is their universal preschool provision for three-to-six-year-olds and, in the case of France and Belgium, substantial child care subsidies or services for children below the age of three. Preschools are open for eight hours a day in Italy and match the lengthy school day in

leaves are usually paid as a percentage of income and are shorter, ranging from four months in France to 13 months in Sweden. 11 France has mandatory joint taxation of married couples; Belgium has optional joint taxation, whereas Italy does not. See O’Donoghue and Sutherland (1998). Joint taxation depressed married women’s employment because in a steeply progressive tax system, the “second income” that is piled on top of the first is taxed at high marginal rates.

61

Belgium and France.12 Although this does not entirely cover the work day, these programs are free and universally available. How can we explain these patterns of public policy? One approach argues that where social democratic parties are politically powerful, they have created “womanfriendly” welfare states that support mothers’ employment (Esping-Andersen 1999; Huber and Stephens 2000). The Scandinavian countries seem to be proof of the argument, as they have powerful social democratic parties, widely-available child care, and high rates of mothers’ employment. A related view holds that the key for the Scandinavian countries was the mobilization of women within labor unions and social democratic parties, which persuaded these actors to embrace mothers’ employment (Naumann 2005). These studies have much to recommend them, particularly as the extensive public sector of the Nordic welfare states has been a major draw, and support, for working mothers. Nonetheless, juxtaposing one particularly quantifiable outcome – publicly-provided or funded child care provision for under-threes – with social democratic power resources shows a mismatch between social democratic power and child care provision (Figure 1). The historical record also reveals that social democratic parties have not always been very committed champions of gender equality. Instead, these parties have behaved differently on gender issues in different countries, at different time points (Quataert 2001). Rather than assume social democratic parties to be inevitable enthusiasts for mothers’ employment, we need to understand why their behavior has differed in various contexts and time periods. A more fruitful approach takes into account the deeper cultural or historic factors that have shaped attitudes towards mothers’ employment and then influenced the environment within which political parties operate. In the case of mothers’ employment, social class shapes views on this issue such that working class people often hold more traditional orientations towards mothers’ employment than middle-class, white collar workers. For that reason, it is logical that social democratic parties would not be enthusiastic champions of public child care or parental leave policies while their base is largely blue collar. With the decline of the working class, however, these parties usually seek the middle class vote, creating an opening for a more feminist perspective on women’s employment and other issues.

12 Often 8: 30–3: 30 in Belgium, 8: 30–4: 30 in France, sometimes with one afternoon off per week.

62

Figure 1 Social Democratic Political Power and Child Care Provision

Social Democratic power

Children in public child care

90 80 70

percentages

60 50 40 30 20 10 0 Austria

Belgium

Denmark

Finland

France

Germany

Italy

Norway

Sweden

Social Democratic power is a measure of the average percentage of cabinet seats held by social democrats, 1960–2000. The child care measure is the percentage of children under the age of three in publiclyprovided or funded programs. Sources: Evelyne Huber, Charles Ragin, and John D. Stephens, Comparative Welfare States Data Set, Northwestern University and University of North Carolina, 1997; child care data: see table two.

Yet, their ability to do so is likely to be affected by the presence or absence of parties that champion more conservative gender orientations and could politicize the issue of working mothers. If conservative parties have the potential to lure working and middle class voters, this could make it especially difficult for social democratic parties to adopt a more “post-materialist” stance toward gender equality concerns. More generally, the presence of powerful Christian democratic parties in a political system may be indicative of the strength of traditional values in the population and could influence the strategies of their political competitors. Strong agrarian parties could have a similar effect: these parties were important allies of social democrats in Scandinavia in the construction of universalistic social programs, but would not necessarily support policies such as public child care, which their rural voters often had less use for. Thus, while the decline of the agrarian party in Sweden created an opening for policies to promote mothers’ employment, the strength of the agrarian party in Finland had the opposite effect, producing a somewhat more traditional cast to workfamily policies (Hiilamo and Kangas 2005; Bergqvist et al. 1999; Sipilä and Korpinen 1998).

63

The behavior of political parties may also be affected by the landscape of public policies that were created in an earlier time period. Many welfare states have deep historical roots, their foundations having been laid in the late 19th and first few decades of the 20th century. During that time, social democratic parties were often excluded from the political system, while others – including liberal, agrarian, republican, and religious parties – were more significant. Yet, it was in that period that many states first became involved in educational and family policies, including preschool education, maternity leaves, and family welfare programs. These early decisions institutionalized the state’s role in certain policy areas while also establishing norms about state responsibility for child and family affairs. For example, incorporating preschool education into a national education ministry creates both an institutional actor that will defend and expand these programs, but also generates expectations about continuing public sector involvement in this area. More generally, decisions in the first half of the 20th century created a division between the state and often religiouslybased voluntary sector, either squeezing out these voluntary groups or giving them a privileged role in the delivery of education and/or social services (Schmid 1995). To uncover the roots of party systems in Western Europe, and these early public policies for children and families, we concur with the analyses of van Kersbergen and Manow and employ the analysis of Seymour Martin Lipset and Stein Rokkan (1967). Nineteenth-century cleavages and conflicts were central to the creation of modern political systems in Western Europe. While many welfare state scholars have focused on the class cleavage generated by the industrial revolution, the national revolution often produced cleavages over religion that were just as important in shaping European political development. The Reformation, the French Revolution, and the creation and expansion of the nation state all challenged the hegemony of the Catholic Church and generated sectarian divisions in many countries. The resulting conflicts, and their resolution, shaped the structure of European party systems. Where religious conflicts were intense, parties of religious defense generally formed, whereas the absence of these divisions often impeded the creation of such parties. Clericalanticlerical divisions also affected the nature of left and liberal parties, influencing the strength of their anticlericalism and whether they would develop a more accommodating stance towards religious actors.

64

Patterns of religious conflicts also influenced early public policies for women, children, and families. In many countries, conflicts between secular and religious forces – and/or between competing religious groups – crystallized around the development of mass education systems in the late 19th and early 20th centuries. As education was long in the province of churches, the expansion of state responsibility in this area, and the aspiration of liberals that it be wrested away from churches, generated fierce fights in many European countries. There also were disputes over the division of labor between the state and religious charities in other forms of welfare provision (Schmid 1995). Such conflicts did not occur everywhere, however, and different patterns of churchstate relations and religious conflicts shaped the roots of contemporary educational and welfare provisions. In the Nordic countries, the state would peacefully assume responsibility for children’s education and family welfare, directly providing services where deemed necessary. In countries marked by clerical-anticlerical divisions, anticlerical forces pushed for expanded state responsibility for child and family affairs and were most successful in achieving this in France. Religious forces would gain more influence over policy in Belgium and Italy, but would face contestation by the anti-clerical left. In the accommodationist countries, religious forces would preserve their influence, which would be guaranteed in Germany and the Netherlands through the devolution of responsibility for sensitive policy areas (such as education or family welfare policy) to the voluntary sector. In Austria, the central state would preserve a more direct role in education and family policies, but given the weight of Catholics in politics, state policy would protect the interests of the Catholic Church.

3.2

The Religious Origins of Gendered Welfare Regimes

3.2.1 Church-State Fusion and Religious Homogeneity in the Nordic Countries The roots of the Nordic “passion for women’s equality” lie in its distinctive religious heritage (Graubaud 1986). When the Reformation came to Scandinavia in the 16th century, Protestantism became the state religion but the state gained the upper hand. The subordination of religious to civil authority, and the religious homogeneity of these countries, minimized the development of religious cleavages (Martin 1969).

65

Over the course of the 18th and 19th centuries, political authority was gradually secularized. While at times, national churches fought these developments, the Nordic countries were generally spared the kind of strife caused by the religious question in the rest of Europe (Stephens 1979). Instead, a “beneficent circle” of secularization ensued (Martin 1969; Gustafsson 1987). The de facto incorporation of the Lutheran church into the state machinery put these countries on a trajectory of relatively peaceful secularization and political centralization. This is not to imply that these countries were devoid of religious divisions, or that their pathways to secularization were identical. In the late 19th and early 20th centuries, revivalist political movements emerged in reaction to the “spiritual deadness” of the established churches (Madeley 1977). In Norway, this rural revivalism overlapped with center-periphery cleavages and helped ensure a greater politicization of religion. While Denmark and Sweden were already culturally uniform at the time of peasant mobilization in the 19th century, Danish domination of Norway (until 1814) produced a governing elite that endured through the 19th century and resided in the cities, spoke Danish, and attended the high church. This created a combination of economic, cultural, territorial, and religious grievances of peasants in rural areas, which had a lasting impact on Norwegian politics (Rokkan 1966). Sundbeck (1994) argues that Finnish secularization also was less beneficent, reflecting distinctive elements of Finnish state-building. One consequence perhaps is that levels of religiosity have been higher in Finland and Norway than in Sweden or Denmark, even though church attendance had dropped significantly in all of these countries by the mid-20th century. These patterns of religious divisions, or lack thereof, critically shaped Nordic political development in the 19th and early 20th centuries. In Norway, the aforementioned religious and center-periphery cleavages endured through much of the 20th century, producing the only Christian Democratic party of any electoral significance in Scandinavia, the Christian People’s Party (KrF) (Rokkan 1966, 74–8; Stephens 1979, 136–7). The KrF became one of the four main parties on the center-right, serving in governments and at times providing prime ministers for these governments as well. In the rest of Scandinavia, small, Christian-based parties formed in the decades following the Second World War, but with the exception of the Norwegian Christian party these parties have been fairly weak (Madeley 1977: 269–70; van Kersbergen and Manow).

66

More generally, religion came to play an increasingly marginal role in society and politics in the 20th century. Many scholars trace the early and advanced secularization of the Nordic countries to the early fusion of church and state and relatively minor conflicts around religion. Such divisions are usually seen as promoting competition between churches, or between religious and secular authority, and thus institutional vigor. Instead, the established churches of the Nordic countries became bureaucracies that steadily lost influence over the population (Martin 1969; Sundbeck 1995, 87–91). While most people remained members of the established church, actual church attendance plunged.13 The lack of a strong religious cleavage, and weakness of societal religiosity, also affected the political left. The working class was not divided between secular and religious unions and parties, as in continental Europe, enabling Scandinavian social democratic parties to build a unified working class base. The lack of religiously-based mobilization also enabled the greater centralization of political power, something that would be advantageous for social democrats. Across Europe, left parties often favored state intervention over subsidiarity or deference to civil society, and the Scandinavian social democrats were better able to realize this vision. The exception here was Norway, where a religiously-based voluntary sector continued to play an important role in both the education system and the provision of social services (Nicolaysen 2001). The Nordic pattern of church-state fusion, weak religious cleavages, and advanced secularization affected child- and family-related policies of the early 20th century. With the fusion of church and state, the traditional responsibility of churches for family well-being was transferred from the church to the state. While religious actors initially maintained their involvement in family matters as state officials, they lost this influence with the gradual secularization of public authority (Anderson, this volume). This is evident in the development of mass education. In Sweden, for example, the school administration thoroughly fused secular and religious authorities (Sundbärg 1904). As education was progressively secularized, the National Church

13 The high rates of church membership in these countries reflect the close connection between church and state, making citizenship coterminous with church membership. However, while most people belong to church, they hardly ever attend services. See Halman and Riis (1999), 6–7.

67

was edged out of its traditional role in education (Boli, 1989, 68, 84, 139–40; Gustafsson 1989, 166). Although there were some conflicts between Lutherans and secularists over education and other areas, these disagreements in no way replicated the heated, polarizing disputes that occurred elsewhere in Europe. The growing secularization of these societies and polities also shaped family policies early in the 20th century. The Scandinavian countries were pioneers in the development of family laws that assigned equal status to men and women (Therborn 1993; Bradley 2000). For example, marriage reform laws in the first three decades of the 20th century made spouses fully equal in marriage, allowed divorce by mutual consent, and made husband and wife equal property owners. As Wetterberg et al. remark (2002, p. 10) “rationality and science gradually came to replace religion as a base for legislation and moral legitimacy connected to marriage and sexuality.” The declining influence of religion also contributed to the individualization of welfare state entitlements. In the rest of Europe, women’s entitlement to social benefits was usually tied to her marital status. In Scandinavia, by contrast, women were individually entitled to some early social programs through citizenship, and not by their status as wives (Sainsbury 1996, 63–4). These individualizing elements of family policy would develop further in later decades but were important in establishing the principle that women exist independent of their familial relations. At the same time, however, the assumption early in the 20th century was that women were economically reliant on men and that state policy should neither support nor encourage mothers to work. In this, the Nordic countries were no different from any others in Western Europe, but they were unique in that they also lacked the preschool services that were developing in much of the European continent. The lack of educational services for young children reflects the absence of religiously-based disputes over education. Without competition between clerical and anti-clerical forces, or between competing religious groups, there was no imperative to extend the education system beyond six- or seven-year old children – the age at which primary school began. As the succeeding sections will show, in countries with sharper religious cleavages and competition between churches and secularists over education, these conflicts spilled over into the domain of early childhood education and spurred the growth of these programs (Morgan 2002). In the Nordic countries, by contrast, where there was little such competition, such programs remained very underdeveloped

68

while education authorities focused their attention on mass education for children above the age of seven. In sum, secularization and the fusion of church and state produced an active state with regard to family matters, albeit one that had not yet engaged in questions of mothers’ employment and the care of young children. Instead, we can see the potential for state intervention in reshaping gender relations and affecting family life through the individualization of some welfare benefits and pioneering, genderegalitarian changes in family law. These secular polities and societies, with their tradition of state involvement in family affairs, would later be transformed into the “women-friendly” welfare states we know today.

3.2.2 Clerical-Anticlerical Divisions in Belgium, France, and Italy Belgium, France, and Italy are Catholic countries with a long history of conflict between clerical and anticlerical forces. The tug-of-war between clericalism and anticlericalism is reflected in a policy profile that mixes the subsidiarity and familialism typical of Christian democratic regimes with active state involvement in the care and education of young children. The balance between these two orientations varies by country according to the relative influence of Catholic and secular political forces. While the Italian welfare state came to be characterized more by “passive familialism,” the sharp clerical-anticlerical divide in France and Belgium produced a greater state role in children’s lives and family affairs. In addition, anti-clericalism shaped left parties such that they would often challenge conservative parties on education and family issues, distinguishing them from their counterparts in other continental European countries. In addition, the weakness of Christian Democracy in France fragmented a repository of traditional gender ideologies on the right, enabling a pragmatic approach to working mothers later in the 20th century. For much of the 19th and 20th centuries, clerical-anticlerical divisions were dominant features of political and social life in France, Belgium, and Italy. The French Revolution brought the first, most violent attack on the Catholic Church, and the reverberations of the Revolution throughout the 19th century spurred challenges to clerical authority. These conflicts shaped the development and ideologies of left and right par-

69

ties. Belgium and Italy were typical of the rest of continental Europe in that the growing friction between religious and secular authority generated political movements designed to preserve the influence of the Catholic Church (Lipset and Rokkan 1967, 15). This initially took the form of voluntary associations dedicated to winning the Church’s goals through social welfare activities, but religion also became the basis of some early political parties. These religious movements evolved into the Christian democratic parties that dominated politics in the post-World War Two era, accompanied by a network in civil society of religiously-based associations, clubs, newspapers, etc (Kalyvas 1996). In France, by contrast, Christian democracy struggled to develop under the opposition of intransigent, anti-democratic forces on the right, the anticlericalism of much of the rest of the political spectrum, and a majoritarian electoral system. Together, these factors impeded the development of a moderate, Christian party early in the 20th century (Irving 1973; Kalyvas 1996; Manow and Palier, this volume). The participation of Catholics in the resistance made possible the birth of the MRP after 1945, and its continued existence was facilitated by a PR electoral system. As Manow and Palier note (this volume), the MRP was a frequent governing party throughout the 1950s and was able to put a strong stamp on the French welfare state. Yet, the MRP collapsed with the French Fourth Republic and the return to a majoritarian electoral system in the Fifth Republic (1958-present). The dominant party on the right became a Gaullist one that violated many of the tenets of Christian democracy. De Gaulle was technocratic and statist in his orientation towards public policy, in opposition to the Catholic principles of subsidiarity. He was also anti-American and suspicious of European integration. This hardly sat well with Christian democratic sympathizers at the time, although they often voted for De Gaulle and his party for lack of better alternatives (Irving 1973, 231). In addition, France did not develop the Catholic civil society found in Belgium and Italy. There was a religious subculture in certain parts of France (such as Normandy and Brittany), which included Catholic voluntary organizations (Berger 1987). Such organizations were more limited in the rest of France, reflecting the weak reach of Catholicism into large parts of the country. One important set of voluntary actors would be family associations, some of which were religiously based. These groups emerged in the late 19th and early 20th century and championed policies for large

70

families (such as family benefits). Similar associations developed in Belgium, but not Italy (Martin and Hassenteufel 1997). The secular-religious divide infused left parties with a potent anti-clericalism. In Belgium, both the socialist and liberal parties have been strongly anti-clerical, with the liberal party focused on the issue of secular education (Gérard-Libois and Mabille 1981, 128; Lorwin 1966, 147).14 Liberals and left parties in Italy were anticlerical and would often mobilize around educational or family issues (Halperin 1947). In France, the Socialist and Communist parties inherited the anti-clerical mantle and fought public funding for Catholic schools. As in Belgium and Italy, the emotions aroused by the schools issue in the post-war decades is indicative of the continuing anticlericalism of liberal and left parties (Orlow 2000). These political conflicts shaped educational and family policies in the late 19th and early 20th centuries. In both Belgium and France, competition for control over education focused largely on primary schools but also came to concern the preschool programs run by the Catholic orders. Rather than leave any educational services in the hands of the Catholic Church, anticlerical Republicans in Belgium and France took control of these programs in the late 19th century, incorporating them into national education systems as non-mandatory services (Dajez 1994, 161–5; Mallinson 1963, 83– 5; Delhaxhe 1989, 24–5). Henceforth, the education of children aged three-to-six would be a state responsibility, although the implementation of this vision would differ based on the relative power of Catholic political forces. In Belgium, Catholics regained political control in 1884 and decentralized control over the education system so that local communes could decide whether to finance a Catholic or secular school (Mallinson 1963, 99–102). This meant that religious voluntary associations would continue to play a major role in the provision of publicly-funded education (Fix 2000) and ensured that there would be fierce competition between laic and clerical groups over educational provision for the succeeding decades, which in turn helped bring about an expansion in the total number of services available. In France, by contrast, secular republicans won their battles with religious forces, and a “reverse principle of subsidiarity” developed in education and social welfare (Ar-

14 While a 1958 compromise on the school issue helped mitigate tensions, education continued to be a sensitive issue through at least the 1960s (Mallinson 1967, 276).

71

chambault 1996, 15–17; Manow and Palier, this volume). Both pre- and primary education became an entirely state-run service, for example, and the congregations who had traditionally run these programs were initially barred from any role in the education system. After decades of conflict, a 1959 law finally allowed publicly-funded religious schools to exist alongside the secular, public ones, although these schools remained far less prevalent than in neighboring Belgium.15 Initially, there was a more collaborative relationship between the state and civil society in the area of social welfare, with charities continuing to run crèches and other welfare services in parts of the country (Weiss 1983; Smith 1998). Nonetheless, the expansion of the welfare state in 1945 involved a substantial increase in state responsibility for family welfare, and a marginalization of the voluntary sector’s role in the provision of crèches and other family-related programs. Clerical-anticlerical conflicts also spurred the expansion of family benefits in France. As declining fertility rates in the late 19th and early decades of the 20th century spurred fears of depopulation in France, conservative Catholics blamed republican attacks on religion and traditional values (Talmy 1962, 60–1). The republican response was pronatalism – a secular, nationalist doctrine that promoted large families in the name of national greatness (Antomarchi 1996, 41; Lenoir 2003, chp. 4). As anxieties over population decline mounted, Catholic conservatives and Republican pronatalists made common cause about the need for generous family policies, including aid to large families, tax measures, efforts to promote infant and maternal health, and, in the 1930s and 1940s, a system of generous family allowances that gave a higher per child benefit to large families (Antomarchi 1996, 39–40; Pedersen 1993). These policies thus signified a further expansion of the state’s role in the lives of families. Belgium developed similar family policies at roughly the same time, reflecting demographic and economic conditions after the First World War, as well as the push by Catholic and socialist laic family associations for such policies (Martin and Hassenteufel 1997, 29–31, 116).16

15 In the mid-1990s, about 60 percent of students in Belgium attended Catholic schools; in France, less than 14 percent of preschool and primary students are in private schools today. 16 The Catholic groups preferred state funding to support charitable initiatives, while the socialist associations pushed for an expanded state role in providing benefits and services to families.

72

In Italy, by contrast, Catholics were better able to preserve their control over educational and social welfare services than they were in France, and there was no similar development of family benefits programs. In the late 19th and early 20th centuries, anti-clerical liberals created a public education system (in which religious teaching would be optional), repressed religious congregations, and secularized the institution of marriage (Halperin 1947, 22–4). Their larger drive to regulate the Church’s vast network of charities would ultimately fail, however, reflecting weaknesses in the Italian state as well as the lack of an ambitious social agenda among the reigning liberals (Quine 2002, 293–5; Lynch, this volume). Responsibility for social welfare programs, including the precursor to the modern preschool, was largely left to religious charities (Pistillo 1989, 157). Although the fascist regime later sought to develop a more coherent set of family and education policies – largely in the service of pronatalism and inculcation of fascist values – this was done in a way that accommodated Church interests. For example, the 1929 Concordat with the Catholic Church required religious instruction in public elementary and secondary schools, and although the fascist regime prioritized preschool education for the first time, the religious orders preserved their monopoly over the scuole materne (Wolff 1980; Marraro 1936, 46–52). The Church would maintain its hegemony over preschools and social welfare programs until well into the 1960s, when it would finally face a spirited challenge from the left (discussed more below). Although secularist republicans challenged Church authority over the socialization of children, left political forces were often lackluster advocates for women’s issues. In France, for example, republicans at times contested women’s legal subordination but did not embrace this cause with the enthusiasm they brought to the education issue. Republican governments in the first decades of the Third Republic brought some changes in family law that improved women’s rights, but many patriarchal elements of the Napoleonic code remained on the books for many more decades, as they did in Belgium (Antomarchi 1996, 39; Marques-Pereira and Paye 1998, 111). One reason may be that republicans in these countries believed women were frequently under the sway of the Catholic Church. This was all the more reason to make sure that their children received a secular education as early as possible, but did not produce much interest in extending women’s political rights (as it was assumed they would vote for Catholics). Instead, republicans embraced familialist policies with luster in both

73

France and Belgium, creating generous family allowances that employed the family, and not the individual, as the defining unit of social policy. Notably, France, Belgium, and Italy were among the last countries in Western Europe to grant women the right to vote. To sum up, clerical-anticlerical divisions produced an expanded state role in education that shaped the development of preschools in Belgium and France. These programs would later be an important set of services for working parents, albeit ones that were seen primarily as education rather than day care. Italy began developing a similar set of services during this period, although churches were more successful in preserving control over these and other social welfare programs. However, although republicans could agree that Church influence should be diminished in the socialization of children, they did not attempt a wider challenge to patriarchal familialism. Indeed, republicans “laicized” the family in their competition with Catholics over family policy, but they eschewed the individualist conceptions that were emerging in Scandinavia. Other legacies of the clerical-anticlerical divisions of the early 20th century include the weakness of Christian democracy in France and the strength of anticlericalism in all three countries.

3.2.3 The Accommodation of Religious Forces in Austria, Germany, and the Netherlands Germany and the Netherlands resemble Belgium in that they both experienced religious conflict in the 19th and early 20th centuries, and a resulting mobilization of religiously-based voluntary associations and political parties. They differ in that conflicts were not solely between Catholics and anti-clericals, as in Belgium, but also between Protestants and Catholics. They also differ in the way these conflicts were resolved. Ultimately, religious forces gained the upper hand in these polities, either as a coalition of Calvinists and Catholics (in the Netherlands) or through a predominantly Catholic Christian Democratic Party (In Germany). Even with their disagreements, Protestants and Catholics were unified around the goal of preserving control over social and educational services. Secular forces were outnumbered, and as the intensity of anticlericalism faded, the left lost its anticlerical bite and struck up a more accommodationist relationship with religious parties. We can see a similar outcome in

74

Austria that was reached through a somewhat different path. Anti-clericalism would ebb relatively quickly and the state maintained a collaborative relationship with the Catholic Church in education and other matters. The result was that religious voluntary organizations would develop much less in the educational and welfare spheres, in part because religious forces had secured their place in Austrian politics earlier than in the Netherlands or Germany. As in the rest of continental Europe, the development of mass education sparked conflict between religious and secular forces. From the mid-19th century until 1920, the school question was an endemic source of antagonism in Dutch political life that pitted both Catholics and orthodox Protestants against liberals. In Germany, disputes over education erupted when the National Liberal party led a drive for complete state sovereignty in education in the 1870s. In Austria, there were similar moves by liberals in the mid-19th century to augment the central state’s responsibility for education and reduce the power of the Catholic Church over this domain. Religious and class cleavages spurred the growth of sub-cultural organizations that become the bedrock from which political parties would later emerge. Throughout much of the 19th century, religious congregations flourished in the Catholic regions of the Netherlands, creating educational and social welfare services for the poor (van Vugt 1991). Protestant and secular equivalents formed as well, laying the basis for the pillarization of Dutch society into four sub-communities – Catholic, Protestant, liberal, and socialist (Lijphart 1984). In Germany, similar Protestant and Catholic charities proliferated after 1830, spurred in part by competition with each other (Dickinson 1996, chps. 1–2). Many of these organizations were later incorporated into two peak associations representing each confession, and a social democratic peak association would form as well. In Austria, three subcultures emerged in the 19th century that would produce political parties and a network of related voluntary organizations: Catholic-conservative; social democratic; and anti-clerical national liberals (Luther and Müller 1992, 1–3). Religious cleavages also led to the creation of religious parties that would dominate politics for much of the 20th century. In the Netherlands, separate Catholic and Protestant parties formed and were often allied with each other, but remained formally separate for most of the 20th century. Clashes over education in the 1870s were one

75

factor driving the formation of the Anti-Revolutionary party in 1879, a party of Orthodox Protestants. The first Catholic party was formed in 1926, but Catholics had already made common cause with the ARP in the late 19th century, a move that enabled them to gain political power and give public funds to non-state religious schools (van Kersbergen, this volume). In Germany, the Catholic Zentrum party formed in 1870 to protect the interests of the Catholic minority. After 1945, the division between Protestants and Catholics was overcome with the formation of the Christian Democratic Union-Christian Social Union (CDU-CSU). In Austria, the Catholics became one of the main political actors during the First Republic, and they would return to that role after the Second World War. The strength of religion in these societies and polities also affected the left. Initially, left parties were strongly anticlerical, but the Austrian, Dutch, and German parties lost their anti-clerical edge. In the Netherlands, religious conflict diminished with the resolution of the schools struggle in favor of equal state support to privately-run schools – a pact between liberals, socialists, and the religious parties that was included in the 1917 Constitution.17 This settlement essentially ended the conflict over education, in contrast to France and Belgium where the issue festered for several more decades. Moreover, in contrast to nearly all other Socialist parties in Europe, the Dutch socialists had already accepted the principle of confessional schools because they did not wish to alienate religiously-minded members of the working class, particularly those living in areas influenced by Orthodox Protestantism (Hansen 1976). There were renewed tensions over education in the 1950s, but the relative mildness of the Dutch social democrats on these questions contrasts with the anti-clericalism of French socialists in the same period (Orlow 2000, 212–13). The German SPD maintained its anti-clericalism through the 1920s, and SPD governments during the Weimar Republic attempted to re-assert control over education and welfare services from the voluntary sector. They ultimately failed in this regard, and the 1922 Youth Welfare Act was a triumph for the Christian organizations, as it enshrined the principle of subsidiarity (Dickinson 1996, chp. 6–7; Hong 1998). Through much of the 1920s, Heinrich Brauns, a member of the Catholic Zentrum

17 The socialists got universal male suffrage in return for their support of state-funded religious education.

76

Party, served as the Minister of Labor and he worked to construct a system of publicly-funded welfare services that would be provided by religiously-based voluntary associations. In the first decade after World War Two, the SPD resumed its traditional anticlericalism, but it abandoned this stance at the 1959 Bad Godesberg party meeting. This would enable the SPD to make important inroads among the Catholic working class, which had previously voted almost entirely for Christian Democrats (Spotts 1973, 338–46). In Austria, the social democrats also softened their stance towards religion by the 1950s, as they sought to attract peasants and middle class voters (Secher 1959, 296–7). The political preeminence of religious forces would also shape educational and family policies. For one thing, the subsidiarity notion was institutionalized in the welfare state, as social services and some educational functions were delegated to religious and other voluntary organizations. In Germany, this gave a great deal of power to the two largest, religious peak associations – the Protestant Diakonisches Hilfwerk and the Catholic association, Caritas. For decades following the Second World War, no social policy could be passed if the two churches were in unanimous opposition to it (Spotts 1973, 286). In the Netherlands, voluntary organizations became preeminent in the delivery of publicly-funded social services (Brenton 1982). This contrasts markedly with France, where centralized provision of services was long the norm. Austria differs in that religious organizations did not face the imperative of developing alternative social and educational services. Instead, the state provided more of these services, and one consequence was a lesser development of early education programs. In the Netherlands and Germany, by contrast, voluntary organizations were more active in providing early education programs, but the part-day or part-week schedule of these programs ensured they would be unhelpful to working parents (Fix 2000). As in the rest of continental Europe, these three countries did not reform many patriarchal elements of family law for several decades after the Second World War. For example, until 1977, the German civil code explicitly depicted women as homemakers, and women had a legal duty to fulfill their home duties before they could justify working outside (Hantrais 1994, 147). In the Netherlands, there were similar measures that denied women legal personhood and that were removed in the decades after the Second World War. The political dominance of confessional parties also re-

77

sulted in restrictive laws on abortion, homosexuality, and other elements of sexual morality until the 1970s (Bussemaker 1998a, 29). In sum, religious divisions in these countries laid the bases for the confessional parties that would dominate politics in these countries. These divisions also helped produce a strong, often religiously-based voluntary sector in the Netherlands and Germany that would be key to the delivery of social and educational services in these welfare states. In Austria, the state maintained greater responsibility for these programs but policy was shaped by the conservative familialism espoused by the ÖVP that went relatively unchallenged by the SPÖ. The result in all three countries would be conservatism with regard to gender roles and the family, but also a lack of left contestation.

3.3

Consequences during the Expansion of the Welfare State

3.3.1

Welfare State Expansion: 1945–1975

In most West European countries, the decades following the Second World War were the pinnacle of the male breadwinner model, which was reflected in and sustained by the social policies of the period. These policies included the mandatory joint taxation of income that discouraged mothers’ employment; high male wages assured through negotiated wage agreements or social conventions; eligibility for social security programs that reflected the assumption of female reliance on male incomes; and, in some countries, generous family allowance programs. Public child care was limited, reflecting the belief that children required full-time maternal care in the early years. In much of Western Europe, these policies hardly changed throughout the period of rapid welfare state expansion (roughly 1945–75). Yet, starting in the mid-to-late 1960s, a few countries – the Nordic states, along with Belgium and France – began developing more supportive policies for working mothers, such as public child care and improved maternity or parental leaves. Sweden went furthest in dismantling the male breadwinner model by abolishing mandatory joint taxation in 1971 and fully individualizing entitlements to social benefits, both of which created incentives for married women to be in paid employment. The other Nordic countries pursued simi-

78

lar policies, although change was slower in Norway and Finland. France and Belgium also preserved some more traditional dimensions to family policy. These shifts reflected both policy legacies and the way in which religion shaped the party systems in these countries. In terms of policy legacies, France and Belgium had inherited an early education system that could be employed to meet the needs of working parents. By the 1970s, preschool provision was nearly universal in both countries and the long school day covered much of the time parents were at work. Both also had institutions for the provision of family benefits. If pushed to do so, funds for these programs could be redeployed to build day care centers or offer subsidies to parents seeking child care help. In the Nordic countries, the state had similarly assumed responsibility for the education and welfare of families, rather than leaving it to voluntary organizations. Although preschool and child care services were very limited at this time, there was little doubt that when new family needs arose, it was the responsibility of the state to deal with them. In countries such as West Germany or the Netherlands, by contrast, the idea of state involvement in family affairs was viewed as unwelcome intervention in the private sphere. Voluntary organizations were a protective layer between states and families through which social programs should be mediated – a layer that was largely lacking in the Nordic countries and France. The other key factor was the way religion influenced the societies and party systems in these countries. Among western countries, religious belief and participation was lowest in France and the Nordic countries (World Values Survey 1981), which made the issue of mothers’ employment less controversial than it would be in much of continental Europe. The lack (or weakness) of religious parties also affected the spectrum of political competition and strategies of political parties. In the Nordic countries, social democrats did not have to compete with religious parties for the working class vote, which strengthened their position in politics. It also facilitated a shift in party strategy with the decline of agrarian parties – long an ally of social democrats – and growth of a large middle class that held more gender egalitarian views. Social democrats could reach out to this expanding population of white collar, often public sector workers, without fear of losing their more traditional, working class base to a religious party (Morgan 2006). They did so by reversing their own, long-standing support for the male breadwinner model and embracing the aim of transforming

79

these societies into ones in which all mothers were in paid work and fathers did a greater share of care work. Where traditional gender ideologies were stronger and represented in the party system, such a shift was made more difficult. Norway, for example, had the strongest Christian party in the region that served in a number of governments in this period. This party politicized the issue of mothers’ employment and helped unify the conservative parties against expansions in public child care (Karvonen 1994; Morgan 2006). In Finland, efforts by the social democrats to expand public child care were stymied by a powerful Agrarian party that championed the interests of rural families, many of whom preferred to receive generous care benefits rather than a place in public day care (Hiilamo and Kangas; Bergqvist et al. 1999). In France, center-right parties governed throughout the 1960s and 1970s and so there was little interest in trying to fundamentally transform gender roles, as was taking place in parts of Scandinavia. In contrast to the Christian democrats governing in other European countries, however, French Gaullists were often pragmatic on questions of gender roles. If women wanted to work and this could be shown to benefit the economy, then center-right policy-makers were inclined to help them, particularly given widely-shared assumptions that family well-being was the state’s responsibility (Morgan 2003). Thus, by the early 1970s, family benefits funds were being redeployed to expand public child care, and in the coming decades France would develop various services and subsidies that helped parents with their child care needs. At the same time, however, some of the more traditional aspects of family policy – such as generous family allowances and joint taxation – remained. This reflects the fact that while conservative forces would not gain preeminence in shaping French family policy, they would not be eliminated either. Conservative family associations (often Catholic) would continue to play a role shaping public policy although they would have to compete with Protestant, secular, and leftist family associations for influence, as well unions that were strongly committed to women’s employment (Martin and Hassenteufel 1997, 45–6). As a result, French family policy balances competing orientations rather than being dominated by either a progressive or traditional orientation. Belgian family policy reflects a similar balance of forces, with conserva-

80

tive family groups well-placed in the decision-making process but lacking decisive control over policy (Martin and Hassenteufel 1997, 33–5; Marques-Pereira and Paye 1998). What differentiates both France and Belgium from most of their neighbors, however, is the fact that left parties became advocates of working mothers by the 1970s. After a push by the Belgian Socialist Party to reform family policy, a 1968 government accord between the socialist and Christian social party determined that public policy should offer both day care for working mothers and financial support to those at home. Similar agreements reached in the early 1970s paved the way for the expansion of the public day care system by also promising to pay some extra family allowances. In the following years, the forces backing day care had the upper hand in these debates and managed to stave off calls for the creation of a special allowance for housewives (Dubois, Humblet, and Deven 1994, 39). Access to public child care expanded, albeit at a slower pace than in France. Nonetheless, the fact that Belgian socialists and trade unions challenged traditional values and got some resources directed toward public child care contrasts with the situation in both Germany and the Netherlands. In Italy, the left also challenged Church control over preschools and social welfare services, and became an advocate for child care, parental leave, a shortened work week, and other policies to help reconcile work and family life (Bimbi and Della Sala 1998, 177). A resurgence of anti-clericalism in the 1960s shaped battles over the public preschool system, as republican, liberal and socialist forces successfully pushed for a nationally funded preschool system that could compete with the existing programs run almost entirely by the Catholic Church (Bimbi and Della Sala, 1998, 181–2; Pistillo 1989, 161). Preschool services would expand in succeeding decades, being valued mostly for their educational dimensions. In 1971, the mobilization of the left also spurred passage of a mandatory five-month maternity leave, as well as a law that defined child care as a social right and transferred a portion of payroll taxes to the regions to develop day care for under-threes (Saraceno 1984, 11; Bimbi and Della Sala 1998, 185). Funds were limited, however, and great regional variations in child care provision developed, reflecting the strength of left power in different parts of Italy. By 1986, for example, 19 percent of children in communist-dominated Emilia-

81

Romagna were in public day care, compared to two percent in Calabria, Molise, or Sicilia (European Commission Network on Childcare 1990).18 In Austria, Germany, and the Netherlands, the lack of interest in policies that would support working mothers reflected the political power of Christian democratic parties and, in the German and Dutch cases, the influence of the religiously-based voluntary sector (Tietze et al. 1989, 64). In the FRG, the CDU-CSU dominated the political scene between 1945 and 1969 and created social programs that reflected Catholic doctrines about justice, solidarity, and the search for a middle course between socialism and free market capitalism (Daly 1999). This vision of social obligations was predicated on the view of the patriarchal family as the basic unit of society that should be protected and sustained. Generous social benefits were designed to help men support their families and the tax code heavily privileged single-earner households (Hantrais 1994, 147; Moeller, 127–8). Social Catholic values not only justified a traditional gender order, but served to block central government intervention in the protected sphere of the family. Per the principle of subsidiarity, social welfare functions were to be met at the lowest possible level, preferably families and voluntary associations (van Kersbergen 1995; Daly 1999). We can find a similar set of policy arrangements in Austria and the Netherlands, although the Austrian voluntary sector was less developed in the welfare and educational sectors (van Kersbergen and Becker 1988). In this context, there was little support for the idea of public child care or other policies that would encourage mothers to work. Mothers’ employment was widely viewed as a sign of government failure; if mothers should have to work, clearly the state was not doing enough to support male breadwinners (Bussemaker 1998b; Kolbe 1999). The left also did little to promote change in the status quo. In 1973, a coalition government dominated by the left came to power in the Netherlands and published the first paper on women’s equality policy in 1977. However, this paper did not offer any discussion of women’s labor force participation and described day care as appropriate only for single mothers (Bussemaker 1998b; Clerkx and Van IJzendoorn 1992, 66). In Germany, the SPD was similarly cautious on this issue, and even though

18 In Bologna, where the communist party has been powerful, 40 percent of children under age three were in public day care by the 1990s, compared to six percent nationally (Corsaro and Emiliani 1992, 95).

82

SPD-FDP coalition governments ruled through the 1970s, there was little change in policy. In Austria, social democrats ruled alone throughout the first half of the 1970s, but they did not endorse public day care or mothers’ employment (Fix 1998, 27). Thus, while some of the more patriarchal elements of family law were reformed under social democratic government in the 1970s, and there was an expansion of partday preschool access, there was little acceptance of the idea that mothers could work outside the home while their children were young.

3.3.2 The Enduring Power of Early Policy Choices Since the 1970s, European countries have, of course, changed a great deal. One notable social change has been secularization. This is particularly striking in the Netherlands, a country in which religiosity was once quite high, but has experienced a steep drop in religious practice and resulting erosion of the traditional pillars of identity and political organization. Similar drops in church attendance and religious belief have occurred in other continental European countries. At the same time, rates of women’s labor force participation have risen everywhere in Western Europe, and while there are signs of convergence in these data, there remain important differences in rates of mothers’ employment. Given these facts, we might expect considerable changes in policies for working mothers, with growing acceptance of maternal employment and support for a state role in helping parents balance work and family. Indeed, there have been changes in these policies, with an emphasis on helping working mothers, since the 1990s. The Netherlands is a particularly dramatic example of these changes, as government policy-makers finally accepted that mothers would and probably should be in paid employment, at least part-time. The Norwegian government also committed substantially greater resources to public child care in the 1990s, producing a significant increase in service availability. In 2003, France underwent a major debate about family policies, and emerged from these debates with a renewed financial commitment to support both mothers who wish to stay home for several years, and those who work outside the home.

83

Nonetheless, there are striking continuities in public policy, as evidenced by significant differences in public child care provision and parental leave policies. Germany and Austria continue to do little to support working mothers, and the Dutch model represents a modernized variant of the male breadwinner model – the one-and-ahalf-earner model in which nearly 90 percent of mothers with young children work part-time. While there clearly have been new policy developments, they look much like the “bounded change” described by Kathleen Thelen (2003) – change that is evident yet still bounded by past policy arrangements and political commitments. Modernizing the male breadwinner model has been the preferred solution rather than a radical overhaul (Gottfried and O’Reilly 2002). In part, these continuities reflect the consequences of economic decline and fiscal strain on the welfare state. Slower-growing economies and maturing pension and health care commitments have made it more difficult for countries to envision bold new initiatives in social policy. Instead, governments devote much of their time to satisfying the demands of existing beneficiaries. In countries where responsibility for child care and parental leave was incorporated into the heart of the welfare state in an earlier period, those services and programs became part of mass expectations of state responsibilities. This, in turn, fueled the expansion of these programs, and prevented states from neglecting the promises they made to middle class parents for help with work-family issues. In countries lacking such a well-established notion of state responsibility for this area, governments have been better able to avoid taking on such burdens. This continuity also is testament to the enduring legacies of historic patterns of religious divisions and politics. Even with the decline in religious belief, there remain substantial differences between countries in the values held by the public about gender roles in general and, specifically, the appropriateness of mothers’ employment. With the passing of time, and decline of religiosity, these differences are decreasingly related to individual religious practice, but are reflective of the experiences of people themselves with work and family issues (Morgan 2006). Where mothers commonly work while their children are young, and where this has been an established practice for many decades, the acceptability of mothers’ employment is high. Where the opposite is true, or if mothers of young children only recently began working in large numbers, skepticism endures about the merits of working mothers. This continues to

84

shape the climate of debate about child care and parental leave policies. In this way, deep historical factors that shaped party ideology and public policy continue to influence beliefs, practices, and the politics of mothers’ employment.

3.3.3 Conclusion We can trace the origins of contemporary policies for working mothers to religious conflicts and cleavages of the late 19th century. These divisions shaped both party systems and early public policies for children and families, creating a tradition of active state involvement in children’s education and family affairs in some countries, and passive familialism in others. In the Nordic countries, secularism also would contribute to the individualization of women with regard to social benefits and law. Such changes would come later, and incompletely, to continental Europe, where religion endured as a more important political and social force. Nonetheless, in countries such as France and Belgium, where anticlerical forces challenged the hegemony of the Catholic Church, the state would expand its responsibility for children’s education and family well-being and later prove pragmatic, rather than moralizing, on the issue of working mothers. In much of the rest of continental Europe, by contrast, public policy would discourage mothers from working while their children were young, an approach that has deep historic roots and has proven slow to change.

Works Cited Antomarchi, Véronique. 1996. “La famille et le Parlement de 1870 à 1914.” Recherches et Prévisions 44 (June): 37–42. Archambault, Edith. 1996. Le Secteur sans but lucratif: Associations et Fondations en France. Paris: Economica. Berger, Suzanne. 1987. “Religious Transformation and the Future of Politics.” Pp. 107–51 in Changing Boundaries of the Political: Essays on the Evolving Balance between the State and Society, Public and Private, in Europe ed. Charles Maier. Cambridge: Cambridge Univ. Press. Bergqvist, Christina, Jaana Kuusipalo, and Auður Styrkarsdóttir. 1999. “The Debate on Childcare Policies.” Pp. 137–57 in Nordic Democracies: Gender and Politics in the Nordic Countries eds. Bergqvist et al. Oslo: Scandinavian University Press. Bimbi, Franca and Vincent Della Sala. 1998. “L’Italie: Concertation sans représentation.” Pp. 173–202 in Qui doit garder le jeune enfant? eds. Jane Jenson and Mariette Sineau. Paris: Librairie Generale de Droit et de Jurisprudence.

85

Boli, John. 1989. New Citizens for a New Society: The Institutional Origins of Mass Schooling in Sweden. Oxford: Pergamon. Bonnet, Carole and Morgane Labbé. 1999. “L’activité professionnelle des femmes après la naissance de leurs deux premiers enfants: l’impact de l’allocation parentale d’éducation,” Etudes et Résultats. Bradley, David. 2000. “Family Laws and Welfare States.” Pp. 37–66 in Nordic Model of Marriage and the Welfare State eds. Kari Melby et al. Copenhagen: Nordic Council of Ministers. Brenton, Maria. 1982. “Changing Relationships in Dutch Social Services.” Journal of Social Policy 11, no. 1: 59–80. Bussemaker, Jet. 1998a. “Gender and the Separation of Spheres in Twentieth Century Dutch Society: Pillarisation, Welfare State Formation and Individualization.” Pp. 25–37 in Gender, Participation and Citizenship in the Netherlands eds. Bussemaker and Rian Voet. Hants, England: Ashgate Publishing Ltd. Bussemaker, Jet. 1998b. “Rationales of Care in Contemporary Welfare States: The Case of Childcare in the Netherlands.” Social Politics 5 (Spring): pages. Clerkx, Lily E. and Marinus H. Van IJzendoorn. 1992. “Child Care in a Dutch Context” Pp. 55– 80 in Child Care in Context: Cross-Cultural Perspectives, eds. Anders G. Broberg, C. Philip Hwang, Michael E. Lamb, Kathleen J. Sternberg. Hillsdale, NJ: Lawrence Erlbaum Associates. Corsaro, William A. and Francesca Emiliani. 1992. “Child Care, Early Education, and Children’s Peer Culture in Italy.” Pp. 81–118 in Child Care in Context: Cross-Cultural Perspectives, eds. Anders G. Broberg, C. Philip Hwang, Michael E. Lamb, Kathleen J. Sternberg. Hillsdale, NJ: Lawrence Erlbaum Associates. Daly, Mary. 1999. “The Functioning Family: Catholicism and Social Policy in Germany and Ireland.” Comparative Social Research 18: 105–33. Dajez, Frédéric. 1994. Les origines de l’école maternelle. Paris: Presses universitaires de France. Delhaxhe, Arlette. 1989. “Early Childhood Care and Education in Belgium.” Pp. 13–37 in How Nations Serve Young Children: Profiles of Child Care and Education in 14 Countries eds. Patricia P. Olmsted and David P. Weikart. Ypsilanti, MI: High/Scope Press. Dickinson, Edward Ross. 1996. The Politics of German Child Welfare from the Empire to the Federal Republic. Cambridge: Harvard University Press. Dubois, Alain, Perrine Humblet, Fred Deven. 1994. “L’accueil des enfants de moins de trois ans,” Courrier hebdomadaire #1463–1464. CRISP. Esping-Andersen, Gøsta. 1999. The Social Foundations of Postindustrial Economies. Oxford: Oxford University Press. European Commission Network on Childcare. 1990. Childcare in the European Community, 1985–90. Brussels: Commission of the EC. Evers, Adelbert, Jane Lewis and Birgit Riedel. 2005. “Developing Child-Care Provision in England and Germany: Problems of Governance.” Journal of European Social Policy 15, no. 3 (August): 195–209. Fix, Birgit. 1998. The Institutionalization of Family Welfare: Division of Labour in the Field of Child Care in Austria and Germany. Mannheimer Zentrum für Europäische Sozialforschung, Arbeitsberech I, Nr. 24.

86

Fix, Birgit. 2000. “Church-State Relations and the Development of Child Care in Austria, Belgium, Germany, and the Netherlands.” Pp. 305–21 in Families and Family Policies in Europe: Comparative Perspectives eds. Astrid Pfenning and Thomas Bahle. Frankfurt am Main: Peter Lang. Gérard-Libois, Jules and Xavier Mabille. 1981. “Belgian Electoral Politics.” Conflict and Coexistence in Belgium: The Dynamics of a Culturally Divided Society ed. Arend Lijphart. Berkeley: Institute of International Studies. Graubaud, Stephen R. 1986. Norden: The Passion for Equality. Oslo: Norwegian University Press. Gottfried, Heidi and Jacqueline O’Reilly. 2002. “Reregulating Breadwinner Models in Socially Conservative Welfare Systems: Comparing Germany and Japan.” Social Politics 9, no. 1: 29– 59. Gustafsson, Gorän. 1987. “Religious Change in the Five Scandinavian Countries, 1930–1980.” Comparative Social Research 10: 145–81. Halman, Loek and Ole Riis. 1999. “Contemporary European Discourses on Religion and Morality.” Pp. 1–18 in Religion in a Secularizing Society: The European’s Religion at the End of the 20th Century, eds. Halman and Riis. Tilburg: Tilburg University Press. Halperin, S. William. 1947. “Italian Anticlericalism, 1871–1914.” Journal of Modern History 19, no. 1 (March): 18–34. Hansen, Erik. 1976. “Marxism, Socialism and the Dutch Primary Schools.” History of Education Quarterly 13, no. 4 (Winter): 367–91. Hantrais, Linda. 1994. “Comparing Family Policy in Britain, France and Germany.” Journal of Social Policy 23, no. 2: 135–60. Hiilamo, Heikki and Olli Kangas. 2005. “Trap for Women or Freedom to Choose? Child Home Care Allowance in Finnish and Swedish Political Discourse.” Paper presented at RC-19. Hong, Young-Sun. 1998. Welfare, Modernity, and the Weimar State. Princeton: Princeton University Press. Huber, Evelyne and John D. Stephens. 2000. “Partisan Governance, Women’s Employment, and the Social Democratic Service State.” American Sociological Review 65 (June): 323–42. Irving, R.E.M. 1973. Christian Democracy in France. London: George Allen & Unwin, Ltd. Kalyvas, Stathis. 1996. The Rise of Christian Democracy in Europe. Ithaca: Cornell University Press. Karvonen, Lauri. 1994. “Christian Parties in Scandinavia: Victory Over the Windmills?” Pp. 121–41 in Christian Democracy in Europe: A Comparative Perspective ed. David Hanley. London: Pinter Publishers. Kolbe, Wiebke. 1999. “Gender and Parenthood in West German Family Politics from the 1960s to the 1980s.” Pp. 133–67 in State Policy and Gender System in the Two German States and Sweden 1945–1989, ed. Rolf Torstendahl. Uppsala: University of Uppsala Press. Lenoir, Rémi. 2003. Généalogie de la morale familiale. Paris: Editions du Seuil. Leira, Arnlaug. 2006. “Parenthood Change and Policy Reform in Scandinavia, 1970s-2000.” Pp. 27–52 in Anne Lise Ellingsæter and Leira, eds., Politicising Parenthood in Scandinavia: Gender Relations in Welfare States. Bristol: The Policy Press.

87

Lijphart, Arend. 1984. The Politics of Accommodation: Pluralism and Democracy in the Netherlands. Berkeley: University of California Press. Lipset, Seymour M. and Stein Rokkan. 1967. “Cleavage Structure, Party Systems, and Voter Alignments: An Introduction.” Pp. 1–64 in Party Systems and Voter Alignments: CrossNational Perspectives eds. Lipset and Rokkan. NY: Free Press. Lorwin, Val R. 1966. “Belgium: Religion, Class and Language in National Politics.” Pp.147–87 in Political Oppositions in Western Democracies ed. Robert A. Dahl. New Haven: Yale University Press. Luther, Kurt Richard and Wolfgang C. Müller. 1992. “Consociationalism and the Austrian Political System.” Pp. XYZ in Politics in Austria: Still a Case of Consociationalism? eds. in Luther and Müller. London: Frank Cass and Co. Madeley, John T.S. 1977. “Scandinavian Christian Democracy: Throwback or Portent?” European Journal of Political Research 5: 267–86. Mallinson, Vernon. 1963. Power and Politics in Belgian Education 1815 to 1961. London: Heinemann. Mallinson, Vernon. 1967. “Education in Belgium.” Comparative Education Review 11, no. 3 (October 1967). Marraro, Howard R. 1936. New Education in Italy. NY: S.F. Vanni, Inc. Martin, Claude and Patrick Hassenteufel. 1997. La représentation des intérêts familiaux en Europe: Allemagne, Belgique, Grande-Bretagne, France, Portugal. Unpublished report for the European Commission, Directorate-General V. Marques-Pereira, Bérengère and Olivier Paye. 1998. “La Belgique: Vices et vertus du pragmatisme,” in Qui doit garder le jeune enfant? Modes d’accueil et travail des mères dans l’Europe en crise eds. Jane Jenson and Mariette Sineau. Paris: Librairie Générale de Droit et de Jurisprudence. Martin, David. 1969. “Notes for a general theory of secularization.” Archives européennes de sociologie 10: 192–201. Moeller, Robert G. Protecting Motherhood: Women and the Family in the Politics of Postwar West Germany. Berkeley: University of California. Morgan, Kimberly J. 2002. “Forging the Frontiers Between State, Church, and Family: Religious Cleavages and the Origins of Early Childhood Education in France, Sweden, and Germany,” Politics and Society 30, no. 1 (March): 113–48. Morgan, Kimberly J. 2003. “The Politics of Mothers’ Employment: France in Comparative Perspective.” World Politics 55, no. 2 (January): 259–89. Morgan, Kimberly J. 2006. Working Mothers and the Welfare State: Religion and the Politics of Work-Family Policies in Western Europe and the United States. Palo Alto: Stanford University Press. Naumann, Ingela K. 2005. “Child Care and Feminism in West Germany and Sweden in the 1960s and 1970s.” Journal of European Social Policy 15, no. 1: 47–63 Nicolaysen, Bente Blanche. 2001. “Voluntary Service Provision in a Strong Welfare State.” Arbeitspapiere – Mannheimer Zentrum für Europäische Sozialforschung, no. 35.

88

O’Connor, Julia, Ann Shola Orloff, and Sheila Shaver. 1999. States, Markets, Families. Gender, Liberalism and Social Policy in Australia, Canada, Great Britain, and the United States Cambridge: Cambridge University Press. O’Donoghue, Cathal and Holly Sutherland. 1998. “Accounting for the Family: The Treatment of Marriage and Children in European Income Tax Systems.” Unicef: Innocenti Occasional Papers Economic and Social Policy Series, no. 65. Orloff, Ann Shola. 1993. “Gender and the Social Rights of Citizenship: The Comparative Analysis of Gender Relations and Welfare States.” American Sociological Review 58 (June): 303–28. Orlow, Dietrich. 2000. Common Destiny: A Comparative History of the Dutch, French, and German Social Democratic Parties, 1945–1969. NY: Berghahn Books 2000. Pedersen, Susan. 1993. Family, Dependence, and the Origins of the Welfare State. Cambridge: Cambridge University Press. Pistillo, Filomena. “Preprimary Education and Care in Italy.” Pp. 151–202 in How Nations Serve Young Children: Profiles of Child Care and Education in 14 Countries eds. Patricia P. Olmsted and David P. Weikart. Ypsilanti, MI: High/Scope Press. Quataert, Jean. 2001. “Socialisms, Feminisms, and Agency: A Long View.” Journal of Modern History 73, no. 3 (September): 603–616. Quine, Maria Sophia. 2002. Italy’s Social Revolution: Charity and Welfare from Liberalism to Fascism. New York: Palgrave. Rokkan, Stein. 1966. “Norway: Numerical Democracy and Corporate Pluralism.” Pp. 70–115 in Political Oppositions in Western Democracies ed. Robert G. Dahl. New Haven: Yale Univ. Press. Sainsbury, Diane. 1996. Gender, Equality, and Welfare States. Cambridge: Cambridge University Press, 1996). Saraceno, Chiara. 1984. “Shifts in Public and Private Boundaries: Women as Mothers and Service Workers in Italian Daycare.” Feminist Studies 10, no. 1 (Spring): PAGES. Schmid, Josef. 1995. “Verbändewohlfahrt im modernen Wohlfahrtsstaat: Strukturbildende Effekte des Staat-Kirche-Konflicts.” Historical Social Research 20, no. 2: 88–118. Secher, Herbert P. 1959. “The Socialist Party of Austria: Principles, Organization and Policies.” Midwest Journal of Political Science 3, no. 3 (August): 277–99. Sipilä, Jorma and Johanna Korpinen. 1998. “Cash versus Child Care Services in Finland.” Social Policy and Administration 32, no. 3 (September): 263–77. Smith, Timothy B. 1998. “Republicans, Catholics, and Social Reform: Lyon, 1870–1920,” French History 12, no. 3 (September): 246–75. Spotts, Frederic. 1973. The Churches and Politics in Germany. Middletown, CT: Wesleyan University Press. Stephens, John D. 1979. “Religion and Politics in Three Northwest European Democracies.” Comparative Social Research 2: 129–57. Sundbärg, Gustav ed. 1904. Education in Sweden. Stockholm: Kungl Boktryckeriet. Sundbeck, Susan. 1994. “Nation and Gender Reflected in Scandinavian Religiousness.” Pp. 129–50 in Scandinavian Values: Religion and Morality in the Nordic Countries, eds. Thorleif Pettersson and Ole Riis. Uppsala: S. Academiae Ubsaliensis.

89

Sundbeck, Susan. 1995. “Tradition and Change in the Nordic Countries.” Pp. XYZ in The PostWar Generation and Establishment Religion: Cross-Cultural Perspectives eds. Wade Clark Roof, Jackson W. Carroll, and David A. Roozen. Boulder: Westview Press. Talmy, Robert. 1962. Histoire du mouvement familial en France 1896–1939 volume 1. Paris: Union Nationale des Caisses d’Allocations Familiales. Thelen, Kathleen. 2003. “How Institutions Evolve: Insights from Comparative Historical Analysis.” Pp. 208–40 in Comparative Historical Analysis in the Social Sciences, eds. James Mahoney and Dietrich Rueschemeyer. Cambridge: Cambridge University Press. Therborn, Göran. 1993. “The Politics of Childhood: The Rights of Children in Modern Times.” Pp. 241–93 in Families of Nations: Patterns of Public Policy in Western Democracies, ed. Frances Castles. Aldershot: Dartmouth Publishing Company Ltd. Tietze, Wolfgang, Hans-Gunther Rossbach, Karin Ufermann. 1989. “Child Care and Early Education in the Federal Republic of Germany.” Pp. 39–85 in How Nations Serve Young Children, eds. Patricia P. Olmsted and David P. Weikart. Ypsilanti, Mich.: High/Scope Press. Van Kersbergen, Kees. 1995. Social Capitalism: A Study of Christian Democracy and the Welfare State. London: Routledge. Van Kersbergen, Kees and Uwe Becker. 1988. “The Netherlands: A Passive Social Democratic Welfare State in a Christian Democratic Ruled Society.” Journal of Social Policy 17, no. 4. Van Vugt, Joos. 1991. “For Charity and Church: The Brother-Teachers of Maastricht.” History of Education 20, no. 3 (1991): 219–43. Weiss, John H. 1983. “Origins of the French Welfare State: Poor Relief in the Third Republic, 1871–1914,” French Historical Studies 13, no. 1 (Spring): 47–78. Wetterberg, Christina Carlsson, Kari Melby and Bente Rosenbeck. 2002. “The Nordic Model of Marriage and the Welfare State.” Paper presented at “Alva Myrdal’s Question to Our Time,” conference at Uppsala University, Sweden, March 6–8. Wolff, Richard J. “Catholicism, Fascism and Italian Education from the Riforma Gentile to the Carta Della Scula 1922–1939.” History of Education Quarterly 20, no. 1 (Spring): 3–26.

90

4

Italy: A Christian Democratic or Clientelist Welfare State?

Julia Lynch

Is the welfare state in Italy, a quintessentially Christian democratic polity, a “Christian democratic welfare state”? The Italian welfare state has most, if not all, of the hallmarks of what is often referred to in the comparative welfare state literature as a Christian democratic welfare state (hereafter CDWS): contributory social insurance programs linked to occupation that reproduce status differentials, a predominance of transfers over services, low female force participation and low employment rates among others. The key explanation in the comparative literature for the emergence of these features in a number of countries has been Christian democratic party strength, usually measured as ‘government participation of Christian democratic parties’ (Huber, Ragin and Stephens 1993; Huber and Stephens 2001). In Italy both the institutional features and the political strength of Democrazia Cristiana (DC, Christian Democratic party) are undeniably present, lending plausibility to the argument that Christian Democracy built the Italian welfare state. Yet this view of the Italian welfare state as organically related to the political strength of a Christian democratic party does not, I shall argue, concern itself adequately with the mechanisms by which Christian democracy as a political phenomenon becomes translated into welfare state structures. Indeed, it tells us little about how Christian democratic parties may have created homologous welfare states, as distinct from the welfare states constructed in polities dominated by social democratic or Liberal parties. A second strain of thought in the comparative literature has done more to uncover the causal mechanisms at work (Becker and Van Kersbergen 1988; Esping-Andersen 1990, 1998; Van Kersbergen 1995). This literature has identified two main pathways, which may operate simultaneously, through which Christian democracy as a social and political phenomenon influences policy outcomes: social Catholic ideology as carried by political parties and other social actors (Esping-Andersen 1990, Van Kersbergen 1995); and the “politics of mediation” (Becker and Van Kersbergen 1988, Van Kersbergen 1995), a particular type of political strategy characteristic of centrist parties based on cross-class, religiously-oriented social coalitions. This literature is of

91

immense value because it insists on specifying the mechanisms underlying the observed correlation between Christian democratic party strength and “Christian democratic” welfare state policies. I shall argue in this chapter, however, that neither of the mechanisms it has identified to date fully explain the development in Italy of policy features commonly regarded as characteristic of a Christian democratic welfare state. A reconstruction of the history of Italy’s welfare state offers least evidence for the impact of religious doctrines on the shape of social policies. The role of the Church as a pressure group promoting particular policies was tightly circumscribed, with the possible exception of education (see Morgan this volume). The role of mass-level demand for specific policy features that accord with social Catholic teachings is also limited. Policy drift (Hacker 2005), rather than demand emanating from a religiouslyinspired society, drives many of the key policy features that others have interpreted as evidence of a Christian democratic imprint. The evidence for the impact of partisan competitive strategies on the shape of Italian social policies is more robust, which would seem to confirm earlier findings (Becker and Van Kersbergen 1988, Van Kersbergen 1995). However, I find that more than the “politics of mediation” practiced by other Christian democratic parties in Continental Europe, it was above all the DC’s extreme reliance on clientelism that led to the development of highly fragmented, status-differentiated, and gendered policies in the Italian welfare state. This chapter’s main purpose, then, is to illustrate the mechanisms through which Christian democracy has affected the Italian welfare state. The importance of the presence of a religious cleavage in politics, as articulated by Van Kersbergen and Manow (this volume), is beyond question. But how does this cleavage matter? How does it become transformed into social policies? Policies implemented by anticlerical Liberal governments lay the groundwork of the Italian welfare state. These were sometimes directed against the interests of the Church, but sometimes conformed to them. Where they did correspond to the Church’s interests, however, they were not implemented for this reason alone. Liberal governments rather implemented these policies despite the fact that the Church also supported them.

92

After World War II the dominant force shaping the Italian welfare state structures that we observe today was indeed the DC, but not because of any Catholic social doctrine that leaders of this party may or may not have espoused. Indeed, as Zuckerman (1979) has observed, ‘[T]here is no evidence that DC leaders have ever sought to apply as policy the principles of Christian democracy’. Instead, the Christian Democratic party’s highly particularistic mode of political competition, which went well beyond what either Catholic social doctrine or the “politics of mediation” would require, is responsible for the policy immobility that, combined with policy drift, creates a highly stratifying and familialist welfare state. This chapter begins by describing the features of the Italian welfare state and of Italy’s political and social environment that have led scholars to interpret Italy as a welfare state defined by and for Christian democracy. The second part of the chapter lays out an alternative interpretation of the Italian welfare state that has little to do with an ideological imprint of Christian democracy. To preview, I argue that Italy’s status-reproducing occupationalist social insurance programs, which over time matured into a welfare state that rewarded older male breadwinners handsomely but crowded out spending on benefits for families and young adults, originated during the Liberal period, and did not reflect pressure from either the Church or a nascent Christian democratic party. The Liberals’ hands-off approach to social assistance, which eventually matured into a familialist welfare state that provided few direct benefits to women, children, and young adults, was conditioned in part by the Church’s monopoly of charity in the nineteenth century. But it was also inspired by Liberal ideologies that privileged fiscal restraint and an underdeveloped administrative apparatus. The closure of state policies to young and female labor-market ‘outsiders’ in the midtwentieth century and beyond was not an outcome enforced or even necessarily desired by post-WWII Christian democrats, even though it was influenced by the Church’s dominance in the charitable provision during the Liberal period. During the post-War period, the DC assumed crucial responsibility for maintaining and intensifying Liberal-era policy patterns. Over time, and in concert with demographic and labor market shifts, these Liberal-era policy patterns gave the Italian welfare state its distinctive occupationalist and familialist cast. But it was the DC’s clientelism, and not its Catholicism, that explains this outcome.

93

4.1

The Christian Democratic Isomorphism of the Italian Welfare State

Italy’s geopolitical location at the heart of the Church’s empire and the DC’s electoral dominance from 1948 through 1992 make it highly plausible that Christian democracy would be an important influence on the development of the welfare state in that country. For this reason, Italy is often described in the comparative welfare state literature as a Christian democratic welfare state, following on the seminal contributions of Esping-Andersen (1990 and 1999) and Van Kersbergen (1995) (see Castles and Mitchell 1993; Huber, Ragin and Stephens 1993). At the level of policy, a CDWS is characterized by occupational social insurance programs that reproduce status differentials; few publicly provided services, particularly for families; a male-breadwinner bias in both tax and transfer systems; and a tendency to devolve authority over delivery and implementation of social policy to nonstate actors. Italy’s welfare state meets many of these criteria. It is occupationalist: social insurance programs like pensions, unemployment benefits, and family allowances do not provide uniform benefits for all citizens, or even all workers. Instead, numerous different programs addressing the same basic risk (e.g. old age, unemployment) draw from distinct, although rarely hypothecated, funds. The result is a ‘jungle’ of highly differentiated provisions (Gorrieri 1972) that preserve the privileged positions of some sectors and grades of workers, and exclude others entirely. The system works to preserve status differentials and validate class and occupation hierarchies. The Italian welfare state is also transfer-heavy, spending only 23 per cent of its total social budget, including health care, on services (as compared to 42 per cent in Sweden and 36 per cent in the UK).1 As a result, families are responsible for providing or purchasing much care for children, the disabled, and the elderly. Even in the realm of cash transfers, familial obligations are only weakly subsidized. Legal entitlements to support for dependents are generous in concept (under the old family allowance scheme in effect through the mid-1990s, household heads could receive family allowances for not only children, but also spouses and parents), but not in quantity:

1 1999 data from Abramovici (2002). Comparative figures for non-health social spending are even more revealing. Only 1.2 per cent of spending on family and children’s benefits in Italy is for services, compared to 6 percent in Sweden.

94

throughout the 1980s and 1990s, spending on cash benefits and services for families with children per child under age 14 was lower in Italy than anywhere else in Europe outside Spain and Portugal.2 At the same time, Italian civil law recognizes an extended network of obligation to provide cash support for family members. Meanstests may include the income not only of spouses and parents of minor children, but also of non-coresident parents, adult children, full and half siblings, and sons-, daughters-, mothers- and fathers- in law (Millar and Warman 1996: 35; see also Naldini 2003). ‘Weak state control’ (Van Kersbergen 1995) over the financing and implementation of social policies has also been a hallmark of the Italian welfare state. Most social policies in Italy, even means-tested social assistance benefits, are financed via payroll deductions, rather than through general revenues or specially earmarked taxes. Cross-subsidization of social insurance programs for different sectors is carried out via a porous financial administration of the funds and weak hypothecation, not through general revenue financing of funds in deficit. In shifting the burden of financing to the social partners, the Italian state also effectively relieved itself of the duty to collect payments. A 1983 report of an independent watchdog group noted a very small number of employees of the labor inspectorate whose duties include enforcement (CENSIS 1983: 237). Perhaps not surprisingly, employees have complained frequently over the years that evasion of social payments by employers is widespread. Indeed, a 1980 study revealed evasion of social security payments for 42 per cent of employees in the over ten thousand firms studied (Regonini 1984, 106). The administration of social benefits in Italy is likewise subject to weak state control. Most old-age and disability pensions, family allowances, and some types of unemployment benefits are administered by the national social security administration (INPS), which is governed on a tripartite basis with the social partners enjoying a strong majority. Patronati, advocacy groups that help individual citizens with the processing of social insurance claims, were until recently financed by the Italian government, but run by and for private organizations like political parties, unions and religious groups. Much institutional care for children, the disabled and the elderly is operated by non-state actors. From the 1960s through the early 1990s, the remaining

2

Data from OECD (2003). Average of yearly amounts from 1985 through 1998.

95

social assistance functions of the welfare state, including a de facto employment service consisting of public jobs, were carried out by a public administration that was thoroughly colonized by political parties that should themselves be considered subsidiary organizations. The occupational fragmentation of Italian social insurance system, its passive familialist orientation resulting from very limited state provision of social services, and spending patterns that reward senior males while granting few independent entitlements to women, young adults, and children, all seem to conform to social Catholic doctrines of organic hierarchy and Subsidiarity. The Italian pension system comprises more than 120 separate public pension funds for different sectors and categories of workers at its peak (Castellino 1976) and is therefore an unusually fragmented occupationalist system. It has been interpreted as a reflection of the social Catholic emphasis on status preservation. The pension-heavy mix of welfare state spending in general is also notable. Italy spends a higher proportion of its social policy resources on the elderly (even adjusting for the demographic makeup of the population) than almost any other OECD country (Lynch 2001), which leaves little room for spending on other priorities like children or the unemployed. Relatively generous public pensions coexist with meager child allowances and unemployment benefits, and almost non-existent family services and active labor market programs. This policy mix, in which appears to fit nicely with the social Catholic doctrine of subsidiarity, allocates large cash transfers to older (usually male) family heads for further redistribution. In return, it requires familial (often female) provision of both financial and non-financial assistance to needy members. I have argued elsewhere (Lynch 2006) that Catholic social doctrine could be interpreted to support other policy options as well, including generous family allowances (see in particular the lit on the French caisses allocation familiales). So a straightforward reading of the ‘natural’ policy outcomes of social Catholic doctrine is not possible. Futhermore, policy features central to the CDWS (occupational fragmentation, familialism, and a male-breadwinner orientation chief among them) are present in most continental European states, even where they trace their roots to ideological and political movements distinct from Christian democracy (see Esping-Andersen 1996 on varieties of Conservatism). So in order to understand the sources of the Italian welfare state’s particular mix of policies, it is necessary to look beyond their face-level

96

compatibility with social Catholic ideologies, and attend closely to the mechanisms through which Christian democracy as a political movement has influenced the formation of these central policy features. The remainder of this chapter argues that central features of the Italian welfare state are the result of the interaction over time between old policies put in place under an anti-Catholic Liberal regime and the clientelist style of political competition preferred by the DC and its allies during the post-War period. The DC (as well as its secular allies) took accommodation and pragmatism to levels well beyond those required by Catholic social doctrine, converting them into the currency of a manifestly clientelist political operation. In the process, they reinforced social policy institutions that originated during a period, perhaps Italy’s only such period, of decidedly anti-Church policy-making. Policy drift, in this case the failure to update policies in line with new developments in labor markets, demographic trends, and family structures, resulted, by the 1970s, in a welfare state characterized by stratification, passive familialism, and a male-breadwinner orientation.

4.1.1 The Secular Roots of Stratification and Occupational Fragmentation The Italian social security system’s basic occupational organizational structure dates to the late nineteenth and early twentieth century. The unification of Italy in 1861 resulted in the seizure of large territories from the Church, and the Italian nation was in many ways founded against the Church. Catholics themselves were excluded by papal order from participation in government and politics. Of course, many Catholics did participate in politics, including those who stood for office under the banner of the Popular party. Still, the case for strong Christian democratic influence in constructing the early policies of the Italian welfare state in Italy appears weak on the face of it. But institutions do change, so we also need to explain the continuity of occupationalism once it is in place. It is possible that the Italian welfare state’s original occupationalist structure was only preserved because of strong support from Catholic political actors and/or the Church. As we shall see, however, successive opportunities for reform in the direction of universalism were blocked not solely by the political power of social Catholic ideologies, but also, and more actively, by other forces.

97

Occupational differentiation in the Italian pension system dates back to the earliest public pension legislation.3 The first public pension provisions in Italy, which took Bismarck’s programs in Germany as a direct model, were for public employees only. In 1898 the first state-supported pension scheme for employees in the private sector was legislated, but inscription was voluntary and the system was designed primarily as a subsidy to encourage provision by private voluntary organizations. The emphasis in this early pension legislation on private voluntary solutions surely reflected to a certain degree social Catholic doctrine as expressed in Rerum Novarum (1891), which preferred joint worker-owner schemes to obligatory state-run social insurance. The Church and politicians of the nascent Catholic political movement could not have been displeased with the outcome, for the limited social insurance law ensured continued Church dominance of social assistance, especially in rural areas, and permitted the Catholic democratic movement to mobilize agricultural workers in the South using the selective benefits of mutualist insurance programs (Ferrera 1993: 214–5). Nevertheless, Church and Christian democratic influences do not seem to have been the most important contributors to the 1898 law’s emphasis on occupational social provisions. Politicians of the day were strongly influenced by the desire to ‘legitimate the Liberal state in the eyes of the working classes’ (Gustapane 1989: 57), and saw socialismo della cattedra along Bismarckian lines as the means to this end. Ferrera (1993: 210–211) reports that the legislative debates leading up to the 1898 provision were dominated by the language of Kathedersozialismus, which had been adopted by many Liberal policy-makers by the 1880s. And while the Church had an interest in promoting mutualism at the expense of comprehensive state-run insurance programs, so too did non-religious actors. The political landscape in the late nineteenth century was still characterized by an uneasy alliance of the forces that had joined to promote the recent unification of the Italian nation: northern political elites and small and medium landowners in the South. These landowners depended on cheap labor from some nine million agricultural laborers, or three quarters of Italy’s dependent workforce (Sepe 1999: 97). Northern political elites were concerned to maintain the support of Southern landowners by preventing a universal state social insurance

3 This analysis of the development of Liberal-era social insurance is reconstructed largely from Ferrera (1993, Chapter 6), Cherubini (1977, Chapters 2–5), and Sepe (1999, Chapters 2–4). Other useful sources of analysis of this period include Bartocci (1999), and INPS (1962).

98

model that would have mandated insurance contributions for the employers of agricultural laborers. Italy’s early pension legislation, then, reflected the étatist impulse of the Liberal period, and served the political ends of both the Liberal politicians who controlled the state policy apparatus and Catholic actors who, however, did not themselves have any direct way of influencing policy outcomes.4 Under the circumstances, the occupationalist imprint of the 1898 legislation (and extensions thereto from 1900–1920) comes as no surprise. What was surprising, at least at the time, was the nearly complete failure of this model to respond to the ‘social question’ with comprehensive insurance against social risks. By 1910 fewer than 350,000 workers – only 2.5 per cent of the dependent workforce – were enrolled in voluntary insurance schemes subsidized by the state (Sepe 1999: 97). Opportunities to revisit the design of social insurance programs arose out of the failure of the 1898 legislation. The persistence of the occupationalist model after 1898 thus demands explanation. But if Christian democracy did not cause the occupationalism of the first wave of legislation – unremarkable given the (self-) exclusion of Catholics from political life under the papal non expedit – neither did it work to preserve the occupationalist model in subsequent waves of reform. Between 1906 and 1912 separate obligatory pension insurance schemes were legislated for particular categories of employees, mainly in the transport, maritime and public or parastatal sectors. Under mounting pressure from the socialist movement and in anticipation of a major suffrage expansion in 1912, Giolitti proposed in 1910 new comprehensive public social insurance that would cover all dependent workers, including those in agriculture. Catholic leaders were not enthusiastic about the proposal, but did express a willingness to support Giolitti (Ferrera 1993: 217). The plan came to nothing as the war in Libya quickly removed social issues from the agenda,

4 Ferrera (1993) takes a more structural view, arguing that the occupational fragmentation of the Italian social insurance system results from longstanding features of the Italian political economy: a highly segmented economy, and a culturally fragmented and ideologically polarized polity. Both make risk-pooling and the development of solidarity difficult (see also Baldwin 1990). Baldwin’s model of welfare state development, and Ferrera’s extension of it, challenge quite directly the notion that occupational fragmentation is rooted in the ideological traditions of social Catholicism.

99

but this attempt to reorganize the Italian social insurance system along less occupationally fragmented lines did not fail because of opposition from Christian democracy. In the aftermath of the First World War, the social question re-emerged with particular intensity, and by 1919 obligatory pensions for blue-collar employees came into law. Liberal Prime Minister Nitti had, with the support of both socialists and some Catholics in the legislature, put forward a reform proposal that would have unified the various funds into a single public scheme and extended obligatory coverage to agricultural workers. However, the 1919 law stopped well short of these goals. Pensions applied only to blue-collar workers with incomes less than 800 Lire per month, and the organization of the social insurance programs remained fragmented along occupational lines. The 1919 law introduced relatively modest changes, in part because employers opposed the more maximalist reforms envisioned by Nitti. Agricultural employers in particular argued that they could not afford obligatory pension insurance and that therefore the only possible direction of reforms was an extension of voluntary insurance (Sepe 1999, 100). The failure of Nitti’s reform agenda has been attributed mostly to opposition from employers (Cherubini 1977: 237). But other political forces also failed to support a radical reorganization of the system along universalistic lines. Divisions in the socialist movement meant that working-class organs no longer spoke with one voice in favor of social reformism, and, despite increasing discussion of a universalistic option in the years leading up to 1919 (Ferrera 1993, 225), many Liberals continued to favor simply extending the old model of state-subsidized occupational insurance (Sepe 1999, 158). A Christian democratic voice was then not a decisive contributor to the already overdetermined outcome of this debate. It is not even clear what a Christian democratic voice at this time would have said. While the Sturzo-De Gasperi wing of the Popular party had supported Nitti’s plan to introduce universal social insurance in 1919 (Ferrera 1993: 229), the Catholic labor movement definitively opposed state-run insurance. In any event, with or without the influence of Christian democracy, the social insurance system remained occupational in nature as the Liberal regime crumbled.

100

Giolitti’s call in 1921 to ‘complete’ and ‘simplify’ the system5 so as to reduce the degree of occupational fragmentation went unheeded as Italy began its slide into Fascism. The onset of the extreme occupational fragmentation that characterizes the Italian social insurance system today dates to the Fascist period. In 1923 the pension system underwent what was to be a major reorganization, joining the separate funds for different industries and different branches of the public service into a single entity, and allowing separate sectoral funds only to provide benefits supplemental to the basic insurance. In practice, however, these reforms did not occur, and the occupational fragmentation of the system actually increased. Cherubini (1977, 270) attributes the growth of special privileges and deficiencies for different sectors to the Fascist politics of playing groups off of one another, offering concessions where necessary to garner support for the regime (e.g. to journalists), and sacrificing those who could not fight back6. Needless to say, insurance for agricultural laborers was once again put on the back burner. Let us take stock. So far we have seen that the original occupationalist design of the Italian social insurance system was inspired primarily by étatist principles, rather than social Catholic ones, and was implemented under a regime in which Catholics and the Church were explicitly excluded from policy-making. Successive attempts to broaden, harmonize, and universalize pension coverage failed, but never because of opposition from Christian democratic forces. As we enter the post-World War II period, the potential for Christian democratic influence became greater, with the decisive rapprochement between the Vatican and the Italian state, and with the DC’s hegemonic political position. And indeed the occupational fragmentation of the Italian pension system continued to increase well through the 1980s. As we shall see, however, occupational fragmentation persisted not because of social Catholic respect for hierarchy and status differentials, but because of the particular style of political competition that the DC engaged in for much of the post-War period.

5

See Giolitti’s Discorsi parlamentari, Vol IV, p. 1894, cited in Sepe (1989: 159).

6 Social policy in Communist East Germany underwent a similar transformation, where loyal groups such as Stasi members, police, party rank and file, and journalists were rewarded with special benefits, in strict opposition, of course, to the Communists’ official policy of eliminating status differentials preserved in the West German Bismarckian model (Manow 1994).

101

4.1.2 The Laissez-faire Roots of Passive Social Assistance Measures Before turning to the development of the Italian WS after WWII, though, we must consider for a moment the pre-War roots of social assistance, as well as those of the social insurance programs we have already examined. The terrain of social assistance, taken together with the explosive growth of occupationalist social insurance programs, has important implications for the eventual familialist orientation of the welfare state in the contemporary period. The Italian welfare state’s emphasis on pensions at the expense of other forms of social spending marks it as a passive, familialist, male-breadwinner centered welfare regime. Locating the origins of this policy mix, either in social Catholic doctrines of subsidiarity or in some other set of causes, is then crucial for understanding the mechanisms through which Christian democracy has influenced the Italian welfare state. The Italian welfare state clearly shows signs of Church influence in the prevalence of non-state, often Church-run, social service provision, and in the under-development of social assistance more generally. The Church in Italy had a near monopoly on charity and institutional care in the late nineteenth century, and staunchly defended its material and spiritual interest in the social assistance sector. Yet Liberal-era politicians overcame the Church’s objections and, in several key areas, voted for legislation that was important to their goals of ‘modernizing’ society and achieving a separation of civil and ecclesiastical authority. The 1887 Cioppino law, for example, which established free compulsory public education, prompted accusations from the Church that the Italian state was engaging in a Kulturkampf, yet only twenty deputies voted against it (Quine 2002: 37). 1890 legislation radically increasing the state’s role in social assistance and charity also passed easily, with 196 votes for and only 98 against, despite Pope Leo XIII’s accusation that politicians supporting the bill harbored an ‘implacable hatred of Jesus Christ’ (cited in Quine 2002: 53). The underdevelopment of state social assistance functions during the Liberal period then cannot be understood purely as a result of the Church’s success in preventing the state from playing on its field. Church opposition cannot account for Liberal politicians’ failure to seize the field of social assistance, a task that they did accomplish in education despite even more strenuous objection from the Church (see also Morgan, this volume).

102

Quine (2002) and Fargion (1997) both argue that the weakness of social assistance and social services in Italy well into the post-war period originate in the Liberal state’s reluctance to take responsibility for the costs – both financial and in terms of developing administrative capacity – of providing adequate social provision directly. The Italian state did attempt to assert control over private charitable institutions, attempting to transform Christian charity into a theoretically more rational, discretionary, and efficient ‘legal charity’ (caritá legale). However, this attempt was half-hearted. Legislation in the pre-World War I period made municipal- and provincial-level administrators responsible for regulating and transforming religious charities into public entities, but these administrators were endowed with neither the financial resources nor the administrative capacities necessary to effectively intrude upon traditional practices. Piedmontese statutes, which were eventually replicated in Republican Italy, did not merely seek to avoid antagonizing the Church: ‘[F]inancial considerations also provided a very compelling reason to resist calls to extend the means of state intervention in the activities of charities. Cavour had long recognized that creating a system of legal charity would cost a lot of money’ (Quine 2002: 34). Decentralization and delegation of administrative control over social assistance responded in part to demands from the parliamentary Left, but it was also ‘a means for the nation’s leaders to offload the financial and bureaucratic responsibilities for monitoring opera pie onto officials in the localities’ (Quine 2002: 44; see also Fargion 1997: 75). Even once the Italian state’s fiscal capacity improved in the last decade of the nineteenth century and the first of the twentieth, policy-makers remained unwilling to assume for the central state the burden of providing social assistance to the masses. Quine attributes this reluctance primarily to a combination of ideological factors (Liberals lacked a sense of state, and ‘never truly grasped the importance of “the social” to the polity“ [2002: 295]) and weak state administrative capacities, the latter of which allowed Church dominance of the charitable sector to continue even when legislation formally wrote it out of the picture. The clientelist political dynamics of legislative trasformismo also played a role in the continuing underdevelopment of national-level policies. Locally constituted public boards consolidated the finances of smaller charities and governed the operation of even the larger institutions. These boards were an important electoral resource for a system that prized the manufacture of electoral majorities at the prefectural level (Quine 2002: 58), especially given that

103

the reported revenues of private charitable institutions in 1880 were nearly equal to the total tax take of the Italian state (Quine 2002: 50). In the Liberal period, then, the Church’s role in perpetuating its own dominance of the social assistance sector was indirect, operating mainly as a default option pursued only because of Liberal politicians’ own preference for limited central government involvement. To sum up, in the late-19th to early-20th century, both the social assistance and social insurance functions of the nascent Italian welfare state took their shape from the interests of Liberal (and to a lesser extent Socialist) political actors, not from the Church. Paradoxically, given Liberal-era anticlericalism, the interests of Liberals, Socialists and the Church were often congruent. But as we have seen, occupationalist social insurance programs and weak state penetration of the social assistance sector occurred not because the Church demanded them, but because key players in the anticlerical regime desired them.

4.2

Clientelism and the Maintenance of Occupationalism after World War II

An opportunity for radical revision of pre-existing social insurance and social assistance programs occurred in Italy after World War II. Wartime conditions had aggravated social problems, and insurance programs had been bankrupted because of runaway inflation and wartime destruction of property. Italy’s reform-minded Communist Party (PCI), the second-largest party after the DC, strongly advocated replacing the old system of occupational social insurance with a universal citizenship-based welfare state. The PCI was not alone. Under the influence of the International Labour Organization7 and policy lessons diffusing from Britain, Italy, like most European countries with occupational welfare states (Ferrera 1993), set up nonpartisan official commissions to investigate the feasibility of introducing citizenship-based welfare programs along the lines of Britain’s Beveridge Plan.

7 Such was the enthusiasm for the Beveridgean model at this time that ILO recommendations included citizenship-based prescriptions, despite the fact that the ILO was dominated by countries with occupationalist social insurance systems.

104

Two nonpartisan Commissions, in 1947 and 1963, recommended that the pension system be overhauled, occupational distinctions minimized, and coverage eventually extended to the self-employed (Commissione d’Aragona 1948, CNEL 1963). But beginning in the 1950s the pension system was gradually extended, under DC tutelage, to new categories of beneficiaries (small farmers, fishermen, artisans, shopkeepers, housewives, etc.), each with its own separate scheme. In the realm of social assistance, politics in the post-War period continued to limit the development of public social services well into the Republican period (Fargion 1997). The post-war Constitution for the first time granted citizens a right to social assistance, redefining charity as a duty of the state. But administrative competence was vested in the regions and municipalities, and the DC’s fears that more autonomous regional governments would lead to electoral gains for the PCI meant that provisions giving regions real powers remained dead-letter entities until the 1970s. At the same time, Fascist-era parastatal entities involved in charitable care, entities that commanded significant resources that could be used as currency in the DC’s clientelist political machine, were given some important functions (Fargion 1997, 88). DC policies and practices limited the competencies of state institutions, at the local and regional level as well as the national level, and the state’s ‘inertia’ in turn allowed the Church to maintain its traditional role in the social service sector (Fargion 1997: 89). In both the Liberal and early post-WWII periods, the Church’s organizational monopoly in the social assistance and social services sectors reinforced the occupationalist welfare state model pursued by secular actors during the Liberal era. This segregation of the Italian welfare state into ‘assistance’ and ‘insurance’ functions today remains a politically contentious issue, in part because the state’s role in the former was never clearly established. Social services and poverty alleviation together today account for only 2 per cent of welfare state spending in Italy (Abramovici 2003). In the realm of social insurance, the basic configuration of early post-War welfare state politics pitted a universalizing Left against an emergent DC with little interest in changing the status quo. The Left in Italy (the PCI, the Socialist PSI, and a majority of the union movement) favored universalizing the pension system for ideological reasons related to the development of universal citizenship in the new Italian republic. The DC, however, if it favored maintaining the occupational status quo, did not

105

do so for parallel reasons of social Catholic ideology. Until 1975, Left-leaning factions motivated by social Catholic ideologies (e.g. Forze Nuove, Base) were excluded altogether from the ruling organs of the party and from policy-making (Zuckerman 1979, 159–163). The factions that dominated the party were motivated not by ideology, but rather by the twin goals of building a party organization independent of the Church hierarchy and maintaining their grip on power within that party.8 Clientelist distribution of public goods and services (state jobs, tax relief, preferential pension treatment) to blocs of reliable supporters assured these goals. In the early post-war period the self-employed and rural Southerners were cherished above all (LaPalombara 1964). Preference voting, which allowed electors to cast a ballot for a particular candidate from a party’s list and hence allowed the party to determine the power and privilege of discrete factions (correnti) within the national party apparatus, was particularly widespread in the South.9 The intraparty competition fostered by preference voting made legislators perhaps more responsive than they would otherwise have been to local economic interests, which encouraged a proliferation of leggine guaranteeing preferential treatment in the social insurance system to microscopically defined economic ‘categories’.10 Once the DC had for several decades been in a position to hand out public jobs as patronage, public sector and parastatal employees became even more crucial sources of support. The priority for pension system policy was thus to generate maximum ‘consensus’ by securing maximum benefits for rather narrowly defined occupational and sectoral groups. Proposals to reform the Italian pension system by creating citizenship-based entitlements to pensions would have threatened the DC’s ‘consensus’ in several ways. Particularism, taxation, and the failure of universalism. The tax system in Italy suffers from a deficit of legitimacy and high levels of evasion. This tax system was itself rooted in clientelist politics and reinforced the occupational nature of the pension system. The failure to effectively tax key clienteles of the DC not only served the DC’s

8 Early in the post-war period, the goal of institutional independence was probably paramount. By the late 1960s, when this had been assured, hanging on to power and privilege was likely the more powerful motivator. 9 On the importance of preference voting and intraparty competition, see e.g. Cazzola (1972), Pasquino (1972), Parisi and Pasquino (1977), Zuckerman (1979). 10 For an indication of the extent of micro-legislation in the pension system, see Maestri (1994).

106

interests, but also slowly eroded the Left’s support for universalism between 1945 and 1970.11 The main obstacle in Italy to implementing the universalistic pension proposals of the early post-War period was the tax system. The reports of the 1947 and 1963 reform commissions both cite the impossibility of adequately assessing and collecting pension contributions, especially among the self-employed, as reasons to continue to provide pension benefits on an occupational basis (Commissione d’Aragona 1948, CNEL 1963a and 1963b). Labor leaders, although they professed support for the idea of universal coverage, feared that any universalization of the system would be paid for out of increased payroll taxes on employees, and so repeatedly called for fiscal reform as a prerequisite to pension reform.12 The DC’s unfunded extension of the pension system to new clienteles in the 1950s and 1960s worked to erode labor movement support for universalization. By the early 1960s, Italy’s pension system was categorized by two types of funds: those running a surplus, primarily the FPLD (Fondo pensioni lavoratori dipendenti, the main industrial employees’ fund) and the special funds for small groups like journalists; and those running large deficits, primarily the funds for agricultural workers, artisans and shop-keepers. The latter funds had been set up in the late 1950s and early 1960s by DC governments in order to benefit clienteles that were particularly important to the DC (Regonini 1996, 90). These funds ran large deficits because, in service to the clientelistic goals of the reforms, contribution rates were low and benefits made available to people with very limited contributory histories. While the state subsidized these pensions out of general revenues to some degree, in large part it was the employees’ funds that were asked to make up for shortfalls in a form of ‘enforced solidarity’ (Ferrera 1993: 262). By the late 1960s, both Confindustria (the main employers’ confederation) and the unions were complaining about the increased payroll taxes that had become necessary to support this burden, and union support for universalizing the pension system waned.

11 For a more in-depth treatment of this subject, see Lynch (2006, Chapter 6). 12 See, for example, CISL Consiglio Generale (1950: 13); CISL Consiglio Generale (1956: 150); CISL Consiglio Generale (1958: 178); CGIL Segreteria Generale (1962: 226); CGIL (1970); CGILCISL-UIL Segreteria Interconfederale (1970: 201).

107

The one piece of universalism in the Italian pension system was the Social Pension, which originated in tripartite accords in 1964 and provided a pension of last resort to low-income elderly Italians. But the way that the Social Pension was implemented between 1965 and 1970 ultimately undermined the Left’s support for a universal pension system. The 1964 agreement established that Social Pension benefits should be financed out of state revenues deposited in a new Social Fund. It was also agreed that the level of the benefit would be sufficient to provide for a ‘decent’ standard of living, as stipulated in Article 38 of the 1947 Constitution. The first steps towards a truly universal pension benefit had been taken. By 1968, however, both Confindustria and the labor unions had lodged complaints that state contributions to the Social Fund were inadequate to cover its costs, and the fund was instead drawing resources from the FPLD. It had become clear by this point that Social Pensions would have to be paid out of employees’ pockets if they were to be paid at all, and union support for further universalization of the pension system (as well as for upgrading the level of the Social Pension benefit to an adequate social minimum) faded (Ferrera 1993: 262). Failure to implement existing tax laws further eroded the Left’s support for universalism. The reluctance of successive administrations to execute the tax laws continued even after the definitive defeat of proposals to establish a universal pension system. Evasion of pension contributions for employees continued to be an important issue from the late 1970s onward. In 1978 the CGIL and the PCI began to complain quite insistently about the failure of employers to make contributions on behalf of their employees. This practice contributed to the deficits that had begun to plague even the stronger funds administered by INPS. A 1978 union proposal to make evasion of payments a criminal offense was defeated by the DC government, in part because the proposal would have transferred the separate system of occupational injury compensation to INPS, which was controlled by the unions, and thus make unavailable a key source of patronage for the government (Regini and Regonini 1981, 233). Both the tax laws (the failure to adequately tax key clienteles of the DC) and their implementation (the failure to enforce those laws that were in place) sprang from the particularistic mode of political competition pursued by politicians of Italy’s leading political parties during the post-war period. The inability to collect revenues from key economic sectors in turn made it impossible to universalize the pension system without placing an excessive burden on employees. This altered the Italian Left’s

108

preferences with regard to universalization, and made reform of the pension system along more universalistic lines nearly impossible for a period of almost 30 years. Occupationalism persisted because the DC’s behavior, not its ideologies, affected the policy preferences of both Christian democratic and Left political actors. The pension system as a by-product of clientelism. If particularistic political practices prevented the implementation of a tax system capable of sustaining a political coalition for universal pensions, clientelism also had more direct effects on the pension system. Clientelism encouraged DC politicians to maintain and even extend the occupational fragmentation of the pension system, because fragmented occupational provisions provided clientelist politicians with important resources. To most observers, Italy’s proliferation of pension provisions, each with its own benefit formula, contribution rate, degree of state subsidy, rules governing retirement age, years of service required to enter into the plan, etc., constitute proof positive that the pension system has been used as a way to attract support from particular groups in the population (see for example Regini and Regonini 1981; Paci 1984; Maestri 1994; Ferrera 1984). So too does the practice of staffing pension agencies with party supporters, and of deciding pension claims based on an applicant’s party affiliation (Regonini 1996: 90–91). This perception is not limited to the scholarly community, of course. Even the under-secretary of Labor, in the midst of the 1982 pension debate, dispatched a telegram accusing two representatives of the majority coalition of ‘rampant particularism for pernicious electoral reasons’ (Regonini 1984: 108). Christian Democratic politicians advocated extending pension benefits to the selfemployed on very generous terms during the 1950s and 1960s as part of a strategy to purchase loyalty from these groups. During the post-war period, state employees, however, were the most important targets of clientelist pension legislation, for obvious reasons. As the staff of public and quasi-public organizations came to be dominated by supporters of the governing parties, public sector pension benefits took on special cachet with politicians associated with these parties. Special provisions for public sector employees reach back into the Fascist period, as we have seen (see also Cherubini 1977 and Paci 1984). But public sector employees were the recipients of the largest volume of pension legislation during the period 1948–1983, and attention to this sector came predominantly from DC and PSI lawmakers (Maestri 1994: chapter 4).

109

Multiple, differentiated benefit categories are not just ex post evidence that politicians used the welfare state to target benefits to small groups of voters. The existence of such fragmentation also made it easier to justify new forms of discretionary targeting of benefits like a better replacement rate here, or a shorter reference income period there. Occupational pensions that provide different levels and types of benefits for different groups of workers are, clearly, a gold mine for politicians who use particularistic strategies to compete for votes and win elections. Provisions tailor-made for small segments of the electorate are visible and valuable to the beneficiaries, and hence to their benefactors. But as the fragmentation of the Italian welfare state increased, the very complexity and opacity of what Italian commentators have come to call ‘micro-corporatism’ comes to protect the politicians who engage in it. A thicket of highly specialized provisions makes it difficult for the public (and sometimes even for policy-makers) to know when changes have occurred, and even harder for them to understand what the consequences of such changes might be for the public interest. For all of these reasons, politicians who compete using patronage have been loath to see occupational fragmentation overturned, or even reformed. The clientelist behavior of the DC (and later PSI) also drew parties of the opposition into a mode of pension policy-making that has stimulated the accretion of occupationally derived social rights (as well as exacerbated the Italian pension system’s problem with expenditures). As we have noted, in principle the Italian Left supported a universalistic pension scheme. But in a context of particularistic behavior by the ruling DC, unions had a strong incentive to maintain the pre-existing occupational funds, over which they had some modicum of control, rather than giving everything over to a central, DC-run universal scheme (Ferrera 1993). The interplay of clientelism and fragmentation affected the Left’s preferences and behavior, with the opposition parties and unions coming to support particularistic fragmentation under the threat of an even more damaging particularistic universalism. In the 1950s and 1960s, the Left opposed the DC’s practice of extending pension rights to new groups without making these new groups responsible for financing them. However, the Left was too weak to block this practice (Regini and Regonini 1981: 122). But even after 1969, when they had more control over pension legislation, unions opposed moves to universalize the pension system in part because they

110

feared that a state-run system would bring their constituencies into the clientelist orbit of the DC, as had happened earlier with agricultural employees (Regini and Regonini 1981: 127). Instead, they followed a strategy of attempting to upgrade benefits for their constituencies to the level enjoyed by public sector employees (Baccaro 1999). Ferrera notes that the major pension reform in 1969, which granted many benefits to industrial workers that matched those already enjoyed by public sector employees, marked ‘the enlargement of the spoils system to include the PCI and unions, opening the way to that “assistential grand coalition” responsible for the profound imbalances that characterized the Italian welfare state in the years to come’ (1993: 267). The Left and the unions tried to match every gain made by public sector workers (and other privileged clients of the ruling parties), and to the extent that they succeeded, in turn the privileged clients demanded more privileges. The occupational fragmentation of the system generated pressures for ‘leapfrogging’, leading to an upward spiral of benefit levels and a downward spiral of contributory requirements in the 1970s and 1980s. The result was extraordinarily rapid growth of pension expenditures, and a subsequent crowding out of other social spending (Franco 1993). These developments are attributable in the first instance to the particularistic mode of competition engaged in by the DC and the PSI, whose politicians derived personal benefit from a fragmented universal system of pension provision even when it was clear that this system was bankrupting the public coffers and making it difficult to finance other needed social goods. But the strategy of the opposition parties and of labor unions has also been conditioned by this mode of competition among the dominant parties, and thus the Left ultimately shares responsibility for the continued expansion of the pension sector along occupational lines. Social Catholicism as an ideology is perhaps the force least plausibly responsible for the fragmented occupational system that has generated such high pension expenditures beginning in the 1970s. Italy’s fragmented occupational pension system then provides crucial resources for clientelist politicians, whose opposition to reforming such a system in turn provides occupationalism with a means of self-propagation. The combination of fragmented occupationalism and political particularism also altered the incentives and behavior even of political actors who would normally prefer a citizenship-based, rather than an occupational, pension system. Eventually they too came to support the fragmented

111

pension system status quo. Thus in the post-World War II period, just as during the Liberal and Fascist periods, Social Catholic ideologies were neither the cause of the Italian pension system’s occupational fragmentation nor the agent of its preservation. Pension system occupationalism and the DC’s particularistic politics form a resilient and self-reinforcing dyad. These political-institutional characteristics of the Italian welfare state are, as a result, able to shape the flow of benefits to different constituencies over long periods of time. As we shall see in the next section, the ‘stickiness’ afforded by this combination of clientelist politics and fragmented occupational pension programs in turn promotes a social policy mix that directs the lion’s share of resources to ageing male family heads. This policy orientation has often been interpreted as a result of Christian democratic subsidiarity doctrine, but it is better understood as a byproduct of the original choice for occupationalism taken during the Liberal period, reinforced by the DC’s clientelism during the post-War period.

4.3

Post-War Policy Drift and the Creation of a Familialist Welfare State

As we have seen, the DC’s clientelism made Italy’s fragmented occupational pension system highly resistant to reform, even once it became clear that pension expenditures were far exceeding available resources and crowing out other forms of social spending. In recent years, this phenomenon has provoked anxieties about intergenerational equity and population decline, in addition to the usual concerns over fiscal sustainability (see for example Commissione Onofri 1997; Rossi 1997). The reasons for concern lie in the fact that, in a society that has experienced a dramatic decline in fertility since World War II, the burden of caring for both children and ageing parents falls mostly on prime-age women, but older males receive the most generous welfare state benefits. At the same time, very weak unemployment benefits and active labor market policies promote the long-term dependence of ‘youth’ unemployed (up to age 30) on their parents. These pathologies of the Italian welfare state are well-known, and usefully summarized by Esping-Andersen (1996). The pension-heavy mix of welfare spending has important effects on the structure of Italian families. Weak active labor market policies and the lack of unemployment benefits for first-time job-seekers force many young

112

adults to delay independent family formation and remain in the households of their fathers and mothers. A paucity of publicly provided care services also discourages female labor force participation, since adult women are often called upon to provide care for pre-school and school-aged children, as well as elderly parents and parentsin-law. These lacunae in Italian social policy are acceptable to voters in large part because a generous pension system keeps resources flowing into the hands of older men, who then redistribute the resources that subsidize the (labor market) inactivity of their wives and grown children (Jurado and Naldini 1996). Subsidiarity doctrine could not have invented a better way to preserve traditional patterns of authority.13 Ironically, though, subsidiarity doctrine was not the motivating force behind the Italian welfare state’s pension-heavy policy mix. Elderly-oriented welfare states like those in Italy are better interpreted as the product of early institutional choices about program design and the competitive strategies of politicians (Lynch 2006). This takes us once again back to policies chosen by anticlerical policy-makers in the Liberal period, and sustained by the clientelism of the DC. Liberal policy choices and the DC’s clientelist style are linked to the pension-heavy mix of the Italian welfare state in two steps. First, over the course of the twentieth century, fast-growing occupational social insurance programs and minimal state activity in social assistance have come to encourage a pension-heavy spending mix. Second, as we have seen, clientelist politics reinforce occupational social programs, keeping them in place for long enough to produce their distinctive effects on the ageorientation of social spending. Taken together, these two steps imply that the familialism and male-breadwinner orientation of the Italian welfare state is in large part a result of unchecked policy drift. That is, occupational social programs put into place during the Liberal period result, in the late twentieth century, in a pension-heavy policy mix that was not the intention of the welfare state’s founders. The clientelist practices of the DC are a crucial ingredient of this policy drift, because they lock occupationalist policy institutions into place despite important changes in social structures and labor markets.14

13 Ultimately, however, to the extent that these policies also may have contributed to delayed marriage and declining fertility, they may also paradoxically undermine traditional family structures. 14 Hacker (forthcoming) describes similar processes of policy drift in the American welfare state.

113

Occupationalism has led to a pension-heavy welfare state in Italy slowly, gradually, and largely as an unintended consequence of program decisions made in the late nineteenth and early twentieth centuries. The occupational welfare state programs inaugurated in the Liberal period protected what was at that time the members of a youthful core workforce and their families, leaving responsibility for social assistance for the poor elderly and families with no male breadwinner in the hands of the Church. But these same social policy institutions imply different patterns of social spending on different social groups in the late twentieth century than they did one hundred years ago. As public employment-related pensions have expanded to cover more people in more sectors of the economy, and as large numbers of fully vested pensioners have come to draw generous benefits from a mature public occupational pension system, more and more welfare state spending has come to be directed at the elderly rather than prime-age workers. This increasingly expensive occupational pension system is financed through mounting payroll taxes that discourage new hires, ensuring that public spending on occupational benefits even for active workers in the protected core of the labor market increasingly benefits older (usually male) workers. The Italian system of unemployment supports, for example, allocates the most generous benefits (CIG and mobilitá) primarily to older workers, while among the 60 per cent of the unemployed in Italy who are under age 30 (Eurostat 1998), only 4 per cent receive unemployment insurance or assistance benefits (Schmid and Reissert 1996). The apparent familialism and male-breadwinner focus of the Italian welfare state results from very high levels of spending on pensioners, directing public resources primarily into the hands of older male family heads and crowding out other forms of social spending like services for families or active labor market policies to benefit first-time job-seekers. Relatively static welfare state policies have interacted with dynamic social structures and markets for labor and insurance to create a pattern of social policy spending that has matured, or ‘drifted’, over time. But what kept these welfare state institutions static? Clientelism has served as a powerful brake on the development of citizenship-based alternatives to the basic occupational structure set in place in the Liberal period. High pension spending is a result of clientelist politicians’ use of the fragmented occupa-

114

tionalist system The preservation of that system is in turn due to the benefits, such as divisibility and opacity, that it provides to clientelist politicians. In this sense, the familialism of the Italian welfare state is Christian democratic in origin. But this familialism is not a result of social Catholicism implemented by a Christian democratic party. Indeed, the secularization of Italian society over time suggests that the impact on social policy of societal Catholicism should be visible in Italian social policy earlier, rather than later. It is relevant to note that other apparently familialist aspects of the Italian welfare state are also of rather recent vintage. Both means-tests that take into account the financial resources of extended families and obligations to provide care services for non-nuclear family have been implemented only recently, as policy makers searched for savings in social programs already squeezed to the limits by growing pension expenditures (Saraceno 1994; Addis 1998).

4.4

Conclusion

This chapter finds little evidence that Christian democracy has determined the shape of the Italian welfare state through the mechanisms of parties that carry and enforce a particular set of socio-religious doctrines. The fragmented occupational nature of the public pension system is not an outgrowth of organicist social Catholic doctrines. Rather, it results from the DC’s clientelist extension of an occupational pension system established based on étatist principles. Neither can the familialism of a welfare state that is dominated by pension spending, squeezes cash benefits for families, and provides little in the way of social services, be attributed to Catholic subsidiarity doctrine. Instead, we need to look to long-term processes of welfare state change and non-change to explain why the Italian welfare state directs most of its resources to older male family heads. Italy’s welfare state institutions give the illusion of a social Catholic imprint, an impression that is heightened by the fact that the Church and the Christian democratic party have both significantly influenced the development of these institutions. But while the ideological foundations of social Catholicism have inspired generations of Catholic activists, and while the ‘little tradition’ of social Catholicism may have pro-

115

duced public demand for status-reinforcing and familialist welfare policies15, Italy’s is not a welfare state that is either predominantly inspired by elite- or mass-level social Catholicism or constituted by a set of policy compromises designed to ensure cross-class harmony. The occupational segmentation of the Italian welfare state was a result of policy battles fought out among the competing secular interests of Liberal Italy. The passive familialism of a welfare state that leaves social service provision to non-state actors derives from the Liberal state’s choice not to challenge the Church’s control over this domain. And the male-breadwinner orientation that is so much in evidence today results, quite unintentionally, from the DC’s strategic interest in maintaining institutions that end up crowding out spending on entitlements for women and children. Each of these attributes of the Italian welfare state was probably encouraged and facilitated by the fact that it did not contradict ideologies and values nurtured in society by a powerful Catholic church and lay Catholic activists. But both the origins of the Italian welfare state and the persistence of particular institutional forms that appear to conform with social Catholic doctrines are better explained by the behavior than by the ideologies of Christian democratic (and non-Christian democratic) political actors.

Bibliography Abramovici, Gerard. 2003. Social Protection: Expenditure on Cash Benefits and on Benefits in Kind. Statistics in Focus Theme 3 – 16/2002 (Eurostat). Addis, Elisabetta. 1998. Gender in the Italian Welfare State Reforms. Florence, Italy: European Forum, European University Institute. Seminar Paper WS/1. Baccaro, Lucio. 1999. The Organizational Consequences of Democracy: Labor Unions and Economic Reforms in Contemporary Italy. PhD diss., Sloan School of Management and Department of Political Science, Massachusetts Institute of Technology, Cambridge, MA. Baldwin, Peter. 1990. The Politics of Social Solidarity: Class Bases in the European Welfare State, 1875–1975. New York: Cambridge University Press. Bartocci, Enzo. 1999. Le politiche sociali nell’Italia liberale: 1861–1919. Rome: Donzelli. Castellino, Onorato. 1976. Il Labirinto Delle Pensioni. Bologna: Il Mulino. 15 Note however that the roughly 40 per cent of Italian voters who have supported Left parties and their goals of welfare state universalism belie a monolithic interpretation of public opinion in this regard.

116

Castles, Francis. 1994a. On Religion and Public Policy: Does Catholicism make a difference? European Journal of Political Research 25 (1): 19–40. Castles, Francis. 1994b. On religion and public policy: The case for covariance. Reply to Göran Therborn. European Journal of Political Research 26 (1): 111–115. Castles, Francis and Mitchell. 1993. Worlds of Welfare and Families of nations. In Families of Nations: Patterns of Public Policy in Western Democracies, ed. F. Castles. Dartmouth: Aldershot. Cazzola, Franco. 1972. Partiti, correnti e voto di preferenza. Rivista italiana di Scienza politica 2 (3): 569–588. CENSIS. 1983. Spesa pubblica e politica sociale. Libro bianco sulla crisi dello stato assistenziale. Milan: Angeli. GIL. 1970. La CGIL dal VI al VII congresso. Atti e documenti. Rome: Editrice Sindacale Italiana. CGIL Segreteria Generale. 1962. Sui miglioramentio ai pensionati e la riforma generale sull’assicurazione pensioni. In La CGIL dal 5 al 6 Congresso: Atti e documenti, edited by CGIL. Rome: Editrice Sindacale Italiana. CGIL-CISL-UIL Segreteria Interconfederale. 1970. CGIL-CISL-UIL a Rumor sulla politica economica e sociale. In La CGIL dal VII all’ VIII Congresso. Rome: Editrice Sindacale Italiana. Cherubini, A. 1977. Storia della previdenza sociale in Italia (1869–1960). Rome: Riuniti. CISL Consiglio Generale. 1950. Mozione sulle linee d’indirizzo e sugli obiettivi dell’azione sindacale. In Documenti ufficiali dal 1950 al 1958, edited by CISL. Rome: CISL. CCC. 1956. Mozione relative all’attuazione del Piano Vanoni per lo sviluppo del reddito e dell’occupazione. In Documenti ufficiali dal 1950 al 1958, edited by CISL Ufficio Studi. Rome: CISL. CCC. 1958. Raccomandazione della CISL ai partiti nella imminenza delle elezioni politiche. In Documenti ufficiali dal 1950 al 1958, edited by CISL. Rome: CISL. CNEL. 1963a. Osservazioni e proposte sulla riforma della previdenza sociale. Rome: CNEL. CCC. 1963b. Relazione preliminare sulla riforma della previdenza sociale. Rome: CNEL. Commissione D’Aragona (Commissione per la riforma della previdenza sociale). 1948. Relazione sui lavori della Commissione, 4 luglio 1947 – 29 febbraio 1948. Rome: Ministero del Lavoro e della Previdenza Sociale. Commissione Onofri (Commissione per l’analisi delle compatabilitá macroeconomiche della spesa sociali). 1997. Relazione Finale. [accessed May 2006] http://www.csmb.unimo.it/docrb/54_commissione_onofri1997.pdf. Esping-Andersen, Gøsta. 1990. The Three Worlds of Welfare Capitalism. Princeton: Princeton University Press. CCC. 1996. Welfare States without Work: The Impasse of Labour Shedding and Familialism in Continental European Social Policy. In Welfare States in Transition, ed. G. Esping-Andersen. London: Sage. CCC. 1999. Social Foundations of Postindustrial Economies. Oxford: Oxford University Press. Eurostat. 1998. Labour Force Survey Results 1996. Luxembourg: Office for Official Publications of the European Communities.

117

Fargion, Valeria. 1997. Geografia della cittadinanza socialein Italia: Regioni e politiche assistenziali dagli anni settanta agli anni novanta. Bologna: Il Mulino. Ferrera, Maurizio. 1984. Il welfare state in Italia: Sviluppo e crisi in prospettiva comparata. Bologna: Il Mulino. CCC. 1993. Modelli di solidarietá: Politica e riforme sociali nelle democrazie. Bologna: Il Mulino. Franco, Daniele. 1993. L’espansione della spesa pubblica in Italia. Bologna: Il Mulino. Gorrieri, Ermanno. 1972. La giungla retributiva. Bologna: Mulino. Gustapane, Enrico. 1989. Le origine del sistema previdenziale: la Cassa nazionale di previdenza per l’invaliditá e per la vacchiaia degli operaia. Previdenza Sociale 1: 35–98. Hacker, Jacob. 2005. Policy Drift: The Hidden Politics of US Welfare State Retrenchment.” In Beyond Continuity: Institutional Change in Advanced Political Economies, ed. W. Streeck and K. Thelen. New York: Oxford University Press. Huber, Evelyne, Charles Ragin and John D. Stephens. 1993. Social Democracy, Christian Democracy, Constitutional Structure and the Welfare State. American Journal of Sociology 99 (3), 711–749. Istituto Nazionale della Previdenza Sociale (INPS). 1962. Per una storia della previdenza sociale in Italia: studi e documenti. Rome: INPS. Jurado Guerrero, Teresa, and Manuela Naldini. 1996. Is the South So Different? Italian and Spanish Families in Comparative Perspective. South European Society and Politics 1 (3): 42–66. Kersbergen, Kees van. 1995. Social Capitalism: A Study of Christian Democracy and the Welfare State. London: Routledge. Kersbergen, Kees van, and Uwe Becker. 1988. The Netherlands: A Passive Social Democratic Welfare State in a Christian Democratic Ruled Society. Journal of Social Policy 17 (4): 477–99. LaPalombara, Joseph. 1964. Interest Groups in Italian Politics. Princeton: Princeton University Press. Lynch, Julia. 2001. The Age-Orientation of Social Policy Regimes in OECD Countries. Journal of Social Policy 30 (3): 411–36. Lynch, Julia. 2006. Age in the Welfare State: The Origins of Social Spending on Pensioners, Workers and Children. New York: Cambridge University Press. Manow, Philip. 1994. Die Sozialversicherung in der DDR und der BRD, 1945–1990: Über die Fortschrittlichkeit rückschrittlicher Institutionen. Politische Vierteljahresschrift 35 (): 40–61. Maestri, Giovanni Ezio. 1994. Rappresentanza degli interessi, partiti e consenso: Giungla pensionistica, clientelismo e competizione politica in Italia. Milan: FrancoAngeli. Millar, Jane, and Jane Warman. 1996. Family Obligations in Europe. London: Family Policy Studies Centre. Naldini, Manuela. 2003. The Family in Mediterranean Welfare State. London: Frank Cass. Navarro, Vicente, John Schmitt, and Javier Astudillo. 2004. Is Globalisation Undermining the Welfare State? Cambridge Journal of Economics 28 (1): 133–152. OECD. 2003. OECD Social Expenditures Database (SOCX). OECD 2003 [accessed November 2003], www.sourceoecd.org.

118

Paci, Massimo. 1984. Il sistema di Welfare italiano tra tradizione clientelare e prospettive di riforma. In Welfare state all’italiana, ed. U. Ascoli. Rome: LaTerza. Parisi, Arturo and Gianfranco Pasquino. 1977. Relazioni partiti-elettorali e tipi di voto. In Continuitá e mutamento elettorale in Italia, ed. A. Parisi and G. Pasquino. Bologna: Il Mulino. Pasquino, Gianfranco. 1972. Le radici del frazionismo e il voto di rpeferenza. Rivista Italiana di Scienza Politica 2 (2): 353–368. Quine, Maria Sophia. 2002. Italy’s Social Revolution: Charity and Welfare from Liberalism to Fascism. New York: Palgrave. Regini, Marino, and Gloria Regonini. 1981. La politica delle pensioni in Italia: Il ruolo del movimento sindacale. Giornale di diritto del lavoro e di relazioni industriali 10: 217–42. Regonini, Gloria. 1984. Il sistema pensionistico: Risorse e vincoli. In Welfare state all’italiana, ed.U. Ascoli. Rome: La Terza. CCC. 1996. Partiti e pensioni: Legami mancanti. In Il gigante dai piedi di argilla: La crisi del regime partitocratico in Italia, ed. M. Cotta and P. Isernia. Bologna: Il Mulino. Ritter, Gerhard. 1983. Social Welfare in Germany and Britain. New York: Berg. Rossi, Nicola. 1997. Meno ai padri, piú ai figli. Bologna: Il Mulino. Saraceno, Chiara. 1994. The Ambivalent Familism of the Italian Welfare State. Social Politics 1 (1): 60–82. Schmid, Gunther and Bernd Reissert. 1996. Unemployment Compensation and Labour Market Transitions. In International Handbook of Labour Market Policy and Evaluation, edited by G. Schmid, J. O’Reilly and K. Schomann. Cheltenham, UK and Brookfield, MA: Edward Elgar. Sepe, Stefano. Le amministrazzioni della sicurezza sociale nell’Italia unita (1861–1998). Milan: Giuffre. Therborn, Göran. 1994. Another way of taking religion seriously: Comment on Francis G. Castles. European Journal of Political Research 26 (1): 103–110. Zuckerman, Alan. 1979. The Politics of Faction: Christian Democratic Rule in Italy. New Haven: Yale University Press.

119

5

Religion and the Welfare State in the Netherlands1

Kees van Kersbergen

5.1

Introduction

Two major puzzles have occupied research on the Dutch welfare state. First, why was social policy so late to emerge and why did it remain underdeveloped for such a long time? Second, why did the Dutch welfare state in the period after the Second World War so suddenly and rapidly expand to become one of the most comprehensive, generous and passive of its sort? The second puzzle is to some extent the reverse image of the first, because ‘catching up’ is only possible for a country that first lagged behind European developments. The generosity and passive character of the Dutch welfare state in the 1960s and 1970s, however, are a distinct conundrum. As is stressed in the introductory chapter of this volume, there are three major sources of variation in welfare state development: the timing of the development of the self-regulating market in the 19th century, the long standing religious differences since the Reformation and the cleavage structuring in politics as an effect of the choice of electoral system in the beginning of the 20th century (in the Dutch case in 1917). Measured against such indicators as wealth and urbanization, the Netherlands between 1850 and 1890 qualifies as a developed and modernized economy that, according to the modernization tradition in welfare state research, should have developed modern social policies early on. Because it did not, the country can be classified as a welfare state laggard or latecomer. The country’s economic wealth and development stemmed from commerce, shipping, agriculture, banking and finance. Industrialization did not really take off until the 1890s and a slow maturation of industry did not took place until after 1895 (Klein 1980; Van Zanden and Griffiths 1989; Lademacher 1993). In fact, the Dutch economy would never really develop an industrial economic system comparable to Belgium or Germany, although it did, of course, develop the Polanyian self-regulating market.

1 I would like to thank Philip Manow for his critical comments and helpful suggestions. This chapter draws and expands on earlier work on the Dutch welfare state (see Van Kersbergen

120

If we look at labour law, it seems that the Dutch development was only somewhat late compared to events elsewhere in Europe. The first piece of this type of modern social legislation in the Netherlands was the Child Law of 1874 (legislated by the Liberal Van Houten), which put a curb on female and child labour (Van Oirschot 1950; Wolffram 2003). Then followed the Labour Law of 1889 (that was to combat dangerous and excessive labour of young and female workers and put a legal ban on all child labour) and the Workplace Safety Law of 1895. As Obinger (this volume) shows, Switzerland is a parallel case. Here labour legislation is very early and social protection comes relatively late. This is taken as evidence for the thesis that Liberal (Reformed-Protestant) currents played a distinctive role in early modern social policy (Manow 2004). The model of Protestant asceticism could fully justify labour safety, but everything beyond that fell in the realm of individual responsibility and discipline. Historical research (Hoogenboom 2004) shows that laws that mimic Bismarckian social insurance were circulating already in the 1880s and proposed in the 1890s. However, modern social legislation was not introduced until well into the 20th century. The reform of the old Poor Law dates from 1912, only to be replaced by the Public Assistance Act of 1963/65. The Invalidity Act was introduced in 1913, the first Unemployment Decree dates from 1917, but was not replaced by a full-fledged Unemployment Insurance until 1952. The Netherlands was therefore late in the introduction of social insurance for industrial workers and the relevant question is: what and who blocked the timely introduction of these laws? This chapter attempts to provide the solution to the puzzles of lateness and generosity-cum-passivity. It is structured as follows. Section 2 documents religious segmentation in the Netherlands and highlights that powerful orthodox Protestants were uninterested if not very reluctant in developing modern social policy. In addition, a Liberal anti-interventionist legacy, vulnerable Catholicism and weak Socialism further explain late development. In section 3 I provide a solution to the second puzzle. The postwar transformation of the political landscape marginalized Calvinist politics and brought to power a Roman–Red coalition that had economic recovery and growth and social policy development as its core programme. In the 1960s, Liberal– Confessional (mainly Catholic) coalitions came to power that rapidly expanded the

and Becker (2003 [1988]); Van Kersbergen 1995: chapter 6; Van Kersbergen 1997).

121

system under favourable conditions of economic growth and reinforced the already in-built, path dependent expansionary mechanisms. Section 4 concludes.

5.2

Cleavage Structure, Pillarization and Late Social Policy Development

Religious segmentation has characterized the history of the Netherlands since the Reformation and the revolt against the Spanish (1568–1648). The Peace of Munster (1648) settled the religious war and established the independence of the Dutch republic. The political impact of the fundamental religious cleavage on Dutch society, both expressed as a Liberal–Confessional antithesis and as a Calvinist–Catholic conflict, can hardly be overstated. Religious strife and dissension are major continuities in the history of this nation. In fact, religious conflict is the single most important feature of what constitutes the nation. The religious cleavage controlled Dutch politics during the second half of the nineteenth century; it was the single most important cause for the rise of the ‘pillarized’ structure of the political system (1917–1967); and it determined the dominance of Christian and Catholic parties during most of the twentieth century. Dutch Catholics were a large, oppressed minority in the nineteenth century and much of their social and political activities involved their much sought after emancipation (Bax 1995). The re-establishment of the Catholic hierarchy in the 1850s, allowed by the Liberal government, had aroused the anger of the Protestants and reinforced the Calvinist–Liberal conflict, but at the same time signified the inauguration of a long and laborious struggle for Catholic emancipation (Kossmann 1978: 278). Nevertheless, the Calvinists were first to organize politically. The orthodox Protestants had a hard time deciding who were the bigger enemy: the Catholics or the Liberals. Clearly, according to myth, the Dutch nation was founded in the Reformation and the struggle against Catholic Spain. Public life and public education in the Dutch Republic (until 1795) were clearly dominated by the Dutch Reformed church (Woltjer 1992: 27). Since the Constitution of 1798 of the Batavian Republic that separated church and state, there was formal freedom and equality of religion, but in the eye of the political, economic and cultural elite Dutch public life and the nation were indisputably still Protestant (Van Rooden 1999), albeit of the liberal, freethinking type that

122

the Dutch Reformed church had gradually embraced and that the orthodox Calvinists fiercely opposed. The orthodox Calvinists seceded from the Dutch Reformed church in 1886. Anti-Catholicism was part and parcel of the Calvinist worldview. Politically, however, the Liberals and their anti-religious policies (especially in education) were directly threatening Calvinist freedom. The political issue of the public financing of private schools in the end re-drafted the impact of the religious cleavage by moderating the Calvinist–Catholic conflict and by reinforcing the antithesis between the Liberals on the one hand and the religious forces on the other hand. The school issue itself had induced the political organization of orthodox Calvinism already in the 1870s. Moreover, the school issue also fundamentally reorganized the political landscape with respect to other crucial issues such as the extension of the suffrage and the social question, and supplied the foundation for the social and political pattern of highly integrated yet segregated sub-cultural organization named pillarization. The politically dominant Liberals intended to create a universal, state-controlled and religiously neutral educational system and passed a law for that purpose in 1878. This law obviously threatened to topple the Confessional schools and led to the foundation of the first Dutch mass party, the Anti-Revolutionaire Partij (ARP) in 1879 by the Calvinist minister Abraham Kuyper. The party was anti-revolutionary to the extent that it rejected the tenets of the French revolution and the enlightenment. The term, therefore, refers to ‘a distinction between “counterrevolutionaries” who would wish to reverse history and “antirevolutionaries” who, while rejecting the entire spirit of 1789, would yet recognize change’ (Daalder 1966: 199, footnote 27). The ARP was not reactionary, but rather ‘isolationist’, struggling for the survival of the Calvinist minority and the limitation of state intervention in what Kuyper believed to be the vital and essentially free spheres of social life (the family, the church, the school). It was a genuine party with a cross-class appeal, although it was disproportionally supported by the orthodox lower middle classes and lacked substantial workingclass support. Among its leaders were middle class men as well as aristocrats and patriciates. The party not only organized the Calvinist electorate, but also those who were denied the right to vote (the mass of the people behind the voters). In this sense the ARP was indeed an evangelical people’s party (Irving 1979: 196). The ARP was successful electorally and on average controlled about 20 per cent of seats in the

123

Tweede Kamer (lower house) between 1888 and 1913. In the inter-war years its electoral appeal declined gradually to approximately 14 per cent of seats (Mackie and Rose 1991: 331–5). Its record in government, however, is more impressive. During 1888–1891 it formed the first Confessional coalition with the Catholics. This was a novelty in Dutch politics and was repeated in 1901–1905 and 1908–1913. The ARP played a pivotal role in government formation during the entire period 1918–1940 and it provided the prime-minister five times. Catholic social and political emancipation was notoriously slow and although the priest Herman Schaepman had already written a Catholic political program in 1897, it took until 1926 before an authentic Catholic party, the Roomsch Katholieke Staatspartij (RKSP), was founded (see Kalyvas 1996). This does not imply, however, that Catholics, who were in many ways treated as second-class citizens, refrained from participating actively in national politics. On the contrary, they played at first a small, but gradually a more important role in several coalition governments with the AntiRevolutionaries between 1888 and 1940, especially but not exclusively in social policy. The school question encouraged Catholic political mobilization too and detached them from the secular Liberals, whom they initially had supported because an alliance with the Liberals promised greater chances of liberty for the Catholic populace. But the Liberal attack on Confessional schools brought them nearer to the orthodox Protestants. The paradoxical aspect in the development of Dutch Catholic politics is that it was in fact greatly encouraged by the political progress of the Calvinists. At the time of the school question, Kuyper formulated his political theory of the ‘antithesis’. According to this theory, it was no longer the conflict between the true faith of Calvinism and the false belief of Catholicism that fundamentally divided the Dutch politically, but rather the cleavage between faith (of whatever creed) and incredulity. This doctrine cemented the anti-Liberal coalition of the Confessional forces. Quite unexpectedly, therefore, the Catholics had found a partner in the Calvinist party through which they could exercise some political influence in their attempts to improve the situation of their brethren, particularly promoting social reform and taking responsibility for it. Although the inclusion in national politics mitigated the sense of inferiority among Catholics, their leaders were eager to avoid new waves of anti-papism that they ex-

124

pected to follow from further Catholic mobilization (Wolinetz 1983: 648). In addition, Catholic politics was no different in its internal discord from its equivalent elsewhere in Europe at the same time, as a result of which the formation of a unitary Catholic party was further delayed (Irving 1979: 199; Kalyvas 1996). Yet Catholics did establish electoral committees at the turn of the century that survived until the RKSP replaced them in 1926. Similar to other Catholic parties, the RKSP was an outright Confessional party and comprised an amalgam of the political and social organizations of the Catholic sub-culture, which by that time was already well-established. In terms of electoral strength, the various Catholic representatives-without-a-party commanded an average of over 26 per cent of seats in the Tweede Kamer between 1888 and 1925. The RKSP won about 30 per cent of seats between 1929 and 1940, almost perfectly reflecting the numerical strength of the Catholic minority in the Netherlands. The most decisive event in Dutch political history for both Catholic and Protestant politics before World War II undoubtedly concerned the ‘Pacification’ of 1917, that is the simultaneous settlement of several sharply dividing issues. The new constitution of 1917 granted the financial equalization of public and private primary schools and introduced universal manhood suffrage (and a provision for female suffrage) under a proportional system in a single national constituency. As Gladdish (1991: 28) rightly put it, ‘(…) the new electoral system recognised, reflected and reinforced the crystallisation of political sub-cultures (…)’ which was to determine Dutch politics at least until the mid-1960s. In line with the theoretical thrust of this volume’s introduction, the proportional representation (PR) electoral system did indeed lead to a firm translation of the religious cleavages into politics and partly repressed the class cleavage, particularly because the Catholic pillar organized Catholic workers too. Moreover, PR led to the political marginalization of the Liberals and stimulated the consolidation of distinctly modern centralized, hierarchical and programmatic mass parties (Van der Kolk and Thomassen 2006). These parties represented the cleavage structure of Dutch society and reinforced it institutionally. In the summary of Van der Kolk and Thomassen (2006: 122): The electoral system designed in 1917 and fine-tuned between 1917 and 1937 reflected a core element of Dutch politics and Dutch society. Until the 1960s, ‘pillarization’ was the dominant social characteristic of the Netherlands. Catholics, secular workers, and Protestants all established their own social networks, including labour unions, newspapers, broadcasting organizations, etc. Each network, or pillar, was represented by its own political party. Many members of the pillars voted for the political party representing

125

their pillar and elections were often seen as a mere census. The electoral system guaranteed all major groups to be represented in parliament. Since the relative strength of the groups did not change very much, and because turning out to vote was compulsory, the net change of seats between two elections was virtually negligible.

The social and political system of pillarization together with the proportional system was particularly beneficial to the mobilization of Catholic power in the inter-war years. The RKSP, as the political representative of a comprehensive network of Catholic social and cultural organizations, became the single largest party in parliament. Moreover, pillarization contributed to the political dominance of Confessionalism in more general terms. The Confessional parties together systematically managed to control a comfortable absolute majority in the Tweede Kamer as well as in government. The RKSP represented the tradition of Catholic social concern, especially through its labour wing, and through the notion of subsidiarity attached importance to public social intervention. The Protestant parties’ concern was with self-organization and religious freedom, especially in education. In 1901, the Confessional parties won the elections at the cost of the Liberals who – in vain – had hoped that their (faltering) social policies would bring them the workers’ vote. The Protestant parties and the Catholics could form a majority coalition. Kuyper himself took the lead, but refused to support an increase in public expenditure, not even for subsidizing religious schools. The ARP had developed the concept of ‘sovereignty in one’s own circle’ primarily as a means to secure the autonomy of lower societal circles and spheres, including the family, the church and social organizations, from the state. All authority was argued to derive from God and the essentially free circles of social and economic life did not have to acknowledge the authority of the state: ‘(…) the highest authority in any circle has nothing above itself but God, and the state cannot intervene and cannot command on the basis of its own power’ (Kuyper 1898: 79, my translation). In Kuyper’s eyes, Calvinism defended God’s right and sovereign authority in the social spheres of life and protested against the omnipotence of the state and against ‘the horrible perception as if there were no Law above and outside the prevailing laws (…) (Kuyper 1898: 86).

126

In social policy matters, the boundaries of intervention were reached with the protection of the labour of the adult male worker. Kuyper lost the remnants of his ‘social face’ in the wake of the railway-strikes of 1903, to which he as responsible minister of the interior reacted with the so-called ‘strangling’ laws: he made strikes illegal for government and railway personnel (Wolffram 2003: 140–3). It is no surprise that even the Protestant philosopher Woldring and the Protestant sociologist Kuiper (1980: 40– 51), in their study of the Protestant critique of society, qualified the contribution of the ARP to the solution of the social question as ‘disappointing’. Abraham Kuyper, once in government (1901–1906), in spite of the ambitions of the ARP, did not try to accomplish anything in the area of social legislation, while he even rejected (but could not block) the (limited) social policy initiatives of his kindred spirit Syb Talma (member of government from 1908 to 1913). It has long remained unclear why Kuyper was so inactive in the field of social policy, especially as the official government programme of the coalition that he led contained several proposals for new social laws (e.g. a sickness law, a disability and old age pension, an accident law). Marcel Hoogenboom’s (2004: 135) historical research, however, has revealed that Kuyper had a secret pact with the leader of CHU, a conservative and anti-Papist union of Dutch-Reformed aristocrats, not to propose these social laws, unless additional financial means would become available from extra import taxes. Taxes, however, were not raised, as a result of which the social laws were blocked. Hoogenboom (2004: 136) comments: ‘In this way the development of social security in the Netherlands in the first half of the 20th came to an almost complete standstill’. Generally speaking, this is an example typical of the internal politics of compromise in religious parties. Once a party mobilizes both lower and middle-class interests, as religious parties (in this case the ARP) did, they need to find both an internal compromise and an external coalition in all questions of redistribution (see Manow and Palier, this volume, who discuss the compromise between the Catholic workers and the Catholic patronat in the French case). Note, however, that internal compromises and crossparty coalition-building do not necessarily always lead to anti-interventionist solutions, as the analysis (below) of the Catholic party and its coalition strategies will demonstrate. The ARP’s general approach towards social intervention was strongly influenced by internal party compromises and external coalition building, but received its distinc-

127

tive anti-interventionist character because it had to be adjusted continuously to the theory of sovereignty in one’s own circle. This governed the party’s ideology and political practice well until the 1970s. It is not surprising, therefore, that this party frequently voted against social legislation and as a government party did not take responsibility for social policy (Van Putten 1985: 175). Moreover, it was the ARP that largely had been responsible for a detrimental crisis policy in the 1930s. The AntiRevolutionary attempt to cope with the economic slump was permeated by dogmatic economic liberalism (Klein 1980: 5) that squared well with the political ideology, while the more active state intervention characteristic of the post-war era clearly conflicted with the traditional anti-statism of the ARP. The Dutch Protestants in the closing decades of the nineteenth century were entirely engaged in their struggle against the education policy of the Liberal governments, while the Catholics also developed ideas on social policy. This development seems in line with developments elsewhere, especially in Germany. Since the entire state elite in Germany was Protestant, they were preoccupied with the Kulturkampf as one strategy to construct the nation that was directed against the Catholics. Protestantism as a religion was private and individualized in character, but, just as in the Netherlands, at the same time the ‘unofficial’ state religion. Catholics had more eyes for the misery of the working class, which they tried to mobilize, and were already developing ideas on social policy at that time. Did in the Dutch case Protestant politics contribute to the delay of welfare state development (cf. Manow 2004)? Yes, to some extent, as the government record of Kuyper’s ARP demonstrated. But to a larger extent one must conclude, on the basis of the most recent and best historical studies available, that orthodox Protestantism was by and large inconsequential in the realm of social policy (see for instance Roebroek and Hertogh 1998; Wolffram 2003). Hoogenboom (2004) shows that the ARP had no positive or negative impact on social legislation in the early 20th century after the Kuyper-government, even though the party at several occasions made some halfhearted effort to put a halt to the Liberal, Social Democratic and Catholic reform zeal. Between 1888 and 1891, when the first coalition between Calvinists and Catholics governed, it was the Catholic Minister Ruijs van Beerenbroek who was responsible for introducing the Labour Law (1889). And even though the Calvinists occasionally voted against social legislation and governments depended on their support, their

128

weight was apparently insufficient to hinder the adoption of some social bills. When between 1908 and 1913 the Calvinists took direct responsibility for social legislation, they primarily executed what had already been prepared in the 1890s by the Liberals (Wolffram 2003). It seemed that Dutch orthodox Protestants around 1900 were primarily disinterested in social reform to the extent that they would not even withdraw their support for government when they lost a parliamentary battle over new social legislation that they objected to. This lack of social concern and interest in social issues put the Calvinists apart from the political current of Christian Democracy that became one of the pillars of the expansion of the Dutch welfare state. In fact, it was political Catholicism that was the prime motor behind the establishment of the cross-Confessional Christian Democracy in the Netherlands in the mid-1970s. It was not until the Catholic party started to lose its working class support (both electorally and organizationally with the deConfessionalization of the Catholic labour movement in the 1960s; see Roes 1985) that it took the initiative to establish a cross-class and cross-Confessional Christian Democratic party. The merger of the denominational parties into the CDA was facilitated by an ideological rapprochement of political Catholicism and Protestantism (Zwart 1993). By the end of the 1960s, the ARP had radicalized its ideology and had been transformed into a radical-evangelical political party with a much more positive view on the role of state intervention in social matters. However, by that time, the entire building of the Dutch welfare state had already been erected under the Roman–Red coalitions between the Catholic party and Social Democracy in the 1950s, while two final major pieces of social legislation – the General Social Assistance Act of 1963 and the new Disability Law of 1967 – had been introduced under the Catholic–Liberal coalitions of the 1960s and by Catholic ministers. Recently there has emerged a ‘revisionist’ school in Dutch historiography (see De Haan 2003) that de-emphasizes the role of religious forces, pillarization and consociational democracy and instead focuses on how a Dutch nation could evolve around moral (e.g. Hoekstra 2005) and social (e.g. Wolffram 2003) issues despite religious fragmentation and separate institutions for religious communities. Hoogenboom (2004) argues that pillarization and Confessional politics can only partly explain ‘lateness’ of Dutch welfare state development. An equally important factor was the continuing power of the ancien régime in modernizing Dutch society, particularly the

129

landed aristocracy and, peculiar for the Dutch trading and commercial society, the rich urban patriciate. Both late development and the mix of pre-modern and modern forms of social security are said to be explained by this phenomenon. Although Hoogenboom presents compelling evidence to support his position, he at the same time reinforces the more traditional or mainstream position on the importance of religion by showing that the estates cleavage ran squarely within the religious communities of orthodox Protestants and Catholics. The ruling estate of aristocrats and patriciates converged around a politics of opposing and blocking modern social reforms and favouring pre-modern solutions in the late 19th century. But the crucial difference in my view was this: within the Catholic pillar these forces met stiff opposition from the Catholic workers, while no such strong counterforces were present within the orthodox Protestant circles. In other words, in order to understand the peculiarities of the history of the Dutch welfare state and its ‘lateness’, it is still imperative to focus on the role of Calvinism and its political representative, the ARP. More generally, this underscores the earlier observation on the importance of the game of intra-party class politics and external coalition building that religious parties have to play and the diverse outcomes these can produce.

5.3

Cross-cleavage and Cross-pillar Coalitions and the Postwar Expansion of the Welfare State

The Dutch welfare state developed into a highly passive, transfer-oriented, servicelean yet highly generous system that was tailored to income replacement for the typical male breadwinner–female carer household. Moreover, one of the conspicuous aspects of the Dutch welfare state is that active labour market policies never assumed any prominent place within the institutional framework of social policy. At best full employment was considered to be an effect of the attainment of other macroeconomic ends, such as wage moderation. The commitment was to financial compensation for unemployment and other labour market risks, that is, to income replacement (Therborn 1986; Braun 1988). Why did the Dutch welfare state develop rapidly into this passive yet generous transfer state? In a nutshell, the expansion of the Dutch welfare state hinged on a coalition between Catholic and Social Democratic reformist forces and the exclusion of anti-

130

interventionist orthodox Protestantism and free market Liberalism (Van Kersbergen and Becker 2003 [1988]; Roebroek 1993; Van Kersbergen 1995). The origins of the passive and generous architecture of the welfare state lie in the first decades after the war, when the critical, constitutional decisions on the institutional configuration of the welfare state were made. It was the agreement between the Catholic party and Social Democracy over the need for macroeconomic management of the economy and social policy as a means to ease the pain of postwar reconstruction, which functioned as the cement of the so-called Roman–Red coalitions that governed the Netherlands until 1958. This political coalition was embedded in a nation-wide cross-pillar and cross-class accord on how to reconstruct the nation after 15 years of crisis and war. In retrospect then, the Second World War provided much less of a break in the political history of the Netherlands than one might have expected. The reappearance of the pre-war pillarized social and political cleavage structures and with it the resurfacing of Confessional political dominance indicated a clear tendency towards restoration. The experience of the Second World War and the Nazi-occupation had not led to a substantial change in the two-cleavage logic of Dutch politics. Yet, there was an attempt to innovate in order to block the restoration of pre-war pillar dominated elite politics. The most promising line was a new political movement, which had as its main objective the ‘breakthrough’ (doorbraak) of the old multi-party political system and social cleavage based pillarization. The main driving force behind the new Dutch People’s Movement (Nederlandse Volksbeweging, NVB) consisted of the elite of the Socialist party (SDAP) that wished to get rid of its Marxist heritage and sought to transform itself through the NVB into a broader people’s party that could appeal to the (Catholic, Protestant, and secular) middle class too. Anticlericalism too had to recede to the background for this purpose. In addition, the struggle for renewal was backed by left wing Liberals and Catholic groups united around their resistance against cleavagebased politics and an equivocal longing for unity of the Dutch people. They had their eye on the working class vote. This blend of miscellaneous political currents found its common ground in a set of ideological concepts, labelled ‘personalist socialism’, which was an odd mixture of Social Democratic, Liberal, and Catholic–Corporatist conceptions, inspired by the general doctrines of Socialism, Humanism and Christian

131

Democracy that was to legitimize this attempt at cross-class and cross-cleavage political mobilization. The shared conservative element consisted of ‘(…) a return to Christian norms, emphasis on the family as the natural basis of society, the idea of community instead of pluralism’ (Von der Dunk 1982: 197), while the progressive legacy was embodied in an ideal of a society not characterized by religious and class conflicts and the nationalization of the crucial means of production. The personalist element can be seen as ‘a christian–social variant of the old liberal individualism, in which human dignity occupied a central place’ (Von der Dunk 1982: 197). Personalism was interpreted as the unfolding of the human personality at the service of (…) a solid, just and inspired community. The concept implied a critique of the ‘collectivistic ideologies’ of national socialism as well as of Marxist–Leninism. It could be a formula to denounce class attachment and class struggle. In connection with the notion of ‘socialism’ it was understood as the quest for a just society, for a radical social policy on the basis of communitarianism (…); no class struggle and no classless society either, but regulation of the capitalist market economy and the foundation of labour relations in the legal order. (Bank 1978: 20, my translation).

The formulation of this ideology was mainly furnished by Social Democratic reformism and Catholic corporatist notions. Among the reform-minded political groups there was broad agreement on an ethical principle concerning the need for a change in popular mentality in a Christian–Humanist direction and on the demolition of the old political system. Disagreement, however, existed as to the role of the state and the institutionalization of social and economic relations (Böhl et al. 1981: 36), the latter being the main conflict between Socialist collectivism and Catholic corporatism. The ‘breakthrough’-movement, however, did not succeed in altering the twocleavage logic of party political competition in the Netherlands and existing mass political parties continued to profit and reinforce pillarized society. Still, the movement was a success (see Bosmans 1982) in three areas. First, it led to the foundation of a modernized Social Democratic party with a Personalist–Socialist, almost social Christian character. Second, it helped the modernization of the Catholic party, according it an even more reformist disposition. Finally, it facilitated the construction of a government coalition between Catholics and Social Democrats, which established the foundation of a crucial qualitative change in macroeconomic and social

132

policymaking and politically backed the broad capital–labour covenant in industrial relations. Unlike in Germany, the Dutch Confessional parties were incapable of unifying as a single block and there existed considerable disagreement between the Protestant parties on the one hand and the Catholic party on the other hand concerning the relationship between state, market and family, that is to say the three major social institutions of a nation’s welfare regime. It was the ARP that in the 1940s and 1950s refused to accept increasing state intervention. In fact, the early electoral pamphlets of the Anti-Revolutionaries rejected both the alleged ‘state socialism’ of the Social Democrats and the idea of subsidiarity of Catholicism as too interventionist. The orthodox Protestants opted for strong liberal social and economic policies. The ideological concept of Kuyper’s ‘sovereignty in one’s own circle’ remained central and was understood as the idea that could protect the freedom and autonomy of the self-governing communities of Protestantism. It contained a much more restricted view on state intervention than the Catholic theory of subsidiarity. Whereas Catholics could formulate intervention in a positive sense as the duty of the state to assist social units at lower levels of society when in need, the orthodox Protestants could only do so (from the state’s point of view) negatively as the state’s duty of non-interference with the ‘life circles’. It was this political controversy between the Protestant and Catholic parties that forestalled the unification of Christian Democratic forces and contributed to the emergence of a Catholic–Social Democratic alliance. It was therefore this conflict on the state–church dimension between Catholics and Calvinists that facilitated a coalition on the left between Social Democracy and the Catholic party, reinforced by the ARP’s liberal-economic position on the left–right dimension. This political coalition was part of an even wider ranging cross-cleavage and crosspillar social accord on how the nation should recover from the destruction and suffering that 15 years of depression and war had caused. This nation-wide accord consisted, on the one hand, of an export-oriented economic model that stressed low wages and moderate wage development (and later industrialization) as the sine qua non for economic recovery and development in an open economy, and, on the other hand, a social model that aimed at social peace and offered compensation in the form of transfers for the expected negative social effects of low wages and adjusting to

133

world market requirements. Dutch developments manifested many of the features later analytically interpreted as ‘democratic corporatism’ in the political economy models of the 1980s (e.g. Katzenstein 1985). Specifically, the vulnerability that small, open economies faced favoured the expansion of the public economy so as to reduce uncertainty via social guarantees, full employment, and more active political management of the economy. The causal chain set in motion was that this small open economy developed democratic corporatist structures as a way to enhance domestic cross-class and crosscleavage consensus, facilitate economic adjustments, and maintain international competitiveness. Consensus building via democratic corporatism was promoted by the presence of Social Democratic and Catholic labour movements, that demanded a say in policy development and were especially keen to secure full employment. Employers from all pillars feared full employment because it would drive up wages. Wage moderation in exchange for social policy under the condition of full employment became thus the positive-sum solution between employers and employees that sealed the postwar social contract (see e.g. Swenson 1991). Social policy functioned as a central means to mediate social and political conflicts. The politics of social welfare in the Netherlands has been an indispensable constituent of consociationalism and therefore of the development of a stable democracy in a segmented society. Without the development of an extensive and generous system of social compensation, the consensus democracy of the Netherlands would never have prospered as it did. The coalition between Catholic controlled Christian Democracy and Social Democracy has played a substantial and beneficial political role by contributing to the stability of Dutch democracy (see Koole and Daalder 2002; Van Waarden 2002). The likelihood of this political coalition had been greatly increased by a number of historical events and intra-political developments. First of all, the resistance of the Catholic hierarchy against Catholic–Socialist co-operation, dating from the early 1920s, had been tempered as a result of the evolution of Dutch socialism into a Christian–Humanist version of personalist Social Democracy. This tempered the political saliency of the religious cleavage. Second, during the economic crisis in the 1930s Catholic and Socialist socio-economic policy proposals had rapidly converged as they

134

both had sought to cater their working class clientele. This moderated the political significance of the class cleavage. Finally, by the late 1940s the political projects of Catholicism and Social Democracy had considerably converged on three major areas: global macro-economic demand management, social policy and family policy. With this convergence, potentially divisive policy issues receded to the background. Ultimately, the pact was sealed by anti-communism which was broadly shared by political Catholicism and Social Democracy and was politically relevant as the Communist Party of the Netherlands (CPN) polled over 10 per cent of the vote in the first national elections (1946) after the war. The experience of economic crisis, war and fascist oppression had contributed to a temporary radicalization of societal forces, which, in turn, facilitated reforms. If radicalization did not have an immediate effect on the reformist inclination, it fuelled at least anti-Communism in the Dutch context, particularly of the Catholic labour movement, consequently consolidating the construction of a Roman–Red coalition. Anti-Communism not only accelerated the integration of the non-Communist labour movement into the institutional framework of democratic corporatism, but also stimulated social policy as a means to cut the grass from under the feet of societal discontent fostered by the Communist labour movement and party. AntiCommunism also functioned as a convenient catalyst in the reconstruction of Catholic unity and the reconstitution of the Catholic cross-class pillar after the war. The cross-cleavage and cross-class social and political pact that integrated the Catholic and Social Democratic labour movement into the social and political structures of policymaking and backed the political coalition between the Catholic and Social Democratic party, got further momentum in the context of a comprehensive attempt to stimulate industrial development in a still predominantly rural and trading society (Böhl et al. 1981; Fortuyn 1980, 1983; Ter Heide 1986; De Wolff and Driehuis 1980) and to reconstruct Dutch labour relations further in democratic corporatist fashion (Windmuller 1969; Hemerijck 1992). This intensification of the labour–capital accord consisted of a Catholic–Social Democratic compromise of corporatist reorganization and the introduction of global, indicative or technocratic planning as the lubricant of the consensus machine that was beginning to take shape.

135

The macroeconomic policies of the Roman–Red coalitions were dominated by one leading goal: export-led recovery (De Wolff and Driehuis 1980: 37). The conditions for the export orientation of the Dutch economy were found in further economic integration into the European markets, tax relief for entrepreneurs, a restrictive wage and price policy, the improvement of the infrastructure, an extensive regulatory socio-economic framework (Van Eijk 1980; Ter Heide 1986) and compensatory social policies to be initiated and legislated by central government, but administered by semi-public organizations of labour and capital or other interest groups (see extensively Cox 1993). Monetary policy became the principal responsibility of the Dutch Bank which was nationalized in 1948. Five official goals of economic policy that were to guide state intervention until the mid-1960s: full employment, economic growth, a reasonable income distribution, an equilibrium on the balance of payments, and price stability. The export-oriented macroeconomic policy of industrialization originated in a ‘Dutch conception of underdevelopment’ (Therborn 1989: 209), activated by the demographic pressure of a fast growing population and the international pressure to decolonize the largest overseas colony, Indonesia. The (expected) economic loss of the Dutch Indies was feared to deprive Dutch capital of an important investment possibility. Released funds had to be reallocated and export-led industrialization would provide an outlet by the opening up of new markets. At the same time, labour quiescence and wage restraint were seen as preconditions for profitable investment and a successful industrialization strategy. A tight incomes policy until the early 1960s kept wages and inflation low and indeed improved the international competitiveness of the Dutch economy. Since labour was cheap, employment increased steadily, particularly in the industrial and service sector. Both the Christian and the Socialist unions accepted the restriction on wage development, mainly because of its favourable effect on employment, but also in exchange for the extension of social security. Only in this sense, that is in connection with wage restraint for competitive reasons, can full employment be argued ever to have been an explicit policy goal. It was a secondary target because the underlying theory was that export-led growth founded on wage restraint in combination with global measures would automatically lead to an optimal allocation of labour power. There was virtually no discussion on (active) labour market policy and no instru-

136

ments were developed in this direction. Social Democratic attempts to politicize this issue, in fact, encountered Confessional opposition. Generous replacement rates could be demanded (and in fact were demanded by the Catholic party) to be included in social legislation, since they were to replace quite moderate family wages. In addition, full employment kept the cost of social security low. Yet, given the generous structure of benefits and the absence of active labour market policies, social spending obtained a path dependent built-in tendency to rise rapidly under less favourable economic conditions. In addition to increasing coverage and generosity, this explains why, no matter what colour of the ruling coalition, social spending tended to expand in the 1960s and 1970s. Both the Catholic party and the Social Democrats promoted the family as the cornerstone of society and especially the former had a history of resisting the entry of women on the labour market. Since the 1880s in the practice of municipalities and since 1904 by national law employment policy had been characterized by measures to prohibit married women to work in periods of rising unemployment so as to redistribute work among men. For instance, when in 1924 unemployment increased, the Catholic minister Ruys de Beerenbrouck ordered by Royal Decree (Koninklijk Besluit) to fire female public employees under 45 when they got married. A Royal Decree of 1932 simply outlawed the hiring of married women as public employees and a policy guideline of 1934 instructed all government institutions not to hire women, whether married or not, and to discourage female employment more generally through their policies towards private business. The Catholic leader Romme proposed a law in 1937 to the effect that ‘Eene gehuwde vrouw mag geen arbeid verrichten’ (A married is not allowed to work, my translation, quoted in Blok 1978: 120). Resistance against this law from the non-confessional women’s organisations and female Socialist members of Parliament and the shadow of the impending war effectively prevented the introduction of this war. In 1930, 2.2 per cent of married worked, as compared to 13.4 per cent in France. Whereas around 1960 in most advanced nations (except Norway) approximately 30 per cent of women worked, this was only 16.1 per cent in the Netherlands (Blok 1978: 104). It took until 1957 that the ban on married public employees was lifted. None of the electoral manifestoes of the confessional parties in the first 10 years after the Second

137

World War promoted or even mentioned equal treatment of (married) women on the labour market. Family policy proposed by the KVP consisted of the restoration of traditional family relations (with concrete proposals on fighting divorce and contraceptives) and preparing men for industrialization. The Catholic Party’s think-tank (Centrum voor Staatskundige Vorming) published various politically influential reports on married women and the labour market (1951–1953), highlighting the duty of a married women in the family, based on the ‘spontaneous division of labour between the sexes’, the ‘moral order of peoples’, the ‘biological function of women’, and a woman’s psychological inclination to care. The observation in 1951, incidentally, was that since only 35.000 married women worked (1 per cent of the labour force) this was not a particularly pressing problem anyway. The effect of the converging views on the family as the cornerstone of society was clearly visible in the complex of incomes policies (taxation, social security, wages, prices) that was shaped as far as possible according to such notions (see Morgan, this volume) and produced the male breadwinner–female carer model of Dutch social security. Social policy originally assumed that male breadwinners were entitled to social security benefits and that such entitlements had to be sufficient to replace a family income. Women were, so to speak, insured through the wage labour of their husbands and, in exchange for this income guarantee, women were assumed to perform their family caring tasks as well as a certain amount of voluntary work. The result was a system of social insurance characterized by entitlements differentiated according to position in the household and by the absence of specific social rights for married women in particular (Bussemaker and Van Kersbergen 1994). The fact that both the Catholic and the Social Democratic party took a traditional stance on the role of the family as the fundamental constituent of society with defined gender roles did not necessarily imply that family-friendly policies became a paramount characteristic of the Dutch welfare state. On the contrary, it meant that all social policies were permeated by a bias towards reproducing the traditional division of labour between men and women. Social policy was only meant to strengthen the family to the extent that it preserved its traditional character. In this way it reinforced the existing patriarchical structure of Dutch society in the form of a gendered breadwinner–carer model of social policy that was elaborated to the extreme. In this sense, the arrival of modern social policy initially did not modify a resilient socio-cultural

138

trait of Dutch of society that consisted of a sharp separation of male paid employment and female housework that had existed since the 17th century (Pott-Buter 1993; Schama 1988). Catholic (and initially also Social Democratic) family policy was not primarily about assisting the family, but about preserving its gendered character. This is why it ultimately had, in fact, the paradoxical effect of making family life extremely difficult, reinforcing disincentives for women to enter the labour market and to make a career and the immense difficulties of families to combine and reconcile work and care later resulting in declining fertility rates, something the Catholics obviously never intended to accomplish but which they helped to produce nonetheless. Traditionally the Catholic organizations had been strong in the pillarized organization of social work. Public intervention in this area, however, resorted under the Ministry of Social Affair, a Social Democratic bulwark. The Catholic party, however, successfully dissociated social services from this Ministry and relocated them to a new department of social services established in 1952. This department became the embodiment of Catholic subsidiarity in the Netherlands in the sense that it decentralized the execution of subsidized policy to religious institutions and the private initiative as far as possible (Böhl 1981: 252–262). This partly explains the conspicuous ‘underdevelopment’ of public social services in the Netherlands. De Bruijn (1989: 170–171, my translation) comments that the key of family policy towards the vulnerable family was the ‘preservation of the closed family with the intrinsic omnipresence of the mother. Modernization threatened to break open the family. The whole network of family policy instruments and organizations was meant to prevent this; the vulnerable family needed protection and support’. Under conditions of increasing pressure from economic Liberalism – both from the Liberal party and the right wing of the Confessional parties – to withdraw from intervention in wage determination, passive social policy remained virtually the only fundamental consensus between the Social Democrats and the Catholics (Ter Heide 1986). The favourable economic conditions of the late 1950s and early 1960s and the near full employment, however, placed the strict wage policy of the Roman–Red coalitions under pressure. Given the tight labour market, employers had already started to pay higher wages than were permitted while price control had never been as strict as wage restraint anyway. One of the reasons why the coalition between Catholicism

139

and Social Democracy broke up was that the latter stubbornly refused to give up wage policy as the last bastion of interventionism, while the former had started giving in on the demands of the employers and their representatives within the party to ease the control. Within the Catholic party in the 1960s, therefore, a new social coalition was constructed that leaned much more to the (liberal) right and inaugurated the political coalition between the Catholics and the Liberals. The guided-wage system broke down entirely in 1964, under one of these Confessional–Liberal coalitions that ruled the Netherlands between 1958 and 1972. Other than the guided wage policy, the Liberal–Confessional coalitions did not fundamentally alter economic and social policy. Rather, under Catholic leadership, the coalition not only kept the basic contours of the export scenario and compensatory social policy intact, but considerably expanded its scope and performance. Once in place, the model gained momentum and developed its own logic of path dependent expansion. A wage policy was never abandoned entirely either. Under political insistence of the Catholic Minister of Social Affairs (G.M.J. Veldkamp), first unions and employers agreed upon a minimum wage and then the government introduced a statutory minimum wage. This paved the way for another central characteristic of the passive welfare state: the extraordinarily pervasive system of linking the development of wages in the private sector to the wages in the public sector and to the development of social security benefits. This form of ‘automatic government’ reinforced the persistent growth of public spending that was the result of increasing government commitments in the realm of social policy, especially when later the minimum wage was raised considerably more than the rise in productivity (Bos 2006). The loosening of the wage restriction in the late 1950s and its eventual abolishment contributed to an economic boom in the second half of the 1960s which, in turn, resulted in a large scale reconstruction of the economy. Sudden, shock-wise wage increases caused a rationalization of the labour intensive industries. Employers started to invest in labour saving techniques and massive lay-offs followed, initiating the ‘welfare without work’ syndrome (Esping-Andersen 1996; Hemerijck and Van Kersbergen 2004) that is so typical of the continental welfare states. Here the passive social system started to turn against itself. The generous disability insurance, introduced in 1967 as a hallmark of Catholic social policy, provided the employers and unions with a convenient opportunity to ease the pain of rationalization for their

140

workers. Rather than becoming unemployed, many older and economically less functional workers claimed a right to a disability benefit, which secured them 80 percent of their former earnings until the age of retirement. Releasing personnel via the disability scheme created a considerable pool of hidden unemployment which was accepted by workers and unions in exchange for the generosity of the scheme. This dramatically illustrates the passivity of the transferoriented welfare system. It established what perhaps disrespectfully can be called the benefit-machine of the Dutch welfare state. The Netherlands could afford generosity to its workers partly as a result of favourable economic conditions, but also because the nation reached a semi-OPEC status in the 1960s as a result of the discovery and profitable exploitation of natural gas resources in the northern province of Groningen (Lubbers and Lemckert 1980). Next to the factors already noted, two other conditions appear to have been critical for the expansion of the Dutch model. These conditions concern the unparalleled economic growth and prosperity of the Dutch economy in the 1960s and the initially gradual, but in the end quite sudden structural changes of society in the form of depillarization and deconfessionalization, eventually culminating in the progressive political and cultural conjuncture of the permissive society (Braun and Van Kersbergen 1986). One argument is that the growth of the welfare state can be seen as an effect of depillarization. Welfare state resources may have started to compensate for the loss of milieu support that the pillars had enjoyed before. However, this interpretation is difficult to sustain, simply because the period 1945–1967 was the heyday of pillarization and until roughly the mid-1960s there was nothing to compensate. It is, in fact, more plausible to turn around the causal arrow (as Bax 1990, 1995 does). Economic growth helped to provide the resources for welfare state development. Modern social policies involved a professionalization of services and meant more state involvement in income maintenance schemes. The broad social accord, moreover, had already introduced a bigger regulatory role of the state in pillarized industrial relations. According to Bax (1995: 12), all these developments made individual households less dependent on churches, unions, extended family and all kinds of significant social collectivities and thus contributed to the erosion of social control by pillars (…). De-pillarization and the associated individualization of Dutch society implied that the vertical lines of social cleavage withered away as did the consensus nature of Dutch social conflict management and that, consequently, horizontal lines of social and economic division came more in the open.

141

In other words, the welfare state freed people from their dependence on the pillarized welfare institutions and caused a decline of the influence of the (religious) ideologies on social, cultural and political life. This is a more plausible explanation, although there is one serious problem here too. The welfare state may have had this emancipating effect on individuals and thus helped to erode the pillars’ power, but at the same time it continued to deliver its services mainly through the pillarized institutions themselves, even long after these institutions had lost their religious character. I cannot develop this point in this chapter, but my hunch is that the correlation between welfare state development and depillarization is probably spurious and in any case much more complex than a simple linear relationship (in whatever direction) would suggest. The hypothesis proposed here for further research is that rapid economic development indeed spurred a new phase of modernization and brought along the correlates of demographic change, urbanization, individualization and secularization. These latter changes influenced both depillarization and the expansion of the welfare state. Depillarization then promoted the further expansion of the welfare state because the Confessional parties could no longer rely on the ‘automatic vote’ of their pillar members and electoral competition for the working class vote intensified. In addition, both economic growth and the social protection that the welfare state offered significantly increased human existential security and this, in turn, reduced the need for religion and thus for religiously based organizations, as Norris and Inglehart have (2004) demonstrated. So, however one looks at the causality between depillarization and welfare state expansion, the fact is that depillarization and deconfessionalization had a vast impact on Confessional politics, as exemplified by the radicalization of Anti-Revolutionary politics and the gradual social democratization of the Catholic labour movement in the 1970s. The Catholic party in particular was forced to adopt more radical social policy demands in order not to loose its labour wing (or the working class vote) altogether, which in the early 1970s facilitated the return of Social Democratic–Catholic co-operation that made a final effort for further expansion until the oil crisis and stagflation painfully made clear that the welfare state had reached its limits.

142

5.4

Conclusion

Why was social policy so late to emerge and why did it remain underdeveloped for such a long time? Why did the Dutch welfare state in the period after the World War II so suddenly and rapidly expand to become one of the most comprehensive, generous and passive of its sort? If anywhere, it was in the Netherlands that the mobilization of the socio-economic cleavage via Socialist or Social Democratic parties was suppressed by the pervasive presence of the religious cleavage and religious fragmentation. The electoral system of proportional representation translated and reinforced the cleavage structure in multiple political parties representing cleavage based social pillars. Under such conditions, orthodox Protestants – a social and political minority – could come to power in alternating coalitions. The orthodox minority, however, was uninterested if not openly reluctant in developing modern social policy. Liberal anti-interventionism lined up with the Protestants’ mistrust of the state and the still vulnerable Catholics and weak Socialists were unable to gather enough support in favour of more generous social policy. The explanation for the speedy expansion and the passivity of the welfare state after World War II in a sense is the mirror image of late development. Because Calvinist and Liberal politics were marginalized in the immediate period after the war, the way was paved for a pro-welfare Roman–Red coalition between Social Catholics and Social Democrats. Christian Democracy blocked the type of active (labour market) social policies that the Social Democrats favoured, but supported a passive transfer regime. In the 1960s and 1970s, depillarization intensified the competition over the working class vote, which drove up social spending. Under favourable economic conditions path dependent mechanisms of ‘automatic government’, increasing coverage and generosity and the passiveness of the transfer system contributed to the further expansion of the Dutch welfare state until, as elsewhere, the economic reality of stagflation put a hold on expansion and ushered in decades of austerity.

143

Literature Bank, J (1978), Opkomst en Ondergang van de Nederlandse Volks Beweging (NVB), Deventer, Kluwer. Bax, E.H. (1990), Modernization and Cleavage in Dutch Society. A Study of Long Term Economic and Social Change, Aldershot, Avebury. Bax, E.H. (1995), Cleavage in Dutch society: Changing patterns of social and economic discrimination, Paper presented at the Conference on Political, Economic and Social Racism, Thessaloniki, Macedonia, Greece, May 14–18, 1995. Blok, E. (1978), Loonarbeid van Vrouwen in Nederland 1945–1955, Nijmegen: SUN. Böhl, H. de Liagre, J. Nekkers and L. Slot (eds.) (1981), Nederland Industrialiseert! Politieke en Ideologische Strijd rondom het Naoorlogse Industrialisatiebeleid 1945–1955, Nijmegen, SUN. Bos, F. (2006), De Nederlandse Collectieve Uitgaven in Historisch Perspectief, Den Haag, Centraal Planbureau. Bosmans, J. (1982), ‘Het maatschappelijk-politieke leven in Nederland 1945–1980’, Algemene Geschiedenis der Nederlanden, 15, Nieuwste Tijd, Haarlem, Fibula-Van Dishoeck. Braun, D. (1988), Die Niederländische Weg in die Massenarbeitslosigkeit (1973–1981), Amsterdam, Doctoral Dissertation. Braun, D. and K. van Kersbergen (1986), ‘Wendepolitik und politische Kräfteverhältnisse in den Niederlanden’, Blätter für deutsche und internationale Politik, 31, 7. Bussemaker, J. and K. van Kersbergen (1994), ‘Gender and welfare states: some theoretical reflections’, in D. Sainsbury (ed.), Gendering Welfare States, London, Sage. Cox, R.H. (1993), The Development of the Dutch Welfare State. From Workers’ Insurance to Universal Entitlement, Pittsburgh, Univerity of Pittsburgh Press. Daalder, H. (1966), ‘The Netherlands: Opposition in a segmented society’, in R.A. Dahl (ed.), Political Opposition in Western Democracies, New Haven and London, Yale University Press. De Bruijn, J. (1989), Haar Werk. Vrouwenarbeid en Arbeidssociologie in Historisch en Emancipatorisch Perspectief, Amsterdam: SUA. De Haan, I. (2003), Het beginsel van leven en wasdom: de constitutie van de Nederlandse politiek in de negentiende eeuw, Amsterdam: Wereldbibliotheek. De Wolff, P. and W. Driehuis (1980), ‘A description of post war economic developments and economic policy in The Netherlands’, in R.T. Griffiths (ed.), The Economy and Politics of The Netherlands Since 1945, The Hague, Martinus Nijhoff. Esping-Andersen, G. (1996), ‘Welfare states without work: the impasse of labour shedding and familialism in continental European social policy’, in idem (ed.), Welfare States in Transition. National Adaptations in Global Economies, London, Sage. Fortuyn, W. S. P. (1980), Sociaal-economische Politiek in Nederland, Groningen, Doctoral Dissertation. Fortuyn, W. S. P. (1983), Kerncijfers 1945–1983 van de Sociaal-economische Ontwikkeling in Nederland. Expansie en Stagnatie, Deventer, Kluwer.

144

Gladdish, K. (1991), Governing from the Centre. Politics and Policy-Making in the Netherlands, London, Hurst/The Hague, SDU. Hemerijck, A. (1992), The Historical Contingencies of Dutch Corporatism, Oxford, Doctoral Dissertation. Hoekstra, H. (2005), Het hart van de natie. Morele verontwaardiging en politieke verandering in Nederland 1870–1919, Amsterdam: Wereldbibliotheek. Irving, R.E.M. (1979), The Christian Democratic Parties of Western Europe, London, Allen and Unwin. Katzenstein, Peter J. (1985), Small States in World Markets. Industrial Policy in Europe, Ithaca, N.Y., Cornell University Press. Kalyvas, S. (1996), The Rise of Christian Democracy in Europe, Ithaca and London, Cornell University Press. Kaufmann, F.-X. (1988), ‘Christentum und Wohlfahrtsstaat’, Zeitschrift für Sozialreform 34: 65–89. Klein, P.W. (1980), ‘The foundations of Dutch prosperity’, in R.T. Griffiths (ed.), The Economy and Politics of the Netherlands since 1945, The Hague, Nijhoff. Koole, R. and H. Daalder (2002), ‘The consociational democracy model and the Netherlands: Ambivalent allies?’, in J. Steiner and T. Ertman (eds) (2002), Consociationalism and Corporatism in Western Europe. Still the Politics of Accommodation? Special Issue of Acta Politica, 37 (Spring/Summer). Korpi, Walter and Joakim Palme (2003), ‘New Politics and Class Politics in the Context of Austerity and Globalization: Welfare State Regress in 18 Countries, 1975–95’, American Political Science Review, 97 (3): 425–446. Kossmann, E.H. (1978), The Low Countries, 1780–1940, Oxford, Oxford University Press. Kuyper, A. (1898), Het Calvinisme. Zes Stone-Lezingen in October 1898 te Princeton (N.J.) Gehouden, Kampen, Kok. Lademacher, H. (1993), Geschiedenis van Nederland, Utrecht: Het Spectrum. Lubbers, R. and C. Lemckert (1980), ‘The influence of natural gas on the Dutch economy’, in R.T. Griffiths (ed.), The Economy and Politics of the Netherlands Since 1945, The Hague, Martinus Nijhoff. Mackie, T.T. and Rose, R. (1991), The International Almanac of Electoral History, Houndmills and London, MacMillan. Manow, P. (2004), The ‘Good, the Bad and the Ugly’. Esping-Andersen’s Welfare State Typology and the Religious Roots of the Western Welfare State. MPIfG working paper 04/3, Max Planck Institut für Gesellschaftsforschung, Cologne. Norris, P. and Inglehart, R. (2004), Sacred and Secular: Religion and Politics Worldwide, Cambridge: Cambridge University Press. Pott-Buter, Hettie A. (1993), Facts and Fairy Tales about Female Labor, Family and Fertility: A Seven-Country Comparison, 1850–1990, Amsterdam, Amsterdam University Press. Roebroek, J. (1993), The Imprisoned State. The Paradoxical Relationship Between State and Society, Tilburg, Katholieke Universiteit Brabant.

145

Roebroek, J. and M. Hertogh (1998), ‘De Beschavende Invloed des Tijds’. Twee Eeuwen Sociale Politiek, Verzorgingsstaat en Sociale Zekerheid in Nederland, Den Haag: VUGA. Roes, J. (ed.) (1985), Katholieke Arbeidersbeweging. Studies over KAB en NKV in de Economische en Politieke Ontwikkeling van Nederland na 1945, Baarn, Ambo. Ter Heide, F.J. (1986), Ordening en Verdeling. Besluitvorming over Sociaal-economisch Beleid in Nederland, 1949–1958, Kampen, Kok Agora. Schama, Simon (1988), The Embarrassment Of Riches. An Interpretation Of Dutch Culture In The Golden Age, Berkeley, University of California Press. Swenson, Peter (1991), Bringing capital back in, or social democracy reconsidered: Employer power, cross-class alliances, and centralization of industrial relations in Denmark and Sweden, World politics, 43 (4): 513–544 Therborn, G. (1986), Why Some Peoples Are More Unemployed Than Others, London, Verso. Therborn, G. (1989) ‘“Pillarization” and “popular movements”‘, in F.G. Castles (ed.), The Comparative History of Public Policy, Cambridge, Polity Press. van der Kolk, Henk and Jacques Thomassen (2006), ‘The Dutch Electoral System on Trial’, Acta Politica, 41 (2): 117–132. Van Eijk, C.J. (1980), ‘Counter-inflation policy in the Netherlands’, in R.T. Griffiths (ed.), The Economy and Politics of the Netherlands Since 1945, The Hague, Martinus Nijhoff. Van Kersbergen, K. (1995), Social Capitalism. A Study of Christian Democracy and the Welfare State, London, Routledge. Van Kersbergen, K. (1997), ‘Christian Democracy in the Netherlands and its influence on economic and social policy, in E. Lamberts (ed.), Christian Democracy in the European Union, 1945–1995, Leuven, Leuven University Press. Van Kersbergen, K. and U. Becker (2003 [1988]), ‘The Netherlands: A passive social democratic welfare state in a Christian democratic ruled society’, in J.T.S. Madely (ed.), Religion and Politics, Vol. 15 of D. Arter (general editor), The International Library of Politics and Comparative Government, Dartmouth, Ashgate (originally published in Journal of Social Policy, Vol 17, 1988). Van Kersbergen, K., P. Lucardie, and H.-M. ten Napel (eds) (1993), Geloven in Macht. De Christen-Democratie in Nederland, Amsterdam: Het Spinhuis. Van Kerbergen, K. and A. Hemerijck (2004), ‘Christian democracy, social democracy and the continental ‘welfare without work’ syndrome’, Social Policy Review, 16: 167–186. Van Oirschot, F. (1950), Beknopte geschiedenis der sociale kwestie. Ontstaan en oplossingen, Roermond en Maaseik: J.J. Romen & Zonen. Van Putten, J. (1985), Politieke Stromingen, Utrecht and Antwerpen, Spectrum. Van Rooden, Peter (1999), ‘History, the nation and religion: The transformations of the Dutch religious past, in Peter van de Veer and Hartmut Lehmann (eds), Nation and Religion. Perspectives on Europe and Asia, Princeton, NJ: Princeton University Press, 96–111. Van Waarden, F. (2002), ‘Dutch consociationalism and corporatism. A case of institutional persistence’, in J. Steiner and T. Ertman (eds) (2002), Consociationalism and Corporatism in Western Europe. Still the Politics of Accommodation? Special Issue of Acta Politica, 37 (Spring/Summer).

146

Van Zanden, J.L. and R.T. Griffiths (1989), Economische geschiedenis van Nederland in de 20e eeuw, Utrecht: Het Spectrum. Von der Dunk, H. (1982), ‘Conservatism in the Netherlands’, in Z. Layton-Henry, Conservative Politics in Western Europe, London and Basingstoke, MacMillan. Windmuller, J.P. (1969), Labor Relations in the Netherlands, Ithaca, Cornell University Press. Woldring, H.E.S. and Kuiper, D.Th. (1980), Reformatorische Maatschappijkritiek. Ontwikkelingen op het Gebied van Sociale Filosofie en Sociologie in de Kring van het Nederlandse Protestantisme van de 19e Eeuw tot Heden, Kampen, Kok. Wolffram, D.J. (2003), Vrij van wat neerdukt en beklemt. Staat, gemeenschap, sociale politiek, 1870–1918, Amsterdam: Wereldbibliotheek. Wolinetz, S.B. (1983), ‘The Netherlands’, in V.E. MacHale and S. Skowronski (eds) Political Parties of Europe, Westport, Greenwood Press. Woltjer, J.J. (1992), Recent verleden. De geschiedenis van Nederland in de twintigste eeuw, Amsterdam: Balans. Zwart, R. (1993) ‘Ideologie en macht. De christelijke partijen en de vorming van het CDA’, in K. van Kersbergen, P. Lucardie en H.-M. ten Napel (1993), Geloven in Macht. De ChristenDemocratie in Nederland, Amsterdam, Het Spinhuis.

147

6

A Conservative Welfare State Regime without Christian Democracy? The French Etat-providence, 1880–1960*

Philip Manow and Bruno Palier

6.1

Introduction

The French welfare state resembles in many respects the Bismarckian blueprint of a conservative regime type: wage deducted contributions and wage related benefits, a fragmentation of social insurance schemes along occupational lines, and the predominance of funds (caisses) managed according to principles of self-administration (autogestion) with the prominent participation of unions. The French Etat-providence also shares many of the problems that plague continental welfare state today: runaway costs, adverse labour market effects due to high non-wage labor costs leading to poor job growth especially in low productivity services, and low employment ratios as a result of generous early retirement schemes. And like its continental counterparts, the French welfare states is also often perceived as being ‘immovable’ and ‘frozen’ as it largely resists any political attempts at reform (but see Palier/Bonoli 2001). In the comparative literature the French welfare state has usually been classified as a conservative, continental, Catholic, or Bismarckian regime (Esping-Andersen 1990; Huber, Ragin et al. 1993; Huber and Stephens 2001; Scharpf and Schmidt 2001; Palier 2004). However, it sits somewhat uneasily with this classification. In fact, the French Etat-providence is not particularly conservative, patriarchic or Catholic when it comes to family policy (Morgan 2002, 2003, 2004, this volume; Talmy 1962) and its strong emphasis on universal social assistance has a much more ‘Beveridgian’ than ‘Bismarckian’ flavor to it (Levy 2000: 312–13). Some scholars even see the French welfare state as belonging to a distinct southern European regime type rather than to the family of the continental-corporatist regimes, a view supported by the long-time absence of an unemployment insurance and the dominance of pension payments in public social spending – features shared by most of the other southern welfare states (Ferrera 1996; Ferrera 1997).

* We are grateful for very helpful comments by Patrick Hassenteufel, Kees van Kersbergen and Kimberly Morgan.

148

In the context of the theoretical framework proposed in this volume (van Kersbergen and Manow, this volume) the French case provides us with a critical case study, since it confronts us with at least three, possibly interrelated puzzles. The first puzzle is that the state-church conflict in France, by all measures, has been particularly pronounced in the last decades of the 19th and the first decades of the 20th century. Yet, in contrast to most other continental countries, this conflict did not trigger the formation of a party of religious defense during the Third Republic, between 1871 and 1940 (cf. Kalyvas 1996). When a Christian Democratic party finally appeared on the scene after World War II, it failed to establish itself successfully in the French party system. After initial electoral success, the Mouvement Républicain Populaire (MRP) performed rather poorly in the second half of the 1950s and finally dissolved in 1967. In contrast to most other countries on Europe’s continent, the religious cleavage in France failed to achieve long-term ‘particization’ (Stoll 2005). France seems to be an outlier also in a second respect: While all other countries on Europe’s continent (plus Scandinavia) had introduced proportional representation by 1919 at the latest, France throughout the Third Republic stuck to the old two round/two ballot system (Caramani 2000). PR was introduced as late as 1945, only to be abolished again soon with the advent of the Fifth Republic in 1958. With the new constitution, France returned to the old electoral system, now in combination with a more presidentialist political system. Given that this volume ascribes critical importance to electoral rules for the party-political representation of societal cleavages, this chapter discusses the impact that the peculiar French electoral rules had on the translation of the religious cleavage into politics. Thirdly and finally, although France differs in these two very important dimensions from most of the other European countries it still has a welfare state that resembles the conservative-continental regimes of its European neighbors in many respects. France therefore represents a particular challenge to an explanation which claims that it had been the interaction of proportional representation with a virulent secular/confessional cleavage which explains the strength of Christian Democratic parties in continental European party systems, parties which then played a critical role in the build up of the continental conservative welfare state regime (van Kersbergen and Manow, this volume). To preview the response to the three puzzles which we will give in the following sections: we show that indeed the secular/confessional cleavage was not only present

149

during the period of early welfare state formation around the turn of the century, but that it was then clearly dominating the capital/labour cleavage in late industrializing France. State/church conflicts left their imprint on early welfare state building, which did not yet address the new risks of an industrializing society, but focused more on the social responsibility of the French state towards the needy and destitute. Not social insurance, but social assistance and public education was what preoccupied the liberal reformers and their Catholic adversaries during the early phase of French welfare state building. This explains much of the state’s strong role and the high degree of universalism in both these domains. However, whereas the secular confessional cleavage proved to be particularly strong in late 19th and early 20th century France, the fierce state/church conflicts over education and social policy did not lead to the formation of a Catholic party. The quasi-majoritarian features of the two round/two ballot system go a long way in explaining this non-formation of a Christian Democratic party in France during the Third Republic. As we argue in Section 3, the lack of such a party of religious defense had long-term consequences for the Catholic milieu. Catholic unions, farmers’ associations, youth organizations and other groups from the Catholic ‘Lager’, which were mainly established in the interwar years, lacked a party-political addressee until 1945. They therefore sought independence and autonomy. This proved consequential for the survival of a French Christian Democratic party once such a party finally made its appearance after World War II. Catholic organizations as well as the church hierarchy remained highly reluctant to rally themselves behind the MRP. The church turned away from the MRP because it’s coalition with the strongly anticlerical socialists and Radicals forced the party to compromise on issues that Catholics particularly valued. Moreover, already in 1947 the Gaullist movement had turned into an anti-system party fighting for the fundamental overhaul of the Fourth Republic’s constitution and in many respects de Gaulle appeared to be better able to protect Catholic interests than the MRP. The post-war situation resembled the situation in the 1860s and 1870s, because with the stability of the Fourth Republic in question, organized Catholicism could not bring itself to back the MRP within the republic’s constitutional framework. The Fifth Republic then returned to the old electoral rules, which – in combination with the presidential elements in the constitution – proved to be particularly hostile to the party-political representation of the religious cleavage. Both, the lack of

150

interest group support and the return to the two round/two ballot system then contributed to the weakness of the MRP and its final disappearance from the party landscape in 1967. However, as we also show in Section 3, for the short but critical period of post-war welfare state formation the MRP did exert a very significant and long lasting policy influence. Many of the Bismarckian features of the French welfare state like its high levels of spending, the relatively high degree of organizational fragmentation, the strong role of self-administration, the role of the Christian union in it and the particular emphasis put on family policy clearly have to be attributed to the strong influence that the MRP had during this crucial period of welfare state building. And even when the MRP finally disappeared from the French political landscape, social Catholicism remained influential for French social politics. This chapter is organized as follows: in Section 2 we discuss the role of religious conflicts in the early phase of social legislation in the late 19th century. In Section 3 we describe the immediate post-war years, when the path was set along which the French welfare state would develop in the trente glorieuse or golden years of welfare state growth and expansion that followed. We discuss the role of electoral rules for the late arrival and premature vanishing of Christian Democracy in post-war France and analyze the nexus between the Christian Democratic MRP and the Catholic milieu.

6.2

Consolidating the Nation through the ‘Laïcisation de la Protection Sociale’1

A strong secular–confessional divide was not just present in the French political system of the Third Republic, but it was the dominant political cleavage: “the statechurch conflict developed into the most salient cleavage and the key ideological issue of the Third Republic. It dominated politics at all levels, from parliament to the last village, and overshadowed every other political and social question” (Berger 1987: 113; Kalyvas 1996: 115). It is therefore no surprise that the debates surrounding the early social policy interventions of the French state were dominated by religious concerns. The state–church cleavage coloured the conflicts about schooling (Lanfrey

1

(Renard 1995: 15).

151

2003), child and family policy (Talmy 1962; Pedersen 1993; Morgan 2002, 2003, 2004) and social assistance (Renard 1995: 15–19; Palier 2002: 66–67). The labour question, as one invoking the familiar ‘left’ and ‘right’ camps in the dispute over social legislation, only emerged later in the French politics of welfare (Korpi 1995: 638). Workers were not the prime target of early social legislation in France, but rather mothers, children plus the needy and destitute (Pedersen 1993; cf. Skocpol 1992). Compared to the political trajectories in many other European countries, where early social legislation was intended to break with the national poor law tradition, the French development is peculiar in that the poor law legislation itself marked the beginning of the modern French welfare state (Merrien 1990: 166). The main objective of this legislation was to put public poor relief and social assistance in place where formerly private (charitable) schemes dominated. With this the republicans continued a process of the “laïcisation de l’action sociale” (Merrien 1990: 164) which had set on already in the Second Empire. It is only later, notably with the post-World War II French plan of Sécurité sociale that the principle of insurance finally gained dominance over the principle of assistance (see Palier 2002: 66–71). Important social assistance legislation, such as the law on maltreated or morally abandoned children (1889), the law on medical assistance (1893), the law on assistance for children (1904), and the law on the old-aged, the infirm and incurables (1905), was introduced when the separation between state and church was hotly debated. This debate culminated in the famous Loi 1905, which fully disestablished the French Catholic church. The principle of direct intervention and state responsibility for the ‘valid’ infirm, destitute or old aged ‘which find themselves temporarily or definitely incapable to provide themselves with the necessities of life’ rested on a notion of state responsibility and public obligation to care for the needy. This notion referred to the republican principles of the French Revolution, which seemed particularly fitting around the 100th anniversary of the Revolution in 1889. Among these the principle of laïcism was particularly emphasized (cf. Weiss 1983). The school question had been on the political agenda since 1879 when the republican governments under Freycinet, Duclerc and Gambetta with a series of wide-reaching reforms established the dominance of state schools in primary and secondary education and tried to exclude the church entirely from the education sector. Since 1893 a loose alliance between moderate Republicans, Radicals, Independents and Socialists supported the

152

various governments and in 1902 this group of parties transformed into a more formal Bloc de Gauche. Anticlericalism was the “only policy a group as diverse as the Bloc could agree on” (Luebbert 1991: 30–31) and once the full disestablishment of the French Church was achieved in 1905, the Bloc dissolved (cf. Mayeur 1984: 137–208). The paramount goal of the liberal state elite was to consolidate the nation after France had lost the war against Germany and after the internal turmoil with the violent experience of the Parisian Commune. One way to achieve a French nation was to fight particularistic, in particular religious identities. This is strikingly parallel to Bismarck’s attempts to forge a unified nation after Germany finally had achieved territorial consolidation in the French–German war of 1870/71. Here the fight against Socialists and Catholics, the so-called enemies of the Empire (Reichsfeinde), was a central element in the nation building enterprise. In fact, the attack of French liberals on Church strongholds in education and charity in the name of the new nation state had parallels in many other European countries at around the same period (see Ertman, this volume). Establishing direct state responsibility for the infirm, orphans and the old aged openly and deliberately challenged the dominance of the Catholic Church in caring for the most destitute. It is important to note that notions of subsidiarity as well as notions of self-responsibility, individual thrift or discipline were largely absent in the debates. Instead doctrines of ‘interventionisme’ and ‘solidarisme’ were dominant, reformist schools of thought comparable to those of the German Kathedersozialisten or the British New Liberals. These schools of thought were closely allied to the Radical Party (cf. Stone 1985; cf. Luebbert 1991: 33). Reformers were convinced that only compulsory legislation could alleviate the problems of workers and the poor. They clearly preferred state-sponsored over voluntary and charitable solutions. In order to establish the authority of the state in this domain the various existing assistance schemes came under the central administrative responsibility of the Ministry of the Interior under the general title of assistance publique, under which also the responsibility for public health was soon subsumed (Renard 1994). Where direct state responsibility could not be established, strict control and oversight over the charitable bureaux de bienfaisance and the church-run hospitals were installed (Renard 1995: 16; Ashford 1986: 135–137). Step by step, the pre-eminence of public action over

153

any private forms of welfare provision was established (Renard 1987; Archambault 1997: 18–19). The social assistance legislation supported also the creeping expropriation of local charities through the municipalities, so that the new public activities in the realm of social assistance and poor relief became a process in which previously privately provided help and assistance were converted into public help and assistance (Renard 1995: 16 and 25, endnote 18). The double purpose was to establish a central state system of welfare provision and to crowd out the church in this domain. This was also done by recruiting only devoted republicans and certainly not Catholics as fonctionnaires to run the new public assistance administration (Renard 1995: 16). The fierce conflicts over the responsibility for primary and secondary education first preceded and then went together with the attempts of the French state to replace Catholic charity with social assistance schemes that were financed and administered by the state. The liberal governments that came to power after 1879 immediately engaged in an intense legislative activity in the educational sector under the triple principle of free, compulsory and public, meaning laïcist education (Lamfrey 2003; Archambault 1997: 35–37). Jules Ferry as minister of public instruction (1879–1883) initiated a series of reforms, which aimed at transforming the school into a republican institution designed to transgress the internal strife and the ideological and denominational divisions of the French Third Republic. Starting with the prohibition of Jesuit schools (Décret du 29 mars 1880), the Loi du 16 juin 1881 established free access to primary education and the Loi du 28 mars 1882 rendered primary education obligatory for children of both sexes between six and thirteen. The French state forced departments to establish schools for teachers so as to break the monopoly of the church with respect to the training of school personnel. After the abolishment of the retribution scolaire in the early 1880s, school teachers became civil servants directly paid by the public treasury in 1887. With a clear anti-clerical impetus, the educational laws aimed to separate state and church in primary, secondary and higher education. Already in 1880 the government had replaced all clerics in the central advisory bodies like the Conseil supérieur de l’Instruction publique or the Conseils académiques with university professors. The 1901 law on associations required state approval of associations, and approval was rarely granted (Archambault 1997: 37, 76). The ultimate aim was to exclude the church from the educational sector, an aim finally achieved in 1904 when the Loi du 7 juli forbid congregations any educational activity “of any sort and na-

154

ture”. Subsequently more than 2,500 Catholic schools had to be closed down, and more than 11,000 demands by congregations for state authorization of hospitals or schools were refused. This was one year before the famous Loi 1905 led to the full separation of état and église in education, social policy and any other area (Lamfrey 2003). As a response to this anti-clerical attack Catholics unleashed a veritable guérrilla scolaire that lasted until the First World War. The church and its followers fought with much intransigence for the preservation and public subsidization of private, Catholic schools, but also against a stronger role of the state in social policy. In the church’s view a state playing providence not only undermined individual thrift and prudence, but also threatened to destroy the smallest and holiest unit of social solidarity, the family (Dumons/Pollet 1994: 103–118). While the Catholics vehemently condemned the devastations of ‘excessive economic liberalism’, they at the same fought against any state intervention and refused to give their consent to social protection legislation, since this in their eyes would have meant to participate in the further secularization of French society (Dumons/Pollet 1994: 112). In many respects this conflict constellation between a liberal reform elite eager to consolidate the nation on the one hand, and an ‘intransigent Catholicism’ fiercely fighting against any attempts to exclude the church from education and charity on the other (Dreyfus 1988: 57–61), resembled the conflicts in Italy, Germany, Belgium, Switzerland and the Netherlands at around the same time. Further below we will discuss why, unlike in these latter countries, the strong state/church conflict in France has not led to the formation of a party of religious defence (Kalyvas 1996). While the French governments since the mid-1880s were preoccupied with questions of social assistance and public education, the worker question and the introduction of social insurance programs only later came on the political agenda. This of course reflects the fact that industrialization was simply far less advanced in France than in most of its neighbours (Maddison 1992). In 1911, still more than 56 percent of the population lived in the countryside (Hatzfeld 1988: 284), and still in 1931 agricultural employment with 35.6 percent was larger than industrial employment with 33.7 percent (Mitchell 1988; ILO, various years). The Loi sur les retraites des ouvriers et paysans (ROP, 1910) set the path for later social legislation and prefigured in many respects the institutional setup of future social protection schemes, even though the law itself

155

remained an unsuccessful attempt to respond to the new risks of an industrializing society. Unions, the mutuelle movement, employers, but also Catholics rejected the reform. Unions demanded state financing, mutuelle leaders castigated the compulsory nature of the social insurance, employers complained about the extra-labour costs that the law imposed on them (Merrien 1990: 191–192) and Catholics insisted on the principle of non-intervention and the prime responsibility of church, family and the individual in addressing poverty and economic insecurity (Dumons/Pollet 1994: 103–118). The social insurance scheme that the ROP introduced remained half-way between state-subsidization of voluntary schemes and a fully compulsory social insurance system. The organizing principles of the ROP (Korpi 1995: 639–640) were the following: −

wage-deducted contributions,



different classes of contributions according to income,



entitlement rights conditioned on a record of contributing to the scheme,



administration in the hands of funds or mutuelles (organized either along occupational, syndical, or local principles, complemented with company schemes)



compulsory membership for certain wage groups, optional membership for dependently employed whose income was above the obligatory ceiling (of 3.000 Franc per year),2



self-administration based on the joint and equal financing of the social insurance by employers and employees with tripartite representation of insured, employers and state in the ‘comités de gestion’ (see Korpi 1995 for more details).

These features conformed to those that had been first introduced in an encompassing way with the German social insurance legislation in the 1880s. Contemporaneous to the many social assistance laws, which defined the various groups of needy citizens for which the French state was going to assume direct social responsibility, the state in 1898 also acknowledged the freedom of foundation and membership of the voluntary organizations of self-help, the mutuelles (Rosanvallon

2 Self-employed farmers and tenants, handicraft workers and small shop owners could become member of the insurance on a voluntary basis.

156

1990: 176; Archambault 1997: 34). These themselve had in the meantime become less “dominated by local notables and clerics” and more “republican, laïcist, less hierarchic and more democratic” (Saint-Jours/Dreyfus/Durand 1990: 66, our translation; cf. Dutton 2002: 41). Because the French state for a long time had left social insurance to the voluntary initiative, once it became active in the field it had to face the existence of various caisses and mutuelles that already occupied the terrain. Later legislative initiatives therefore attempted to regulate and expand rather than substitute for the existing infrastructure of caisses, company funds and other organizations of voluntary self-help. The high degree of organizational fragmentation around this period of time was therefore not the reflection of a corporatist ideology of either a Catholic or any other conservative-authoritarian persuasion (cf. Esping-Andersen 1990: 31–32). Growth in membership of the mutuelles was pronounced, with 1.3 million members in 1890, 2 million in 1898 and already 5.4 million members in 1913 (Rosanvallon 1990: 181; Dutton 2002: 42, Figure 4). In 1908, the mutuelles were already paying out around 120,000 to 130,000 pensions (Hatzfeld 1989: 205, fn. 22). However, workers did not make up the major part of mutuelle membership. Only for the best paid workers with highest job security membership was a realistic option. These statesponsored forms of voluntary protection rather attracted small employers, shopkeepers and artisans (Hatzfeld 1989: 211, 213; Dutton 2002: 38). It took until the 1920s and 1930s before social legislation addressed the needs and interests of workers more specifically. Various attempts at subsequent social legislation in the 1920s failed (like the Loi sur l’assurance sociale in 1921 and the initiative to introduce unemployment insurance in 1925) but the Poincaré government in 1928 finally, after long parliamentary battles, succeeded in enacting a law on social insurance. It covered the risks of sickness, old age, invalidity, and maternity (as well as funeral costs). The incoming left government amended the Poincaré legislation, but the law still fell far short of the kind of universal and uniform social protection system that initially had been intended (Merrien 1990: 209–211). In 1930 the amendment established two major social insurance schemes, the régime général, which covered the workers in industry and trade, and another regime (régime agricole) covering workers in agriculture. Institutionally, these laws followed the blueprint of the 1910 Loi sur les retraites, which led to a high degree

157

of administrative fragmentation. The separately organized caisses d’allocations familiales (1932) only added to this already high level of fragmentation and complexity. The caisses d’allocations represent a French peculiarity, as the strong role of family policy in the French welfare state more generally does (Lenoir 1991; 1991a; Talmy 1962, cf. Morgan 2003, 2006). Much of this has been rightly attributed to French pronatalisme, but natalisme in many respects was only the republican response to Catholics’ familialism. A fertility rate below the reproduction rate and the resulting labor shortages in the interwar period lend plausibility and urgency to the Catholics’ claim that unfavourable demographics reflected the nefarious influence which modern individualism, libertinage and the loss of traditional values had achieved on French society (Lenoir 1991; Archambault 1997: 38). With natalisme republicans were then developing a kind of surrogate civic religion. Catholics and anti-clerical Republicans could agree on generous support for families since strong economic interests backed their coalition. The French Catholic patronat had developed a manifest economic interest in the caisses familiales. Before these funds had been made obligatory for all firms by law in 1932, family benefits as wage supplements had been championed by a Catholic patronat on a voluntary basis. What mattered most for these employers was the stabilization of their workforce. But employers valued family wages also because they were an instrument to divide workers, to weaken unions, which always had fought for the ‘equal work, equal wage’- principle, and to avoid general wage increases by granting certain subgroups a higher payment (Petersen 1993; cf. Mares 2003). In 1932 the régime général comprised 78 funds, the agrarian regime 52 funds, and the sickness insurance had 793 funds within the general regime and another 285 funds for the agricultural workers (Korpi 1995: 642). The legislation of the early 1930s built upon the existing landscape of insurance funds, and it therefore is appropriate to characterize social insurance à la francaise as a sort of compromise between the German regime of compulsory social insurance and “mutuellisme à l’anglaise” (Rosanvallon 1990: 181), mutuellisme à l’anglaise here designating the pre-Beveridgian period of the British welfare state with its strong role of friendly societies. This was the compromise from which the social reformers in the immediate post-war years had to start (see Section 3).

158

In the interwar years the labour question came to the forefront of French social policies. At the same time the tensions between the Catholic milieu and the republic had become less marked (Merrien 1990; Debreil 2001). A moderate social Catholicism more and more replaced the instransigent Catholicism that had dominated before the war. During the 1920s the French catholic camp established a rich network of catholic unions, associations, charitable organizations and leisure time groups (Nord 2003). The foundation of the Confédération Francaise des Travailleurs chrétienne (CFTC, 1919), the Fédération national catholique (FNC in 1924), the Jeunesse ouvrière chrétienne (JOC, 1927), the Jeunesse agricole chrétienne (JAC, 1929), the Jeunesse étudiante chrétienne (JEC, 1929), the Ligue ouvrière chrétienne (LOC) and the Mouvement familial rural (MFR) together with many other Catholic associations, newspapers, journals, radio stations and publishing houses established a dense network of Catholic organizations (cf. Nord 2003; Letamendia 1995: 39–43). In this respect the French interwar development was similar to that in other continental countries with a significant Catholic population. Finally, with the Parti démocratie populaire (PDP, 1924), also a Catholic party saw the light of the day. However, in the three parliamentary elections in which it took part (1928, 1932 and 1936), the party never gained more than 3 % of the votes. The PDP remained marginal and compared to the Italian Partito Populare or the German Zentrum of the interwar years strikingly unsuccessful (Delbreil 2001; cf. Warner 1998). If we consider that the Fédération national catholique alone had more than 2 million members in the late 1920s, it is striking that the Parti Democratie Populaire had no more than 395.000 voters in 1932. Obviously, the party massively failed in fully mobilizing the Catholic vote. This failure has much to do with the quasimajoritarian effects of the two round/two ballot electoral system which was in place in interwar France. The Italian Partito Populare and the German Zentrum, in contrast, had prospered under proportional representation. The two rounds/two ballots electoral system leads to a high effective number of parties (like a proportional representation [PR] system), but these parties then ultimately form only two party political camps, a left and a right one (like a majoritarian system). It is the first round which is responsible for a relatively high effective number of parties, since each party is eager to enter the electoral race with its own candidate. The second round via le vote utile then provides parties with incentives to coordinate

159

and cooperate and to form two major camps along the dominant left–right dimension. This does not allow for the representation of any additional, in particular of no religious cleavage, especially since it was often anticlericalism that glued parties together in the second-round coalitions. This has been the French electoral system’s characteristic effect in the Third and then again in the Fifth Republic: the formation of two blocks, a left and a right one, that together dominate politics (Elgie 2005: 126– 127). Parties themselves remained weak under these electoral rules, because they promote a candidate-centred rather than party-centred party system (Carey and Shugart 1995), or in the terminology of the French debate: a ‘rivalité des personnes’ rather than a ‘scrutin d’idées’ (Bernstein 2001: 779), electoral rules favoured local “candidats individuels” over “la mobilisations collectives” (Rousselier 2001: 906). It was also the very small district size (on the arrondissement-level) of the electoral system in the Third Republic that reinforced the local character of interest representation and further discriminated against the Catholic vote, concentrated as it was geographically in the North and the West of the country (cf. Rodden 2005). As we show in the following section, the absence of a Catholic party during the long lasting Third Republic, when industrialization had brought the ‘labor question’ to the forefront of French politics, critically affected the chances of it’s survival once such a Christian democratic party appeared on the scene after World War II.

6.3

Christian Democracy and the French Post-War Welfare State

As in so many other countries on Europe’s continent, the French welfare state was supposed to be a means of social integration. While it had been targeted at the potentially dangerous classes in the late 19th century, establishing state responsibility for destitute or in other ways marginalized French citizens, after the Second World War it became directed at the integration of the Socialist/Communist and Catholic subcultures. The experience of the resistance, in which Communists, Socialists, but also the Christian-democratic Mouvement Républicain Populaire3 had taken part, explains the broad coalition supporting wide reaching social reforms after the Second World War. Until today, Securité sociale is understood as being one of the most important

3

Labelled Mouvement Républicain de Libération until November 1944.

160

achievements of the Libération. Indeed, the French exile government in London under General de Gaulle engaged as early as 1942 in elaborating extensive post-war plans for a thorough reform of the French social protection system (Dutton 2002: 202– 208; Baldwin 1990: 163–169). These plans were oriented towards the principles of universality, obligation and democracy (meaning self-administration) and these three principles had been heavily influenced by the British Beveridge-report of 1942 and by the ensuing debate. The French emigrants had closely followed the British debate since in so many respects the Beveridge report addressed a task that seemed similar to the one that was waiting for them at home: forming something like a national system of social protection out of a highly fragmented arrangement with a strong role of voluntary self-help organizations like friendly societies or mutuelles. However, the French national system of protection finally came to resemble much more a Bismarckian than a Beveridgian welfare state (Ambler 1991: 12–13; Palier 2002: 72–73). Securité sociale remained, in essence, a contribution financed, highly fragmented social insurance for the dependently employed. Given that the literature makes standard reference to the role of Christian Democracy in the setup of the continental welfare regime when explaining these institutional features, we must ask what the role of the Christian Democratic MRP has been in this development. With the MRP the French party system finally got the strong Catholic party it had lacked for so long (see Letamendia 1995; Béthouart 2001). In contrast to its feeble interwar predecessor, the PDP, the MRP was an electorally successful party, at least initially, in the early years of the Fourth Republic. In the first elections for the constitutional assembly in October 1945 the Mouvement gained 23.9 percent of the vote and became the second largest party behind the communist PCF (with 26.2 percent of the vote). In the June 1946 elections the Christian Democrats even surpassed the Communists and became the strongest party with 28.2 percent. Communists and Christian Democrats then again changed rank in the election for the first parliamentary assembly in November of that same year, but the MRP with a vote share of 25.9 percent still proved to be highly successful at the ballot box. With a vote share between 26 and 28 percent the MRP fell only little short of the electoral support that the German CDU (31 %) or the Italian Democrazia Cristiana (35 %) could muster in the first post-war elections. It looked as if political Catholicism finally had arrived and would remain strong in the French party system. Moreover, as a pivot party located be-

161

tween the conservatives, the radical party and later the Gaullist movement on the right and the Communists (PCF) and the Socialists (SFIO) on the left, the MRP took part in almost every government of the Fourth Republic, the only exceptions being the Blum government (November 1946 to January 1947), the Mendes-France government (June 1954 to February 1955) and the Mollet and Bourges-Maunory governments (from February 1956 to September 1957; cf. Dreyfus 1988: 230; Hanley 2002: 130–132). The MRP was the party of the Fourth Republic, as the party identified itself with the republic and was identified with it. The early election success of the MRP certainly reflected the reputation it had achieved as a participant in the Résistance. Yet, its high esteem in post-war France, its early success at the ballot box and its almost uninterrupted government participation during the Fourth Republic could not prevent the party’s steady decline and it’s ultimate failure to establish itself in the French political system (Letamendia 1995; Warner 1998). The MPR dissolved in 1967, but already in the 1950s its electoral record had become much less impressive. With the demise of the MRP, “the confessional– secular cleavage ha[d] disappeared” from the party system (Elgie 2005: 127), while denominational affiliation continued to influence strongly voting behaviour in the Fifth Republic (Michelat/Simon 1977). How can this be explained? Pointing again to the French electoral system seems insufficient since closed-list PR was the electoral mode in place between 1946 and 1958, and this period saw both the MRP’s initial electoral success and its subsequent decline, from strongest to merely fifth strongest party in only five years between 1946 and 1951 (Warner 1998: 559–561). Clearly, after 1958 the MRP did suffer from the return the two round/two ballot system, because of the Catholic vote’s regional concentration and the frequent anticlerical alliances in the second round. It is not surprising that in the debates on electoral reform between 1956 and 1958 the MRP strongly favoured to stick with PR. The MRP preferred PR with only one national district, whereas the Parti Radicale, but also the Independents and Socialists, which successfully had collected the local vote, were ‘sympathetic to a restoration of majority rule which had benefited them historically so greatly’ (Alexander 2004: 210; Bernstein 2001: 781). To be clear, like all stories of political failure the decline of the MRP is to some extent a story of chance events and misfortune, of personal idiosyncrasies, lack of charisma

162

and grave strategic errors (cf. Warner 1998; Letamendia 1995). These factors do not easily lend themselves to systematic analysis. But once we look at the relation (or lack thereof) between the MRP and the existent Catholic milieu, one systematic difference between the French case and the other continental countries becomes apparent: the lack of strong and stable ties between the party on the one hand and the Catholic associational landscape and church hierarchy on the other (Pasture 2001: 243; Delbreil 2001).4 With the high party fragmentation and the high degree of government instability, French interest groups were well advised to look after themselves rather than to seek the protection of their interests in a close attachment to any party. This was particularly true for the Catholic camp, given the long-term absence of a Christian democratic party from the French party system. From the perspective of unions or other interest organizations, ‘going it alone’ therefore seemed to be more promising. In the case of the unions this seemed to correspond to their tradition of syndicalisme anyway. It was therefore not entirely unexpected that the Confédération Francaise des Travailleurs chrétienne, which would have been the MRP’s natural support group, was eager to demonstrate its autonomy and independence from the party at each and every occasion. Already in 1946 the CFTC made it clear that it wanted to remain fully separate from the MRP. The Christian union, second in membership among the French unions, declared its ‘total independence’ from the MRP and finally even declared that holding a position in the MRP would be incompatible with holding a position in the union (Berger 1987: 120; Groux/Moriaux 1991: 7; Pasture 2001: 236, 243). Since its foundation in 1919 the CFTC had learned to survive without close partisan links. A close organizational nexus to the interwar PDP never materialized since this party felt at least as close to the Catholic patronat as to the workers (Delbreil 2001: 80), and the PDP was a marginal party anyway. While the CFTC stood for harmonious relations between capital and labour, a too close attachment to employers had to be avoided. The CFTC’s declaration of independence in 1946 certainly was a reaction to the precarious legitimacy of a Catholic party in the French party system, but also a

4 Carolyn Warner in her comparison between DC and MRP finds a strong and stable effect of membership in Italian Catholic Action groups on the electoral fate of the DC, whereas such an effect is much less evident in the French case (Warner 1998: 563, Table 1).

163

manifestation of the fear that a too close attachment to the MRP would restrict union mobilization exclusively to Catholic workers. A strong current within the CFTC saw in the MRP the danger of a re-confessionalization of politics, while it understood the Christian Union as a mainly ‘secular’ union that had accepted basic tenets of the republic (among them public schooling5; Pasture 2001: 236; Duriez 2001: 65–66). To make this secular status visible, the union dropped the reference to Christianity in its name and renamed itself into Confédération Française Démocratique du Travail (CFDT) in 1964. Apparently, the MRP’s dilemma was that it was too clerical for the unions, and too secular for the church hierarchy, since it “gestured only obliquely to private schools, family values and French ‘spirituality’” (Bell 2000: 103; see below). That stable links between the MRP and the CFTC could not be established (van Kemseke 1997: 178, 180) certainly contributed to the weakness of the MRP – the deconfessionalization of the CFTC has even been made responsible for the “ruin of the Christian Democratic project in France and the dissolution of the MRP in 1967” (Pasture 2001: 243). The weak implantation within the Catholic milieu had already been noted for the PDP (Delbreil 2001: 80; Letamendia 1995: 35) and it was also true for the MRP: “the MRP was never able to install itself firmly in the French Catholic environment” (Kemseke 1997: 185). Not only workers, but also Catholic farmers were reluctant to tie themselves to the MRP in post-war France. Initially, the Catholic agrarian interest groups like the JAC and the Mouvement familial rural (MFR) “seemed to offer a ready-made mass organizational base for the […] activities” of the Christian Democrats (Wright 1953: 541). But after early electoral success in the countryside the MRP could not repeat it in subsequent elections. As was the case with the Christian Union, also the JAC and the MFR refused to bind themselves to the Mouvement and the farmers were ready to switch party soon, either to the Gaullist RPF or other conservative parties. Part of the electoral success of de Gaulle’s RPF was that it successfully tapped the Catholic vote with a corporatist platform, while at the same time it opened the party lists to representatives of the local Catholic organizations. Independent candidates could run on a RPF-list or on the ticket of other parties and their electoral success was based on the local support they could muster. One very impor-

5 The plea for the acceptance of the laïc public school was particularly voiced by the union of public teachers within the CFTC.

164

tant element in this local support structures was the mutuelle of the Sécurité Sociale agricole for farmers and the caisses of the general Social Security scheme for workers (Wright 1953). Crucial was then the political promise – which the Gaullist movement was quick in making – to leave this local autonomy untouched, in the case of farmers complemented with the promise to help generously in financing the Sécurité Sociale agricole with substantial state subventions. A similar logic motivated the largest union, the communist CGT (Confédération générale du Travail) with a membership around eight times as large as that of the second largest CFTC (with 6.5 million CGT-members in 1947; Galant 1955: 69), to vote for a corporatist design of the new social protection system, in which – predominantly – workers would administer the social insurance schemes under moderate state tutelage. The CGT had most fiercely attacked the various attempts at welfare reform in the interwar years as a “bluff democratique” (Galant 1955: 14), and one of its foremost points of critique was the principle of payroll taxation as the welfare state’s dominant financing mode which was castigated as a “chicanerie bourgeoise” (ibid.). The CGT, however, had come to realize in the meantime that contribution finance legitimized self-administration and with its dominant position among all unions, the CGT could rightly expect to dominate the governing boards of the social insurance funds. Given that one could easily foresee that the Communist PCF would not remain part of the national government forever (in fact, it left the government already in 1947 and would remain excluded from power until 1981), the CGT became much more favourable to the idea of the independence and autonomy of the new social insurance schemes and, by implication, also to the principle of contribution financing. Thus, in 1945/46, representatives of the Communist and the Christian unions were for the first time united in their fight for the autonomy of the social insurance schemes and voiced no major protest against the principle of contribution finance (Galant 1955: 37). The church was not unequivocally supporting the MRP either. Tied to the coalition with the vehemently anti-clerical Socialists and Radicals (Hanley 2002: 132), it was far from clear whether the MRP was really the best partner to protect church interests. In fact, the French Christian Democrats were much more active in social policy, where a common ground between the left-Catholic wing and the Socialists was so much easier to find, than in education policy, which the church and the catholic electorate so

165

much valued. In the 1946 constitution the freedom of education had not been explicitly guaranteed, and a considerable faction within the MRP had accepted the republican principle of public schooling, something which the church hierarchy would never do. The episcopate’s strategic calculations also played a role. In the municipal elections of 1947 the MRP had only gained 10 percent of the vote (at the municipal level the old majoritarian system had remained in place), in the 1951 federal election the MRP’s national vote share was cut by half (from 25.9 to 12.6 percent) and both times the Gaullists emerged as the strongest party of the right. Providing the MRP with the church’s full support would have meant risking to bet on the wrong horse (Béthouart 2001: 316–317; Kemseke 1997: 184–185).6 Therefore the French episcopate issued no explicit recommendations to Catholics on how to vote in the 1951 elections (Kemseke 1997: 178), and a considerable degree of Catholic voters apparently found the Gaullist RPF more attractive than the MRP (ibid. 183; Goguel 1981). Gaullists had very cleverly presented themselves as the better defenders of the Catholic cause, especially in the school question (Letamendia 1995: 292–298; Goguel 1981: 109). The Loi Barangé of 1951 granted public payments to all children at school age, including those who went to Catholic schools, that is it broke with the ‘sacred principle’ of laïcisme. The law bore the name of a MRP-deputy, but it had been the Gaullists who had initiated the law in order to force the MRP to posit itself against its coalition partner, the Socialists (van Kemseke 1997: 183, Fn. 28; Bétouart 2001: 322; Letamendia 1995: 296). Recognizing the weakness of the Mouvement Républicain Populaire in the critical question of education7 de Gaulle intended to capture the Catholic vote and to drive a wedge between the anticlerical Socialists (SFIO) and Catholic MRP – with success: right after the 1951 elections the SFIO left the government in protest against the Loi Barangé (Hanley 2002: 128). De Gaulle, himself a practising Catholic, used each occasion to signal his attachment to the Catholic cause (Letamendia 1995: 298). He declared: “je suis un Chrétien social” and described his strategy quite openly: “what we have to do in order to trans-

6 The disregard was mutual: many MRP members with a Resistance background though that the French episcopate had remained loyal to Petain for much too long (Kemseke 1997: 178; Hilaire 1997: 430–431). 7 The MRP could easily be accused of “mollesse dans la défense de l’école catholique” (Letamendia 1995: 294).

166

form us into a new political force is … to occupy the terrain which previously had been occupied by the right, and then subsequently to consolidate and regroup all the forces of Catholic origin. The MRP has not wanted to do this and that is why it has neither presented itself nor defined itself as a Christian Democratic party” (quoted from Béthouart 1997: 327–328; our translation). One of the first measures of the new government under de Gaulle in the Fifth Republic was then to put an end to the “endless conflict of influence between the state and the Catholic church” in education (Archambault 1997: 151). The Loi Debré of 1959 guaranteed full public financing of private schools, once these agree to be under financial and pedagogical oversight by the state (Archambault 1997: 152). Finally the “school war calmed down”, after it had preoccupied French politics for more than 50 years (ibid.).8 At the same time de Gaulle left the social policy sector to the social Catholicism faction of the MRP. Even after 1958 labor minister Paul Bacon of the MRP served for another four years under de Gaulle. It is here, in social and family policy, where the influence of the MRP on France’s post-war order is most visible (Letamendia 1995: 364–366; Bell 2000: 108) and became mainly manifest in the preservation of a high degree of program fragmentation and a strong emphasis on family policy. Essentially the left idea of a unified, national system of social insurance died with the June elections of 1946, when “Socialist strength was deeply eroded and the Catholic left, the MRP, was substantially enhanced” (Ashford 1991: 39). While the communist Minister of Labour remained in office, all other positions in government and parliament that were of relevance for welfare reform were taken by the Christian Democrats (ibid.). And this remained to be the case throughout the Fourth Republic. In 18 cabinets which governed France between 1950 and 1962, it was the MRP that 14 times nominated the minister of labour (of the other

8 It was only in 1984 that the school war saw a revival, when the socialist Mauroy government seemed to challenge the autonomy of the private schools (Loi Chevenement). Mass demonstrations in Paris with over 1 million participants finally forced the socialist government to back down and the following Fabius government dropped the bill (Archambault 1997: 152; Hilaire 1997: 436). Once conservatives tried to abandon the caps with which the Loi Falloux has limited public financial support for private schools in December 1993, this again provoked mass demonstrations with around 600.000 participants in Paris alone. This time it was the anticlerical camp that mobilized. Finally, the Constitutional Court ruled the abolition of the Loi Falloux unconstitutional (Archambault 1997: 156–157; Hilaire 1997: 436). As these events show, the state/church conflict over education is far from being settled in France.

167

four two were Socialists and two other independents very close to the MRP). Nine out of these twelve years Christian Democrats led this portfolio. Paul Bacon was the longest serving minister of labor between 1950 and 1962, and he managed to surround himself with a group of young administrators loyal to the MRP and influential in the social policy debates of the 1960s and 1970s (Béthouart 2001: 321, Béthouart 1996). All major reform decisions in social policy were taken under MRPresponsibility and it was in the first two decades from 1946 to the mid-1960s that the basic institutional architecture or the French welfare state was put in place (see Palier 2002: 109–111). In fact, much of the corporatist outlook of the French welfare state has to be attributed to the MRP’s influence on post-war social policy: The protection of the autonomous status of the social insurance schemes for the self-employed, the introduction of PR-elections for the governing bodies of the social security administration, which granted Christian unions such a strong organizational base, the successful fight against the integration of the Securité Sociale agricole in the overall social security framework as well as the limited integration of the French cadre, in all these cases it was the MRP that could claim responsibility. With these measures the MRP secured the overall Bismarckian or conservative outlook of the French system of social protection (Béthouart 2001: 320–321, 1996). Also with respect to financing and spending the French welfare state came to resemble very much a conservative welfare state regime. In the 1950s and 1960s the French welfare state was generous and transfer heavy. With social security transfers amounting to 13 percent of GDP in 1960 it belonged to the most generous welfare states in the OECD (Cameron 1991: 61–64). Most welfare spending is financed through payroll taxation, while general and specific taxation earmarked for social policy together only accounted for about 10 percent of all welfare state revenue. The French welfare state still today is a big spender, something hard to explain not only within the conventional power resources framework, given the political dominance of conservative parties in the post-war decades and given the low degree of unionization of the French workforce. The French case also sits uneasily with an account that emphasize mainly the importance of electoral rules (Iversen/Soskice 2006; see van Kersbergen and Manow, this volume) given that the French majoritarian electoral system has obviously failed to result in the type of liberal-residual welfare state that this approach would predict.

168

Post-war family policy was strongly influenced by the MRP as well, whereas not many allusions to family policy could be found in the programs of the Communists and the Socialist (Lenoir 1991: 153). As was the case with the labor ministry, the Ministère de la Population et de la Santé was a MRP stronghold as well. In August 1946 Robert Prigent (MRP) saved the caisses d’allocations familiales from integration into Sécurité Sociale (Ambler 1991: 22). He introduced allocations prenatales, augmented significantly birth premiums as well as child allowances, which were extended to the entire population, and prolonged maternity leave to 14 weeks (Béthouart 2001: 320). Other ministers of the MRP who occupied this portfolio (Poinso-Chapuis, Pierre Schneiter, Joseph Fontanet) further contributed to the particular family friendly outlook of the French welfare state (Hilaire 1997: 431). In 1949 family policy made up 40 percent of social security expenditures (Lenoir 1991: 159), and still in 1960 35 percent of total social expenditures went to families. It was only later that steadily increasing pension spending slowed down the growth of family expenditures. Rémi Lenoir summarizes: “Without a doubt we must attribute both the rapid development and the orientation of family policy under the Fourth Republic to the strategic position of the MRP in the political field of this time and also to the alliance in which the MRP was almost always a leading party” (Lenoir 1991: 159; Hilaire 1997: 431, 435–436). While Communists and Catholic unions united in the question of autonomy and independence of Sécurité Sociale (see above), its high degree of organizational fragmentation, the strong role of self-administration and the particular emphasis on family policy must be attributed to the influence of the MRP, which fought together with the interest groups against what they perceived as the threat of universalism and uniformity (Ambler 1991: 22). Interest groups, in particular the unions, had a strong interest in the Sécurité sociale as a stable organizational support structure, which the ephemeral French parties were not able to provide. And it was to the protection of these structures that the interest of the syndicates chretiennes and the syndicates agricoles became primarily targeted. Participation of unions in the administration of the social insurance schemes (autogestion) promised to provide unions and farmers’ associations with “an autonomous sphere for syndical activity” (Berger 1987: 123; cf. Rothstein 1992). Unions gained influence within the social insurance administration through their participation in the elections to the governing boards that were responsible for the ad-

169

ministration of the funds. Initially, the two ordonannces of October 1945 which established the French Sécurité sociale had granted the union which organized the majority of the insured the right to appoint administrators to the governing boards of the insurance funds (Galant 1955: 35–36; Dutton 2002: 215). This favoured the largest union, the CGT, and subsequently met with sharp criticism by the Catholic CFTC. Protest was also voiced by the MRP and the Socialists who feared communist hegemony. As a reaction, the Christian union refused to cooperate in the administrative bodies of the social insurance schemes, thereby granting to the Communists the monopoly of representation in the crucial period of early institution building. After fierce struggles within the provisional government an ordonnance in October 1946 made elections for the governing bodies of Sécurité Sociale obligatory, and proportional representation was the principle that was going to be applied in these elections (Galant 195: 99–102). It was the MRP that had fought intensively for this amendment and it was the influential MRP labour minister Paul Bacon who later counted this among his most important political achievements (Béthouart 1996). In the first elections of 1947 the CFTC gained 26.4 % of the votes (the CGT 59.3 %), and when Force Ouvrière (FO) after a split from the Communist union also participated in the following elections in 1950, the Christian Union and FO together received around 36 % of the vote (against 43.5 of the communist CGT). These vote shares remained relatively stable in the following elections of 1955 and 1962 (Comite d’Histoire 1988: 132) and basically the same results were obtained in the elections to the caisses d’allocations familiales with a joint vote share around 35 % for the Christian Unions plus FO and a vote share around 48 % for the communists (ibid.: 134). One of the main points of contention that had ultimately led to the split between CGT and FO had been the latter’s protest against the politicization of the social protection system. In the following years Force Ouvrière frequently allied with the Christian union to achieve a majority over the Communist CGT on the management boards of the caisses, a pattern of joint management that was to be found in many boards until the early 1990s. The unions’ prominent role in the administration of the social insurance schemes provided them with highly valued resources. Already in the 1950s, the Sécurité Sociale employed more than 70,000 employees, this number more than doubled in the next 20 years and reached 165,000 employees in 1979 (Comité d’Histoire 1988: 267– 268), and around this level it is still today. Given that unions dominated the govern-

170

ing boards of caisses and therefore also elected the presidents, while presidents oversaw the hiring decisions, it comes as no surprise that unions were actually able to implement a ‘closed shop’-principle, that is they often could make union membership a precondition for employment in the social insurance schemes (Galant 1955: 174– 175). The degree of unionization among the personnel of the Sécurité Sociale is therefore high. Moreover, once employed, the personnel of Sécurité Sociale enjoyed remarkably generous working conditions, given that bargaining took place between unions as representatives of the funds and unions as representatives of the workers. No surprise that already contemporaneous observers noted “une certaine confusion entre employeurs et employés” (Comité d’Histoire 1988: 269). The attractive working conditions for workers in the Sécurité Sociale set precedents for union demands in other sectors of the economy and helped union mobilization in them (ibid.: 172). To sum up this section: That France has a Conservative welfare state regime but no Christian democratic party is less of a puzzle once we take into account the MRP’s exceptional policy influence in these formative years of post WW II French welfare state building. In fact, in these years in which PR was the electoral system in place a Christian democratic party was present in the French party system. It was strong and pivotal and it behaved very much like its party homologues in France’s European neighbour countries. For the MRP it was much easier to find a common ground with its anticlerical government partners (Socialists and Radicals) in questions of social policy than in questions which directly touched upon the state/church conflict. The fact that the school question even after the Second World War had still not been settled, because the Catholic church had never accepted its defeat in 1905, permanently destabilized the coalition between anticlerical Socialists and the MRP, whereas the red-black coalition with respect to social protection was as evident in post-war France as in any other continental European country around this time. Closer inspection, therefore, makes the French case not only less of an outlier among the conservative regime types, but even turns it into a strong supporting case for an approach that stresses the importance of the interaction between PR electoral rules, a virulent state/church conflict, which leads to the strong role of parties of religious defence in the European party systems, which then become particularly influential and important for the build up of the continental European post-war welfare state. Given the high degree of institutional inertia of welfare state programmes, the formative post-

171

war period proved critical for the further development of the French welfare state (Palier 2002: 107-165). We therefore can identify a lasting influence of Christian democracy on the French post-war welfare state even if a Christian Democratic party vanished from the French party system in 1967.

6.4

Conclusion

In this chapter we have focused on two particularly important periods of French welfare state formation, the last quarter of the 19th and first decade of the 20th century as well as the immediate post WW II years. Of course, the French Etat providence has not been set up wholly during these times. But the origins of many of its institutional characteristics can be traced back to these two periods. As we argued in the preceding sections, in both periods the religious/ secular conflict had been of enormous importance. Some of the French welfare state’s features that appear rather untypical for a Conservative welfare state regime reflect the fierce state/church conflict that dominated French politics in the early decades of the Third Republic. This conflict left its imprint on early social legislation, especially in the field of social assistance, as well as on education policy with important long-term social policy consequences (Morgan 2003, this volume).

The pre-eminence of the French state in social assistance, which does not leave much room and responsibility for welfare provision to voluntary organizations, distinguishes the French welfare state from the prototypical conservative welfare state regime type. The French welfare state does not have this structuring and empowering of the voluntary sector found in countries like Germany or the Netherlands, and the Third Sector itself is less developed in France than elsewhere on the continent (Archambault 1997). The same is true for the unusual un-patriarchic features of French social policy. The predominance of the state in child care and early education combined with the strong role that family policy plays within the French Etat providence helped securing a much higher degree of female labor force participation than to be expected for a conservative welfare state regime. It was the laic separation between church and state and not the contractual integration of the church into the provision of public goods which characterized the French solution to the fierce conflict between

172

catholic church and secular nation state that waged in all countries of continental Europe in the last quarter of the 19th century.

However, as we showed in the preceding section, where the French welfare state does resemble its continental homologues, it was the influence of Christian Democracy, which must be held responsible for these many institutional similarities. French Christian Democracy was not as strong as elsewhere, but in the short period in which it appeared on the scene, it was strong enough to influence core welfare state features. The high degree of occupational fragmentation, the importance of selfadministration with a prominent role of the Christian union, the dominance of payroll taxation and its correlates (contribution differentiated benefits, length of drawing period dependent on individual contribution record, insurance based on being dependently employed) - all these features had been either introduced or protected by the MRP in the reform debates of the immediate post-war years.

In this critical period of post-war reconstruction, when the French electoral system for the first time followed PR, Christian Democracy was a powerful and pivotal party. And social policy was one of the policy sectors of highest salience for the Mouvement Républicain Populaire. It was also the policy sector, where political agreement was easiest to reach in the coalition governments of the Fourth Republic, in which the MRP ruled together with Socialist, Independents or Radicals (and until 1947, the Communists). The school question, however, remained a highly contested issue between Christian Democrats and the fiercely anti-clerical left, and it was the MRP’s lack of a clear position in this question which finally motivated the church hierarchy’s refusal to back the MRP with its full support, which – in turn – partly explains the premature vanishing of a Christian Democratic party from the French party system in 1967.

References Alexander, G. (2004). France: Reform-mongering between Majority Runoff and Proportionality. Handbook of Electoral System Choice. J. M. Colomer. Houndmills, Basingstoke, Palgrave MacMillan: 209–221. Ambler, John S., 1991: Ideas, Interests and the French Welfare state. In: John S. Ambler (ed.), The French Welfare State. Surviving social and ideological change. New York – London: New York University Press, 1–31.

173

Archambault, Edith, 1997: The nonprofit Sector in France. Manchester – New York: Manchester University Press. Ashford, Douglas E., 1991: Advantages of Complexity: Social Insurance in France. In: John S. Ambler (ed.), The French Welfare State. Surviving social and ideological change. New York – London: New York University Press, 32–57. Ashford, Douglas E., 1986: The Emergence of the Welfare State. Oxford – New York: Blackwell. Bell, David Scott, 2000: Parties and democracy in France: parties under presidentialism. Aldershot: Ashgate. Berger, S. (1987). Religious Transformations and the Future of Politics. Changing Boundaries of the Political. Essays on the evolving balance between the state and society, public and private in Europe. C. S. Maier. New York, Cambridge University Press: 107–149. Bernstein, Serge, 2001: Quatrième République. In: Pascal Perrineau/Dominique Reynié (Ed.), Dictionnaire du Vote. Paris: Presses Universitaires de France, 779–782. Béthouart, Bruno, 2001: Le Mouvement Républicain Populaire. L’entrée des catholiques dans la République française. In: Michael Gehler/Wolfram Kaiser/Helmut Wohnout (Eds.), Christian Democracy in 20th century Europe. Wien – Köln – Weimar: Böhlau, 313–331. Béthouart, Bruno, 1996: Le Ministère du Travail et de la Sécurité sociale: un monopole du MRP de 1950 à 1962. In: Revue d’histoire moderne et contemporaine, 1 (43), 67–105. Comite d’Histoire de la Sécurite Sociale, 1988: La Sécurite Sociale – son Histoire a travers les Textes. Tome III: 1945–1981. Paris: Association pour l’etude de l’Histoire de la Sécurité Sociale. Delbreil, Jean-Claude, 2001: Le parti démocrate populaire. Un parti démocrate chrétien français de l’entre-deux-guerres. In: Michael Gehler/Wolfram Kaiser/Helmut Wohnout (Eds.), Christian Democracy in 20th century Europe. Wien – Köln – Weimar: Böhlau, 77–97. Delbreil, Jean-Claude, 2001a: The French Catholic Left and the Political Parties. In: GerdRainer Horn/Emmanuel Gerard (Eds.), Left Catholicism 1943/1955. Catholics and Society in Western Europe at the Point of Liberation. Leuven: Leuven University Press, 45–63. Delbreil, Jean-Claude, 1990: Centrisme et démocratie-chrétienne. Le Parti Démocrate Populaire des origines au MRP. 1919–1940, Paris, Publication de la Sorbonne. Descamps, Henri, 1981: La Démocratie Chrétienne et le MRP: De 1946 a 1959. Paris: Pichon & Durand-Auzias. Dreyfus, François-Georges, 1988: Histoire de la démocratie chrétienne en France. Paris: Albin Michel. Dumons, Bruno/Gilles Pollet, 1994: L’Etat et les retraites. Genese d’une politique. Belin: Paris. Duriez, Bruno, 2001: Left Wing Catholicism in France. From Catholic Action to the Political Left: The Mouvement populaire des Familles. In: Gerd-Rainer Horn/Emmanuel Gerard (Eds.), Left Catholicism 1943/1955. Catholics and Society in Western Europe at the Point of Liberation. Leuven: Leuven University Press, 64–90. Elgie, R. (2005). France: Stacking the Deck. The Politics of Electoral Systems. M. Gallagher and P. Mitchell. Oxford – New York, Oxford University Press. Esping-Andersen, G. (1990). The Three Worlds of Welfare Capitalism. Cambridge, Polity Press. Ferrera, M. (1996). “The “Southern Model“ of Welfare in Social Europe.“ Journal of European Social Policy 6(1): 17–37.

174

Ferrera, M. (1997). Introduction géneral. Comparer les systèmes de protection sociale en Europe du Sud. MIRE. Paris, Ministere Affaires: 15–26. Galant, Henry, 1955: Histoire politique de la sécurité sociale française 1945–1952. Paris: Armand Colin. Goguel, François, 1981: Chroniques électorales: les scrutins politiques en France de 1945 à nos jours. 1, La Quatrième République. Paris: Presses de la FNSP. Groux, Guy/René Mouriaux, 1991: Le Syndicalisme Chretien en France: Son essor, ses limites. Paris: CEVIPOF. Hanley, David, 2002: Party, Society, Government. Republican Democracy in France. New York: Berghahn Books. Hatzfeld, Henri, 1989: Du Pauperisme a la Sécurite Sociale, 1850–1940. Essai sur les origines de la Sécurite sociale en France. Nancy: Presses Universitaires de Nancy. Hilaire, Yves-Marie, 1977: L’Influence de la Démocratie Chrétienne dans la Sphère de la Religion, la Morale et la Culture en France. In: Emiel Lamberts (Ed.), Christian Democracy in the European Union, 1945/1995. Leuven: Leuven University Press, 427–438. Huber, E., C. Ragin, et al. (1993). “Social Democracy, Christian Democracy, Constitutional Structure, and the Welfare State.“ American Journal of Sociology 3: 711–749. Huber, E. and J. D. Stephens (2001). Welfare State and production regimes in the era of retrenchment. The New Politics of the Welfare State. P. Pierson. Oxford, Oxford University Press. Kalyvas, S. (1996). The rise of Christian Democracy in Europe. Ithaca, NY, Cornell University Press. Kemseke, Peter van, 1977: The Societal Position of Christian Democracy in France. In: Emiel Lamberts (Ed.), Christian Democracy in the European Union, 1945/1995. Leuven: Leuven University Press, 174–188. Korpi, W. (1995). “Un État-providence contesté et fragmenté. Le développment de la citoyenneté sociale en France. Comparaisons avec la Belgique, l’Allemagne, l’Italie et la Suède.“ Revue francaise de science politique 45(4): 632–667. Labbe, Dominique/Antoine Bevort/Maurice Croisat, 1991: Effectifs, Audience et Structures Syndicales en France depuis 1945 (Le cas se la CFTC-CFDT). Insitut d’Etudes Politique de Grenoble. Lanfrey, A, 2003: Secularisation, séparation et guerre scolaire. Les catholiques francais et l’ école, 1881–1911. Paris: Cerf-Histoire. Launay, Michel, 1984: Le syndicalisme chrétien en France de 1885 à nos jours. Paris : Desclée. Lenoir, Rémi, 1991: Family policy in France since 1938. In: John S. Ambler (Ed.), The French Welfare State. Surviving Social and Ideological Change. New York – London: New York University Press, 144–186. Lenoir, Rémi, 1991a: Famille et Politique. Paris: Editions de Minuit. Letamendia, Pierre, 1995: Le MRP, histoire d’un grand parti français. Paris : Beauchesne. Levy, J. D. (2000). France: Directing Adjustment? Welfare and Work in the Open Economy. F. Scharpf and V. Schmidt. Oxford, Oxford University Press: 308–350. Luebbert, G. (1991). Liberalism, Fascism and Social Democracy. Social Classes and the Political Origins of Regimes in Interwar Europe. New York, Oxford University Press. Maddison, A. (1992). Monitoring the World Economy 1820–1992.

175

Mayeur, Francoise, 1981: Histoire de l’ enseignement et de l’éducation, Tome III: 1789–1930. Paris: Éditions Perrin. Mayeur, J.-M. (1984). La vie politique sous la Troisième République, 1870–1940. Paris, Éditions du Seuil. Michelat, Guy/Michel Simon, 1977: Class, religion et comportment politique. Paris: Éditions sociales. Morgan, K. J. (2002). “Forging the Frontiers between State, Church and Family: Religious Cleavages and the Origins of Early Childhood Education in France, Sweden, and Germany.“ Politics & Society 30(1): 113–148. Morgan, K. J. (2003). “The Politics of Mothers’ Employment.“ World Politics 55(January): 259– 289. Morgan, K. J. (2004). The religious origins of the gendered welfare state. The example of work-family policies in Western Europe. Religion and the Welfare State, Cologne. Nord, Philip, 2003: Catholic Culture in Interwar France. In: French Politics, Culture & Society. 21 (3), 1–20. Palier, B. (2002). Gouverner la sécurité sociale. Paris, Presses Universitaires de France. Palier, B. (2002). Gouverner la sécurité sociale. Les réformes du système francais de protection sociale depuis 1945. Paris, Presses Universitaires de France. Palier, B. (2004). “The Politics of the Reforms of Bismarckian Welfare Systems.“ Pasture, Patrick, 2001: Multi-faceted Relations between Christian Trade Unions and Left Catholics in Europe. In: Gerd-Rainer Horn/Emmanuel Gerard (Eds.), Left Catholicism 1943/1955. Catholics and Society in Western Europe at the Point of Liberation. Leuven: Leuven University Press, 228–246. Pedersen, S. (1993). Family Dependence and the Origin of the Welfare State, Britain and France, 1914–1945. Cambridge, Cambridge University Press. Renard, D. (1987). “Assistence publique et bienfaisance privée, 1885–1914.“ Politiques et management public 5(2): 107–128. Renard, D. (1994). “La Direction de l’Assistance publique au ministère de l’Interieur (1885– 1905): Administration nouvelle, politique nouvelle?“ Philanthropies: 187–197. Renard, D. (1995). “Intervention de l’Etat et genèse de la protection sociale en France (1880– 1940).“ Lien social et politiques – RIAC 33: 13–26. Rosanvallon, Pierre, 1990: L’État en France de 1789 à nous jours. Paris: Éditions du Seuil. Saint-Jours, Yves/Michel Dreyfus/Dominique Durand, 1990: Traité de la Sécurite Sociale, Tome V: La Mutualité. Librarie Generale de Droit et de Jurispudence: Paris. Scharpf, F. W. and V. A. Schmidt (2001). Introduction. Welfare and Work in the open Economy Vol. II Diverse Responses to common Challenges. F. W. Scharpf and V. A. Schmidt. New York, Oxford University Press. Stone, J. F. (1985). The Search for Social Peace: Reform Legislation in France, 1890–1914. Albany, State University of New York. Warner, Carolyn M., 1998: Getting out the Vote with Patronage and Thread: The French and Italian Christian Democratic Parties. In: Journal of Interdisciplinary History 28(4), 553–582.

176

Weiss, J. (1983). “Origins of the French Welfare State: Poor Relief and the Third Republic, 1871–1914.“ French Historical Studies 13(1): 47–78. Wright, G. (1953). “Catholics and Peasantry in France.“ Political Science Quarterly 68(4): 526–551.

177

7

Religion and the Consolidation of the Swiss Welfare State, 1848–19451

Herbert Obinger

7.1

Introduction

Switzerland is not only the oldest democratic nation-state, but also the oldest federal polity in Europe. Given deep-rooted societal cleavages of language, ethnicity, religion and a strong divide between rural areas and urban agglomerations, the formation of a multi-cultural nation state in 1848 was a truly revolutionary venture. Leading representatives of the Swiss labour movement pinned their hopes on the democratic institutions to overcome the wide-spread pauperisation of the working-class that had emerged in the wake of the early industrialization of the north-eastern parts of the country. But the dreams of ameliorating the living standard of workers by means of social policy were dashed. The formation of a national welfare state faced many obstacles and took longer than in most continental countries. This is particularly true for social security – neither a public pension scheme nor unemployment insurance was adopted prior to 1945. Why did this hope proved to be an illusion? This chapter examines the driving and braking forces that have shaped Swiss social policy over the period 1848–1945. Special attention is paid to the impact of religion on early welfare state formation. Given the cultural modernization and the decline in religiosity in the post-war period, the impact of religion on social policy should be most pronounced during the formative period of the welfare state. Nevertheless, the fierce state-church conflict of the 19th century and the resulting establishment of parties of religious defence had much more long lasting consequences for welfare state building in Switzerland. My argument is that religion is indeed important for explaining the developmental trajectory of the Swiss welfare state, but it has unfolded its impact on social policy in indirect and complex ways. Two channels of influence were of particular importance. First, the heterogeneous societal make-up exerted a substantial impact on state and institu-

1 I am grateful to the editors for valuable comments and to Julia Moser, Andrea Britze and Janis Vossiek for research assistance.

178

tion building. Federalism was the institutional response to a multi-dimensionally fragmented society and led, in consequence, to a territorial and confessional fragmentation of power resources. In addition, the ”secondary institutions of federalism” (Lijphart 1999) as well as the institutions of direct democracy constituted strong veto points through which opponents of public intrusion into social and economic affairs were empowered to defend the status quo. Together, these factors had a substantial impact on the timing and patterns of social security legislation. Second, the impact of religion on public policy is mediated by the party system and the partisan complexion of the federal government. Similar to the situation in the Netherlands prior to 1980 (van Kersbergen 1995: 43), an interdenominational Christian democratic party did not arise. Protestant voters were mainly absorbed by the laïcist Liberals, whereas genuine Protestant parties lacked electoral strength und therefore never gained a seat in the federal government. Political mobilization of Catholics, in contrast, was strong and the Catholic-conservatives (since 1912 organized in the Catholic People’s Party, renamed in Christian Democratic People’s Party in 1971) have been represented in the national government since 1891. Their entry into government was substantially influenced by the veto powers related to direct democracy. Its frequent use by Catholics in the last quarter of the 19th century forced the politically powerful Liberals to co-opt their former archenemy into government. Hence, the main rivals of the Kulturkampf period have formed coalition governments since the late 19th century. The process of “paradigmatic integration” (Deutsch 1976) was continued when the Peasant, Trade and Citizen’s Party, a conservative agrarian party, received a seat in the federal government in 1929. Finally, World War II paved Social Democrats the way to the corridors of power (see Table 1). In a nutshell, the traditional liberal hegemony was incrementally replaced by a balanced distribution of power as the country gradually moved from majoritarian to consociational practices during the 60 years before 1945. Since the political left was excluded from government until 1943, there was no “redblack” coalition in Switzerland before the 1940s. Patterns of coalition-building were decisively shaped by the multiple institutional veto points which in turn mainly result from the numerous cleavages in Swiss society. At least during the period under scrutiny, the referendum democracy proved to be a stumbling block for a coalition between Christian democrats and Social democracy that can be found in other continental countries. The major reason is that the optional referendum (and federalism)

179

played a diametrically opposed role in the political strategies of both parties. For the Catholics the referendum represented a sort of safety valve to avert majorization and public interference in local affairs. The veto power associated with the referendum forced the Liberals to share the power with the Catholic Party. By contrast, Social Democrats did not attack the reforms initiated by the Liberals since they basically supported liberal efforts to extend the powers of the federal government. Moreover, federalism turned out to be a major impediment against leftist ideas to enhance the responsibilities of the federal government in social and economic affairs. In what follows it will be argued that the lack of a “red-black” coalition and the constitutional rigidities imposed by federalism and direct democracy largely account for the belated take-off of the Swiss welfare state and its pronounced liberal traits (Obinger 1998a). The chapter is organized as follows. Section 2 examines the impact of religion on state-building and shows that federalism is the institutional response to the multiple cleavages inherent in Swiss society. Section 3 analyses the repercussions of federalism and direct democracy on interest formation and the power resources of collective actors and illustrates the ways in which the constitutional setting has changed the social policy preferences and strategies, not only of political and economic actors, but also of religious groups and the churches. Section 4 exemplifies how these factors have influenced social policy making between 1874 and 1945. The final section concludes.

7.2

Religion and State-Building

Switzerland is a religiously mixed society. Though the numerous cleavages in Swiss society are cross-cutting in the sense that linguistic, religious and socio-economic divisions do not cluster territorially (Gruner 1969: 18), religious tensions overshadowed modern state and nation-building from the very outset. By the mid of the 19th century, the share of Roman Catholics amongst the population amounted to approximately 40 percent. Among the then 25 sovereign states (the cantons), which formed a loose confederation in 1815, eleven mostly rural states were predominantly Catholic, ten cantons were Protestant, whereas the remaining four cantons showed more balanced denominational patterns (Linder 2005: 36–37). This religious land-

180

scape was the product of several wars between Catholics and Protestants during the 16th and 17th century. Protestant Churches traditionally were ‘canton churches’ (Kantonalkirchen or Landeskirchen) and thus strongly enmeshed with every canton’s political structure (Gould 1999: 91; Cattacin et al. 2003: 14). Since the position of the churches strongly depended on cantonal authorities, the Protestant clergy did not oppose the Liberals’ efforts to overthrow the ancient order from the 1830s onwards (Gould 1999: 104). But since these plans naturally involved an attack against the Roman Catholic Church and its strong role in domains like education, the Liberals encountered fierce Catholic resistance. When eight Catholic cantons2 formed a special alliance (the so-called Sonderbund) in 1845 and quit the confederation’s body of decision-making (the confederal Diet) one year later, Protestant cantons regarded this move as an act of secession and mobilized troops. Following a short-lasting civil war between Catholic and Protestant cantons, the victorious Liberals (the Freisinn) imposed a Federal Constitution against the will of Catholic-conservative forces in 1848, thereby realizing core liberal ideas such as people’s sovereignty3, secularism and the creation of a national market through the removal of trade barriers between the cantons. The Federal Constitution strongly borrowed from the liberal constitutions that had emerged at the cantonal level from 1830 onwards (Kölz 1998: 27). State-building, therefore, was a bottom-up process supplemented by a transfer of institutions from abroad. By using US-federalism as a blueprint, a federal polity was established that shifted only limited responsibilities to the federal government and guaranteed cantons a powerful influence on the national decision-making process. Federalism was a compromise between the more centralist-oriented Liberals and the federalist Catholic-conservatives (Kölz 1998: 28). Gruner (1969: 23) even sees the formation of a federal state in 1848 as the result of a belated religious war. A symmetric bicameral system and equal representation of the cantons in the Council of States based on the Senate principle provided the smaller Catholic cantons with disproportional influence and veto powers. In addition, any constitutional amendment required high con-

2

Lucerne, Uri, Schwyz, Obwalden, Nidwalden, Zug, Fribourg and Valais.

3 In practice, however, suffrage at the national level was restricted to male citizens until 1971. In addition, the poor and Jews initially were excluded from political participation (see Mooser 1998: 50–52).

181

sensus thresholds. Since 1848, constitutional amendments are subject to a mandatory referendum which requires the approval of voters nationally and in a majority of the cantons. This so-called double majority once again favoured the smaller Catholic cantons. Religious tensions did not disappear, but were temporarily relieved as a result of the new constitutional setting. The Liberals did not impose Protestant churches on Catholic territories, which in addition enjoyed autonomy in many fields of public policy under the Federal Constitution of 1848. Even more important is that federalism changed the political strategy of the defeated Catholics. Their response to the new constitutional setting was a retreat to cantonal strongholds where they began to establish a Catholic sub-society or Sondergesellschaft (Altermatt 1972, 1991). Similar to competitive state building in Canada (cf. Banting 2005), Catholics used the constituent units as a political and cultural fortress to defend a Catholic way of life within a federal polity. While decentralization, bicameralism and constitutional rigidity protected Catholic interests from majorization, the majority rule-based electoral system had the opposite effect, because it gave rise to a liberal hegemony in parliament and government. The Swiss federal government (Federal Council), a collegial body consisting of seven ministers, was exclusively composed of Liberals during the four decades following the formation of the federation in 1848. However, the liberal movement represented a broad ideological spectrum, ranging from Manchester-liberals to Radicals and Democrats who were committed to social policy reforms, anti-clericalism and a powerful federal government (Gruner 1969: 73–79). Though predominantly Protestant in denomination, the Liberals regarded religion as a private affair and championed, in consequence, a clear-cut separation between the state and the various churches. Liberal anti-clericalism mainly focused on the Roman Catholic Church, which was branded as a conservative force and accused of impeding societal progress and modernization. In particular, Liberals considered the Jesuit order “an anathema, as the apotheosis of Catholic obscurantism, intrigue, and subversion” (Gould 1999: 106). This strong anti-Catholic stance, together with the anti-secularist counter-movement that has emerged in the wake of the First Vatican Council, triggered severe tensions between Catholics and Liberals. The resulting Kulturkampf peaked in the early 1870s. Catholic bishops and clergymen like E. Lachat (Basel) and G. Mermillod (Geneve),

182

who strongly backed the dogma of papal infallibility, were defrocked and expelled from the country. This struggle between Catholics and Liberals was fuelled by distinctive and deeply entrenched attitudes towards modernization, state-building and the role of the Roman Church in education. Like in 1848, the Liberals prevailed in this conflict. By means of a total revision of the Federal Constitution in 1874, they finally succeeded in creating a secular state. Though the constitutional revision successfully passed the referendum hurdle4, the former Sonderbund cantons plus two other mostly Catholic cantons (Appenzell Inner Rhoden and Ticino) rejected the overhaul of the constitution, because a series of its newly-adopted provisions radically enforced securalization. The constitution of 1874 banned the pope-loyal Jesuit order and the establishment monasteries and fraternities. In addition, the establishment of new dioceses required the permission of federal authorities and the election of clergymen for parliament was prohibited. Finally, the Roman Catholic Church was not only repelled from primary education and jurisprudence but also lost jurisdiction with respect to marital affairs. The revision of the Federal Constitution was important in two other respects. First, the powers of the central state (including social policy) were enhanced. Second, and more important, the optional referendum was introduced in 1874. By collecting 30,000 (today 50,000) signatures, every parliamentary bill or decree is automatically subject to a (binding) referendum. Yet, this instrument of direct democracy turned out to be a boomerang for the Liberals, because the Catholics did, in contrast to the situation in 1848, not react with a retreat to their cantonal strongholds, but rather utilised the optional referendum to obstruct federal legislation initiated by the Liberals (Neidhart 1970; Altermatt 1991: 146; Bolliger/Zürcher 2004: 73). To avert political dead-lock, the Catholic-conservatives gained one seat in the federal government in 1891. Once co-optated into government, Catholics adjusted their political strategy and abandoned their strict federalist course in the subsequent years (Wigger 1997: 131). From now on they identified the Social Democrats as their main antagonist, mainly because of the Social-democratic party’s increasing Marxist orientation. In consequence, the Liberals and Catholics, the former intransigent opponents, formed a bourgeois bloc against the Social Democratic Party which was excluded

4 A more centralized Federal Constitution had been rejected in 1872, not at least due to strong Catholic resistance.

183

from the federal government for decades. As a result, no red-black coalition emerged in Switzerland before World War II that could have paved the way for the emergence of a more generous welfare state that one finds in other continental countries (van Kersbergen and Manow in this volume). However, the political left could rely on the people’s initiative by which 50,000 (today 100,000) citizens may demand an alteration of the Federal Constitution. Introduced in the year in which Catholics had entered government, this instrument is typically used by the opposition and pressure groups because it offers bottom-up opportunities of agenda-setting. The political left and the unions immediately utilized this instrument to promote social policy legislation. However, it soon turned out that the people’s initiative was neither a vehicle to challenge the bourgeois political cartel nor an effective bottom-up instrument to promote public policy. The reason is that a double majority is required for adopting a constitutional amendment. Internal cohesion provided, the Catholic cantons once again held strong veto powers. The introduction of proportional representation (PR) in 1918 put an end to the liberal hegemony. PR had been a long-time political aim, both of the Social Democrats and the Catholic-conservatives (organised in the Catholic People’s Party since 1912), and was finally introduced against the backdrop of the war-induced domestic turmoil that culminated in the Olten general strike in 1918. Both parties won seats in the National Council when the first election based on PR was held in 1919. But whereas the Catholic People’s Party received a second seat in the federal government at the expense of the Liberals, the left, despite its considerable parliamentary strength, remained excluded from government. The separation of Protestant peasants from the Liberals in the wake of World War I gave rise to a further change in the partisan complexion of the federal government. Since 1929, the bourgeois bloc in the Federal Council consisted of four Liberals, two Catholics, and one representative from the conservative Peasant, Trade and Citizen’s Party (Bauern-, Gewerbe- und Bürgerpartei5) which had its strongholds in (reformed) German-speaking cantons such as Bern and Zurich. Seeing the peasantry as the well of freedom and inclined to anti-modernist ideas, this party strongly advocated an alli-

5

This party is the predecessor of the Swiss People’s Party (Schweizerische Volkspartei, SVP).

184

Table 1 Partisan Complexion of the Federal Council since 1848 Period 1848– 1891

1891– 1919

1919– 1929

1929– 1943

1943– 1953

1953– 1954

1954– 1959

1959– 2003

since 2003

Liberals

7

6

5

4

3

4

3

2

2

Catholic People’s Party (Christian Democrats)



1

2

2

2

2

3

2

1

Peasant, Trade and Citizen’s Party (Swiss People’s Party)







1

1

1

1

1

2

Social Democrats









1





2

2

Party

ance between peasants and the bourgeois camp (König 1998: 50). It was World War II and the military threat of Nazi Germany that finally opened the doors to government for the left (see table 1). In 1943, the Social Democrat Ernst Nobs was elected Federal Councillor after the party had abandoned its anti-militarist stance and backed warloans in order to reorganize the Swiss army. The lack of a red-black coalition gives a first important clue for the belated take-off of the Swiss welfare state. However, the distribution of power-resources and the patterns of coalition-building are not enough for explaining the development of Swiss social policy. Welfare state building cannot be fully understood without analysing Switzerland’s unique institutional framework and its repercussions on social politics and policies. We have already seen in this section that the optional referendum substantially preconfigures coalition-building. But this is far from being the only impact of political institutions on the developmental trajectory of the Swiss welfare state.

7.3

Institutional Repercussions: Actor Constellations, Power Resources, and Coalition Building

The development of social policy was crucially influenced by the repercussions of the constitutional framework which was preconfigured by the (religious) cleavages in Swiss society. The vertical and horizontal division of power had multiple and mutually reinforcing effects on social politics and policies. Federalism (1) gave rise to a

185

territorial and confessional fragmentation of power resources, (2) reined in the powers of the federal government and bureaucracy, (3) reinforced anti-centralist and antietatist norms and values (thereby matching the preferences of Catholics), whereas direct democracy (4) provided a powerful leverage to defend the status quo. 1. A Weak Federal Government The central government was in several respects extremely weak from the outset. To begin with, the federal government had only limited policy jurisdictions. Section 3 of the revised Federal Constitution of 1874 stipulated that the “cantons are sovereign insofar as their sovereignty is not limited by the Federal Constitution and, as such, exercise all rights which are not entrusted to the federal power”. With the exception of labour law, cantons held all responsibilities in social policy. Any reallocation of powers between different tiers of government was subject to a mandatory referendum. Again, this type of referendum provided the smaller Catholic cantons with strong veto powers to avert a centralization of government. In a similar vein, the federal government was fiscally constrained and therefore not in a position to finance generous social programs on its own. In 1874, the federal government could only rely on revenues from tariffs, whereas direct taxes were exclusively levied by the cantons. A national income tax was introduced for the first time during war-time and on a temporary basis only. Again, any transfer of fiscal powers between different tiers of government had to pass the referendum hurdle. Finally, administrative capacities were only weakly developed at the federal level. The lack of an effective federal bureaucracy not only had a detrimental effect on welfare state building but in the long-run also contributed to the emergence of corporatist interest mediation, because the federal government had to rely on the know-how embodied in interest organizations of business and labour for drafting and implementing federal bills (Neidhart 1970). 2. Fragmented Interest Formation and Power Resources Swiss federalism constituted a highly competitive arena for the organization of political, religious and economic interests. As a result, the country’s territorial fragmentation erected high barriers against the formation of national interest organizations and political parties (Gruner 1969: 29). Historically, political parties have first emerged at the local and cantonal level and only the big political parties were able to establish

186

national party organizations. To the present day, the national party headquarters are weak, since the centre of gravity of the party system is still located at the cantonal level. Moreover, a party’s ideological orientation often differs considerably from one canton to another so that national parties show weak programmatic coherence and internal cohesion. Social Democrats (1888) and Liberals (1894) were in the vanguard with regard to the formation of national party organizations. In the 20th century, however, both parties suffered from party splits. While the socialists witnessed the emergence of a Communist party in 1921, the Liberals lost support of the Protestant peasantry in the aftermath of World War I. A secular-conservative party representing agrarian and small business interests was founded at the national level in 1937. Beside socio-economic cleavages, confessional divisions shaped the partisan landscape. An interdenominational Christian-democratic party did not emerge in Switzerland (Gruner 1969: 122). The Catholic-conservatives founded the Catholic People’s Party at the national level in 1912 (renamed in Christian Democratic People’s Party in 1971) after several attempts to establish a national party organisation had failed due to territorial fragmentation and particularism. Like in many continental countries, the Catholic People’s Party undoubtedly was a party of religious defence. Its ideological platform was based on four pillars and showed a remarkable stability over time. Catholics demanded (i) the protection of individual rights from state interference in the fields of education, family, marriage and ecclesiastic affairs, (ii) a guarantee by public authorities for the free development of ecclesiastic institutions, associations and parties, (iii) the protection of municipalities and cantons from federal intrusion, and they favoured (iv) subsidiarity and solidarity as guidelines for public policymaking (see Gruner 1969: 117–118). Its Protestant counterpart, the Protestant People’s Party (Evangelische Volkspartei, EVP) was established in 1919. It was strongly linked with the free churches and the Swiss Federation of Evangelic Workers (see below). Based on the Gospel as its programmatic guide, the party was inclined to a strict morality, insisting on family values and demanding help for the aged and handicapped. Nowadays, the party is mobilizing against abortion, the equality of homosexual partnerships under public law, and Sunday work. From the very outset, however, this confessional party lacked broad political support among Protestant voters. Between 1919 and 1945, it never gained more than one percent of the vote at the national level and therefore remained

187

excluded from government. The protestant vote was rather absorbed by the Liberals and the emerging agrarian parties during the inter-war period. National interest organizations faced many obstacles as well. Economic interest groups organized along confessional and territorial lines. This is first and foremost true for unions. Initially, however, it seemed that the labour movement would be able to bridge the gap between different ideologies and denominations. The Swiss Trade Union Confederation (Schweizerische Gewerkschaftsbund, SGB) was established as a non-confessional umbrella organization of trade unions in 1880 and also the Swiss Federation of Workers (Arbeiterbund) consisted of different ideological and confessional sub-organizations. In subsequent years, however, the SGB became more closely affiliated with the Social Democrats, thus provoking fears within the Roman Catholic Church that Catholic workers would become indoctrinated by socialism (Ruf 1991: 66–67; Wigger 1997: 157). When Social Democrats and the SGB increasingly came under the influence of Marxism at the turn to the 20th century, Catholic workers started to establish Catholic workers’ associations and finally founded the Christian-Social Federation of Unions (Christlich-sozialer Gewerkschaftsbund, CSG6) in 1907. Catholic workers herewith followed an appeal by Pope Leo XIII, who had prompted Catholics to establish Christian-oriented interest organizations in his encyclical Rerum Novarum7. Dating back already to the 1860s, the network of Catholic associations fully unfolded during the first two decades of the 20th century. The period between 1920 and 1940 has even been referred to as the golden age of ‘milieuCatholicism’ (Altermatt 1994: 5). The foundation of the Swiss Catholic People’s Association (Schweizerische Katholische Volksverein) in 1905 was a milestone in this respect. This peak association served as an umbrella organization for a broad range of Catholic associations, which not only were based in the traditional Catholic strongholds but also in the so-called Diaspora cantons. In addition, the Swiss Catholic People’s Association promoted and accelerated the formation of the Catholic People’s Party seven years later.

6

Since 1921 Christlichnationaler Gewerkschaftsbund (CNG).

7 Note that leading Swiss Catholic politicians like Caspar Decurtins had a substantial influence on this encyclical.

188

Despite this strong Catholic mobilization, the Swiss Trade Union Confederation clearly outnumbered the Christian-Social Federation of Unions in union density. In 1918, about 177,000 workers were affiliated with the SGB, whereas only some 8,200 wage earners were members of the Catholic counterpart (Holenstein 1993: 329). Similar developments occurred within the Protestant camp. Since Protestants rejected membership in Catholic unions on religious grounds (Ziegler 1939: 66, 94), Protestant workers, under the influence of the free churches (the Methodist Church in particular), established a Protestant union (the Swiss Federation of Evangelic Workers, SVEA) in 1920. Politically, this union was closely affiliated with the Protestant People’s Party. Starting with only 2,244 members in 1920, membership steadily increased and amounted to 13,000 workers in the late 1930s (Ziegler 1939: 170). In addition to the problem of confessional fragmentation, unions also had to overcome occupational segmentation, because white collar workers were organized under the roof of the Swiss Association of White Collar Workers (VSA). Still other occupational groups formed interest organizations outside of the three peak associations SGB, CSG and VSA. Finally, political fragmentation increased when wage earners sympathetic to the Liberals established the Landesverband Freier Schweizer Arbeiter (LFSA) in 1919, which, however, only played a marginal role in subsequent years (see Table 2). As a result, the Swiss labour movement was divided along confessional, political and occupational lines undercutting, in consequence, the power resources of wage earners as well as the coherence of the labour movement. In contrast to the SGB, the confessional unions as well as the LFSA strictly rejected the class-struggle (Ruf 1991: 67– 69). Instead, Catholic unions championed corporatist policies to balance the tensions between labour and capital in a peaceful fashion. In the 1920s, the already deeply entrenched corporatist ideas gained substantial influence within the Catholic People’s Party and finally became a cornerstone of the party’s social and economic platform of 1929 (Weber 1989). The spread of corporatist ideas was further prompted by the encyclical Quadragesimo Anno promulgated by Pope Pius XI in 1931. Moreover, the 1930s witnessed a contagion of Catholic groups by authoritarian ideas. The party’s youth-faction drifted to the extreme right and aimed to overthrow the consti-

189

Table 2 Membership in the trade unions’ peak associations, 1920–1940 (in per cent) Peak association (political/confessional affiliation)

1920

1930

1940

SGB (social democratic)

74.6

68.7

64.4

VSA (Democrats = left wing of the Liberals)

18.4

19.2

18.4

CSG (Catholic)

5.6

8.3

11.1

SVEA (Protestant)

0.7

2.3

3.5

LFSA (liberal)

0.6

1.5

2.6

Total (=100%)

299,575

282,254

329,935

Source: Ruf (1991: 73), amended by the author.

tution by means of a total revision (Altermatt 1991: 158; Weber 1989: 17–18, 141ff). However, this initiative was rejected in a referendum by a broad majority in 1935. Territorial fragmentation and language barriers also led to a great variety of Protestant churches (Ziegler 1939: 8) and distinct state-church relations across cantons that have survived to the present day (see Cattacin et al. 2003). In order to cope with this cantonal fragmentation and to improve ecclesiastic co-operation across cantons, intercantonal organizations were established in the aftermath of World War I. The Swiss Federation of Evangelic Churches was created in 1920 as an umbrella organization of the cantonal reformed churches. The Protestant free churches, mostly organized as associations, also established a peak association8 in 1919, whereas the Swiss Evangelic Alliance provided a common national platform for the reformed cantonal churches and the Protestant free churches. The Catholic Church, certainly more centralized and homogeneous than the Protestant churches, was nonetheless challenged when liberal Catholics established the Christian Catholic Church in reaction to the dogmas of papal infallibility and the papal primacy of jurisdiction enunciated at the First Vatican Council. However, this occurrence did not undercut the strong influence of the Roman Catholic Church since only some 70,000 former Roman Catholics joined the Christian Catholic Church, which later was awarded state recognition as Landeskirche in some cantons.

8 Verband unabhängiger evangelischer Korporationen (nowadays the Verband evangelischer Freikirchen und Gemeinden in der Schweiz).

190

In sum, federalism and the existence of multiple societal cleavages contributed to a territorial and confessional fragmentation of interest groups, political parties and even the Protestant churches. This segmentation inflated the number of actors and interests involved in policy-making and contributed to a substantial fragmentation of the power resources of the labour movement. Confessional fragmentation was not restricted to wage earners since the peasantry was confessionally divided too. Protestant peasants initially had allied with the Liberals. From the 1920s onwards, however, they were increasingly absorbed by conservative agrarian parties. Catholic peasants, by contrast, represented a core clientele within the Catholic-conservatives, facing there, however, a growing labour faction, especially when the Christian-socials workers became officially integrated into the Catholic People’s Party in 1912. Irrespective of this confessional split and the resulting differences in political affiliation, the Swiss peasantry was part of the bourgeois bloc from the outset. Prior to the 1890s, sporadic coalitions between peasants and the (early) labour movement did exist, however. Early labour protection legislation is a case in point. Initiated by the Democratic wing of the Liberals and supported by the labour movement, labour protection was also backed by the peasantry on the basis of strong anti-liberal and antiindustrial sentiments. But this loose coalition soon collapsed when industrialization and urbanisation created a growing and increasingly radicalized working class (König 1998: 27–29). In sum, neither a red-black coalition nor a sort of red-green coalition similar to the Scandinavian countries was formed in Switzerland. Political institutions provide a compelling answer to this pattern of coalition-building. To begin with, there was simply no paramount need to form such coalitions because the optional referendum was an effective vehicle for interest groups and political parties to defend particularistic interests. The most powerful parties in this respect – those which either could launch an optional referendum or just credibly threat to initiate a referendum – were integrated into to the federal government. The crucial point is that Social Democrats did not exploit this veto power because they had no interest to attack reforms initiated by Liberals that aimed at enhancing the federal powers in social and economic affairs. Exactly the opposite was true for the Catholic People’s Party. Indeed, we will see in section 4 that the Social Democrats supported all pro-welfare state bills subject to a referendum even if the party regarded the proposed social policy measures as

191

totally inadequate. On the other hand, Social Democrats could not utilise the people’s initiative to challenge the bourgeois cartel given the high consensus thresholds required to change the status quo. The requirement of a ‘double-majority’ secured that left-wing initiatives could easily be fended-off by a Catholic veto block. In addition to the multiple veto-points, a strong rivalry between the ‘red’ and ‘black’ unions about the adequate political strategy in social policy-making hampered a black-red coalition. Whereas the red unions championed class-struggle, the Catholic unions aimed to smooth class conflict by means of corporatist policy-making. The conflict over social policy between the ‘red’ and ‘black’ labour movement therefore was mainly a conflict about means, and less about goals. A case in point is the collapse of the Olten general strike of 1918 that resulted from a deteriorating social and economic situation in the wake of World War I. Catholics strictly condemned class struggle and considered the strike as a bolshevist attempt to overthrow the constitutional order (Altermatt 1991: 155–156; Holenstein 1993: 298–301). This strong loyalty to the constitutional order was honoured with a second seat for the Catholic People’s Party in the Federal Council in 1919. Liberals definitely recognized the Catholics as a reliable partner so that the latter ultimately got rid of their image of being Sonderbündler. In consequence, the internal cohesion within the bourgeois bloc increased and it was anti-socialism that kept Liberals and Catholics together. Social Democrats and the ‘red’ unions were branded as vaterlandslose Gesellen (Altermatt 1991: 152). Moreover, the crusade of the Roman Catholic Church against the Social-democratic labour movement peaked in the wake of Olten general strike. In 1920, a declaration (the so-called Bettagsmandat), issued by the Catholic episcopacy, openly condemned Catholics members of socialist workers’ organizations and/or the Social Democratic Party (Spieler 1994: 262–263; Holenstein 1993: 374). Hence, Switzerland fits into the general European pattern emphasized by Thomas Ertman in this volume: Once the fierce conflict between the Liberals and the Roman church and its political allies was settled and the Social Democrats became strong – liberal parties and Catholic parties often formed coalitions against the ‘socialist challenge’.

192

3. Political Strategies and Values: Anti-Centralism and Anti-Etatism According to Paul Pierson, federalism significantly changes the policy preferences and political strategies of social actors. In addition, political or ethnic minorities will attempt to shift policymaking towards whatever arena seems most favourable to their interests so that “struggles for increased political and economic power are likely to become intertwined with questions of jurisdictional control over social policy” (Pierson 1995: 450, 454). Catholics’ strategy to build a counter-culture at the cantonal level is a case in point. The retreat of Catholics from the federal arena in the wake of the events of the 1840s was a strategic response to the secular Federal Constitution imposed by the Liberals. A federal polity strongly based on the idea of local autonomy and the associated devolution of power provided the institutional shelter that allowed Catholics to maintain their traditions, to preserve the influence of the Roman Church in Catholic areas and to create a Catholic counter-culture at the cantonal level. More specifically, Catholics considered the cantons as bulwarks against modernization and its concomitants like big government, bureaucratization and secularization. Since they mostly were concentrated in rural areas and showed strong leanings to the ancient order with its enshrined privileges of the Roman Church, Catholics opposed industrialization and were highly sceptical towards new developments in science and technology. In addition, big government emphasized by the Radical and Democratic wing of the Liberals was perceived as an engine of centralization and thus seen as a threat to traditional Catholic norms and the Catholic way of life. Moreover, the economic backwardness of the Catholic periphery and its economic dependency on the more industrialized liberal centres reinforced anti-modernist and anti-centralist attitudes (Altermatt 1991: 58). Catholics therefore sought to gain political control at the cantonal level in order to protect the Catholic sub-society from intrusion by the Liberals who then controlled the federal arena. Since under the Federal Constitution of 1848 the cantons enjoyed almost exclusive jurisdiction in affairs such as education and welfare, Catholics exploited local policy autonomy to protect the strong influence of the Roman Catholic Church on education and marital affairs, at least until the constitutional revision in 1874. Later, in the 1930s, when corporatist ideologies gained more and more importance against the backdrop of a severe economic and political crisis, some Catholics once again preferred a stronger role for the cantons in social benefit provision. Catholics

193

suggested also a reallocation of social policy responsibilities from the state to corporatist arrangements (Weber 1989). However, anti-centralism and the rejection of big government were not exclusively Catholic sentiments, since similar values were shared by liberal groups in Frenchspeaking cantons who feared a majorization by a German-speaking elite and therefore strongly celebrated federalism and cantonal autonomy. This not only holds true for the Liberal-Conservatives, who were mainly backed by Protestant voters in Latin cantons (Rimli 1951: 145–159), but also for many French-speaking Liberals who strongly opposed centralism and big government (Zimmermann 1948; Gruner 1969). Together with French-speaking Catholics, these groups formed a powerful alliance which attempted to defend the sovereignty and autonomy of the cantons in social and fiscal affairs. 4. Direct Democracy The most effective vehicle for all these anti-etatist and federalist groups was the optional referendum which, in general, provides a powerful leverage to defend the status quo (Linder 2005). This instrument has been frequently utilised by interest organizations of business and labour, albeit for completely different motives. The trade unions and Social Democrats successfully used the optional referendum to oppose welfare state retrenchment, something that occurred for the first time during the inter-war period. Interest organizations of business and peasants relied on this instrument to frustrate welfare state building: 5 out of 8 federal bills aiming at expanding social security were rejected between 1874 and 1945 (table 3). Table 3 Social Policy Related Referendums in Switzerland, 1874–1945 Impact of bill/initiative on social policy Referendum type

Total number

Expansive (rejected)

initiated by (number)

Restrictive (rejected)

initiated by (number)

People’s initiative

3

3 (3)

Trade unions, Social democrats (3)

none



Optional referendum

10

8 (5)

Interest groups of business and peasants, bourgeois parties (10)

2 (2)

Trade Unions, Social Democrats (2)

Note that the optional referendum was introduced in 1874, whereas the people’s initiative was established in 1891. Source: Moser/Obinger (2006).

Given the exclusion of Social Democrats from the federal government until 1943, the party as well as the red unions launched several people’s initiatives to promote the

194

introduction of social programs. Three people’s initiatives were subject to a referendum and all of them were rejected by the voters (Table 3). Overall, Table 3 reveals a strong status quo bias of direct democracy. Vested interest organizations of business and peasants as well as right-wing parties successfully contested new federal social policy legislation. For example, the introduction of health and accident insurance (1900), old age pensions for civil servants (1891) and a general public pension scheme (1931) failed due to a referendum. Once in place, however, the left and trade unions have been successful in defending the status quo in social affairs. Both referendums initiated against propositions by the government to curtail achieved social standards were rejected. These mechanisms and the resulting impact on social policy-making will be illustrated in more detail in the next section which analyzes the process of early welfare state formation between 1874 and 1945.

7.4

Welfare State Consolidation, 1874–1945

So far I have concentrated on the impact of religious cleavages on state building and the ways in which the constitutional setting has influenced actor constellations, the power resources of collective actors, coalition building and political strategies. This section examines how these institutionally mediated effects have shaped social policy making between 1874 and 1945. Particular attention is paid to labour protection, health and accident insurance, old age pensions and income support for families and the unemployed. According to the Federal Constitution of 1874 the cantons were responsible for almost all matters of social policy. In addition, they controlled the most important tax powers, whereas the federal government could only rely on revenues from tariffs. In the absence of federal jurisdiction, social security programs were established both by the cantons, cities and municipalities and by friendly societies, trade unions and entrepreneurs (Zacher 1899; Reichesberg 1906; Krumbiegel 1913; Gruner 1988; Tschudi 1989). However, many cities and cantons remained inactive, whereas in still other cantons, social policy initiatives were frustrated by the use of the referendum procedure. Public unemployment insurance failed to find public consent in the city of Basle in 1900, old age pensions were rejected in Geneva in 1910 and the introduction of

195

minimum wages failed in both Berne (1898) and Zurich (1899). Beside the public programs established in cities and cantons, workers’ associations of both denominations and entrepreneurs set up compensation funds. The number of funds increased from 652 in 1865 to 2,006 in 1903, covering almost 500,000 people at the turn of the twentieth century. In sum, local policy pre-emption has given rise to a patchwork of decentralised social security arrangements, which differed markedly from one another in terms of funding, organisation and levels of benefit provision. Given the central state’s lack of jurisdiction in social and fiscal affairs, welfare state building, similar to state and nation building in general, thus was a bottom-up process. Shifting social policy competencies and fiscal powers to the national level, however, had to overcome the inbuilt rigidities imposed by the Swiss constitution and also had to cope with the widespread anti-centralist attitudes in Swiss society (Saxer 1951). It is thus hardly surprising that the process of welfare state building between 1874 and 1945 moved on from defeat to defeat. Interestingly, however, this process commenced with a remarkable success.

7.4.1 A Success Story: Labour Protection Under section 34 of the Federal Constitution of 1874 the federal government was only empowered to regulate working conditions in factories. The Liberal federal government immediately made use of federal jurisdiction in this field and enacted the Federal Factory Act in 1877 that made Switzerland a European pioneer in terms of labour protection. The Liberals therefore not only promoted market-building (recall that the Federal Constitution of 1848 removed trade barriers between the cantons), but also initiated the Polanyian counter-movement through early efforts of social policy legislation. This puzzle can only be explained if one takes into account the considerable ideological heterogeneity within the liberal movement (Gruner 1969). Nevertheless, other factors also have contributed to this success. The federal competence for legislation certainly was a major cause of why factory legislation could be implemented so early and speedily. Moreover, the federal government could rely on prior cantonal legislation. In the 1860s, the more industrialized cantons such as Argovia, Glarus, Schaffhausen, and Basle established Factory Acts that served as pacemakers for the Federal Factory Act. Last but not least, labour protection is a pure regulatory policy and does therefore not require fiscal resources on a large scale.

196

Nevertheless, the Federal Factory Act, which was also supported by Catholics and peasants on the basis of anti-industrial and anti-liberal attitudes, was attacked by an optional referendum launched by interest groups of business and industry. However, the bill was approved by 51.5 percent of voters in 1877. Based on the Federal Factory Act, the working day for industry was restricted to eleven hours and child labour was prohibited. The Factory Act was replaced by an improved bill in 1914, and the eighthour working day was introduced in 1918. Soon thereafter, however, the bourgeois government decided to increase the working day again. This caused severe tensions both within the Liberal party (Zimmermann 1948) and the Catholic People’s Party. The latter was plagued by internal conflicts between the agrarian faction and the party’s Christian-social wing. Nevertheless, it was the red labour movement that launched an optional referendum against the lifting of working time. The referendum committee was able to collect 200,000 signatures, equivalent to 20 percent of the population entitled to vote (Giovanoli 1932: 410). The plebiscite was held in 1924 and voters rejected the proposal of the federal government. This referendum marks the first time that a federal bill aimed at welfare state retrenchment was rejected by the Swiss voters. Yet, the referendum also empowered business organizations and the political right to attack progressive labour protection legislation. One example is the failure of the federal bill on collective labour law (Bundesgesetz über die Ordnung der Arbeitsverhältnisse). Although the bill was passed unanimously by parliament, the Liberal-Conservative Party and business organizations successfully launched a referendum against the bill in 1920. Fears of federal intrusion in cantonal affairs and resistance against a mounting bureaucracy were the main reasons why the bill was attacked by these groups (Zimmermann 1948: 59). In addition to the regulation of working time and working conditions, a Federal Liability Act was enacted for factory workers in 1881 which was subsequently extended to other groups of workers. It obliged employers to pay compensation to injured workers. Yet, the Federal Liability Act soon turned out as dysfunctional because many injured workers did not receive compensation or did not claim it at the courts in order to avoid dismissal.

197

7.4.2 From Liability Legislation to Health and Accident Insurance In the late 1880s, the malfunctioning of the Federal Liability Act together with the exemplary effects of German and Austrian social insurance legislation convinced the Liberal federal government to adopt a new approach to social policy by advocating health and accident insurance (cf. Maurer 1980: 780–81). Ludwig Forrer, liberal National Councillor, who later also was elected Federal Councillor, vindicated the insurance approach with the slogan: “Employer’s liability means quarrel, insurance stands for peace!”9 However, federal intervention in this field faced various problems. Given the federation’s lack of authority to act in social policy, the legislation process was necessarily split into two stages. The first step was to pass constitutional amendments empowering the federal authorities to undertake social security legislation. This reallocation of jurisdiction proved to be difficult because each amendment was subject to an obligatory referendum, and the Swiss have always been hesitant to entrust the federal government with new powers (Saxer 1951: 222). In particular this was true for the Catholics and liberal groups based in the Latin cantons. Notwithstanding these difficulties, voters approved the constitutional amendment that empowered the federal government with the authority to act in health affairs in 1890. However, this did not automatically lead to a new federal policy. In order for this to occur, the government needed to adopt so-called ‘implementing legislation’. Soon after the constitutional amendment was passed, Ludwig Forrer drafted a bill, the so-called Lex Forrer, which was strongly influenced by Bismarckian health and accident insurance (Krumbiegel 1913). Forrer suggested a mandatory and predominantly contribution-based scheme of workers’ insurance, which would be mainly organized around public and semi-public sickness funds. A public accident insurance company was to be put in charge of mandatory accident insurance. During the deliberations of the bill, leading Catholic corporatists like Josef Beck and Caspar Decurtins suggested that co-operations (Berufsgenossenschaften) should be put in charge of health insurance. This proposal, however, was incompatible with liberal ideas and

9

Quoted from Maurer (1980: 780).

198

Ludwig Forrer consequently replied that the era of the guilds is definitely over (Rutishauser 1935: 164). The bill was finally backed by all major parties (including the Catholic Party) and passed both houses of parliament almost unanimously. Yet, existing private sickness funds argued that competition by public sickness funds would jeopardize business. Industry, small business and farmers opposed increasing non-wage labour costs, while French-speaking liberals rejected compulsory insurance. In contrast to the Social Democrats, the Swiss Federation of Workers branded the bill bureaucratic and disapproved the state supervision of self-administered mutual compensation funds. It therefore launched a people’s initiative in 1893 that proposed a national health service (free medical services) funded by revenues from a tobacco monopoly. However, this initiative failed because the organization was unable to collect the 50,000 signatures necessary for a plebiscite. Business groups and right-wing forces, branding the bill both as collectivist and centralist, were more successful because they were able to collect sufficient signatures to launch an optional referendum against the Lex Forrer. The bill was distributed among the voters in the run-up to the plebiscite. Since it consisted of 400 articles and had a shipping weight of two pounds (Gruner 1988: 640), the Federal Health and Accident Insurance Act became a synonym for federal bureaucracy and big government. König (1998: 32) refers to the Lex Forrer as the first federal bill subject to a referendum that simply was not graspable for ordinary citizens. Not surprisingly, the Swiss rejected the Federal Health and Accident Insurance Act in 1900. The most important reason for its defeat was the compulsory nature of the insurance, which was regarded as the beginning of state-socialism (Maurer 1980: 783). Rutishauser (1935: 167) argues that the referendum beared resemblance to a tax plebiscite. In addition, the resistance of private sickness funds as well as federalist-motivated objections to increased federal expenditure, bureaucracy and state monopolies – attitudes shared by many French-speaking Catholics and Liberals – contributed to the bill’s defeat (Funk 1925: 66–67; Rutishauser 1935: 167; Rimli 1951: 194). Six years after the bill was rejected, a second bill was drafted. The bourgeois federal government was fully aware of the interests backing the referendum (Bundesblatt 1906, vol. VI, p. 252) and thus made far-reaching concessions to the opponents of the Lex Forrer. Employers no longer had to pay contributions and the insured had to pay per capita premiums instead of income-related contributions. Compulsory insurance

199

was abandoned, although the cantons were empowered to introduce compulsory insurance for certain groups. The federal government confined itself to providing positive incentives. The existing private or corporatist sickness funds received federal subsidies if they guaranteed the minimum standards outlined in federal legislation. With respect to accident insurance, employers were released from contributory payments for non-occupational accidents and farmers as well as small business were excluded from coverage. However, in order to prevent ‘cream skimming’ by private insurance companies, the monopoly of the Swiss Accident Insurance Institute was preserved. This gave rise to a second referendum which again was launched by business interest organizations (Rimli 1951: 198). In a tactical manoeuvre aimed to mitigate Catholic resistance against the bill, the federal government decided to place the accident insurance company in Lucerne, a Catholic stronghold in central Switzerland. This strategy turned out to be successful and the redrafted bill was adopted by voters in 1911. This brief account of the adoption of health and accident insurance shows how policy change was not only delayed for nearly two decades, but also fundamentally affected by direct democracy. Early policy pre-emption at the local level, an effect mainly stemming from federalism, was a major impediment to achieving far-reaching policy changes. Given the ‘patchwork quilt’ nature of local social security schemes, the federal government could not act unilaterally and had a limited degree of freedom to replace these arrangements. Since many political and economic interests had already crystallised around the existing decentralised social programs, the providers of these programs, anti-centralist and anti-etatist groups and employers were reluctant to accept federal policy intervention. The defeat of the Lex Forrer crucially shaped subsequent policies as it opened the way for a liberal framework law with universal coverage instead of an imitation of mandatory and class-based health insurance along Bismarckian lines (see Immergut 1992). Employers were exempted from contributory payments so that the cost burden was shifted to the insured which now had to pay per capita premiums and not the earnings-related contributions initially proposed. In addition, the fate of the Lex Forrer crowded out the option of a class-based workers’ insurance for the future because the optional referendum turned out as a powerful vehicle to overthrow alleged privileges for specific occupational groups. This was already evident in 1891, when the Swiss Peasants’ Association had successfully initi-

200

ated an optional referendum against very moderate old age and disability benefits for federal civil servants (cf. Bundesblatt 1890 IV: 301–302). Almost all signatures backing this referendum were collected in Catholic cantons. The widespread mistrust in bureaucracy as well as the special treatment of a particular occupational group put a stop to the bill (Funk 1925: 51).

7.4.3 Income Support for the Unemployed Local policy pre-emption also impeded and delayed the adoption of unemployment insurance. Delay can be explained by several factors: the absence of federal jurisdiction, strong policy feedback from local social security arrangements, institutional factors, and a lack of political will. Already in 1893, the Social democrats had launched a people’s initiative to improve the position of the unemployed, either through federal unemployment insurance or by subsidising existing private funds from the general budget. Yet, this initiative was decisively rejected by voters in 1894. In subsequent years, the bourgeois federal government blocked federal unemployment insurance on cost grounds. Given the politics of non-decision at the federal level, the cantons began to provide subsidies to the existing public and private unemployment compensation funds. By 1914, seven cantons had adopted the so-called Ghent system, that is they provided public subsidies to privately run unemployment insurance schemes (Gruner 1988: 767). Initially on a provisional basis (until 1924), the central state stepped in and provided subsidies from 1916 onwards. Severe social tensions in the aftermath of World War I increased the pressure for political action. Given limited federal powers in this field, the Federal Council enacted a framework law in 1924 that anchored the Ghent system at the federal level. At this point, the federal government limited its intervention to regulating minimum standards and to providing subsidies to the existing unemployment funds, expecting that federal subsidies would stimulate the creation of new unemployment funds at the local and cantonal level. However, this system was ill-equipped to cope with soaring unemployment during the Great Depression. Moreover, economic depression fuelled a political crisis. Mistrust of democracy increased and right wing groups launched an initiative to overhaul the constitution. Corporatist ideas as outlined in the encyclical Quadragesimo Anno gained significant influence in the Catholic People’s Party, which

201

finally also supported the people’s initiative demanding a total revision of the Federal Constitution (Weber 1989). The federal government responded to the economic crisis with protectionist measures and imposed austerity policies to curb inflation. Jean Marie Musy, minister of finance from the Catholic People’s Party, was the most prominent exponent of a strict austerity course. Musy’s deflationary policy package included a 10-percent cut of federal employees’ salaries (Weber 1989). While this plan was fully backed by his party, the red unions launched an optional referendum, which was also supported by left-wing Liberals and smaller bourgeois parties. In 1935, a majority of voters rejected the proposed wage cuts. After the successful referendum against the increase in working time, this referendum was a further success for trade unions in defending the status quo. Motivated by this success, the SGB launched a people’s initiative (the so-called Kriseninitiative), which proposed a series of measures ranging from public employment programs to income support for the unemployed. Because of the stateinterventionist nature of this policy package, this initiative was strictly opposed by the bourgeois bloc, including the Catholic People’s Party. Instead, in line with its corporatist re-orientation the party proposed to relieve the state from social and economic responsibilities and suggested to shift them to occupational co-operations (Weber 1989). Like all social-policy related people’s initiatives voters rejected the Kriseninitiative in 1935. Nevertheless, the SGB soon launched a further initiative to improve the situation of the unemployed. However, this once more met with the strong resistance of the federal government. The SGB backtracked after the federal government had announced to pass a constitutional amendment that would empower the federal government with the jurisdiction to enact unemployment insurance (Sigg 1978: 199). Although this amendment was adopted in 1947, it took thirty more years before mandatory unemployment insurance was finally established.

7.4.4 Old age Pensions: The Defeat of the Lex Schulthess At the turn of the twentieth century, both the Liberals and Social Democrats had put the introduction of a general pension scheme on the political agenda. These early initiatives were rejected by the Federal Council on cost grounds (cf. Bundesblatt 1919, vol. IV, p. 38). During and in the wake of the Olten general strike, also the Catholic-

202

Conservative called for the introduction of old age pensions. Although the Olten general strike utterly failed, the immediate post-war years witnessed many social policy activities to cope with the social disruption caused by the war. Against this backdrop, the federal government announced to establish a federal pension scheme. The government had been put under pressure by a constitutional initiative launched by the Liberal National Councillor Rothenberger in 1920 as well as a petition by the Swiss Peasants’ Association that proposed a tax funded universal pension scheme. In response to the latter the government stated that “it needs no explanation that any attempt to raise public revenues on such a scale is infeasible” (Bundesblatt 1919 vol. IV: 115). Instead, the federal government proposed a scheme that involved mandatory old age, survivors’ and invalidity insurance with universal coverage. The inclusion of the whole resident population was clearly a lesson learnt from the fate of the Lex Forrer. Facing the referendum threat, the federal government rejected a class-based insurance by arguing that “the state must not care to which occupation an individual belongs” (Bundesblatt 1929 vol. II: 186). In addition, it pointed out that “those who have pay but do not have a chance to receive benefits often tend to vote against a bill” (Bundesblatt 1924 vol. II: 731–32). Given the fiscal limitations, the scheme was to be financed mainly from contributions supplemented by public grants. Hence, the burden of financing was mainly shifted to employers and employees to bypass the revenue shortage of the federal government. However, the federal government tied both the introduction and the generosity of the program to enhanced federal taxing powers (Bundesblatt 1919, vol. IV, p. 127 and pp. 149–50). Specifically, the federal government demanded new taxes on tobacco and beer, a federal inheritance tax, and the extension of the alcohol monopoly. Facing the recession of the early twenties, parliament cancelled the beer tax to protect the breweries. The enlargement of the alcohol monopoly was rejected at the ballot box in 1923, while the cantons refused to cede jurisdiction of the inheritance tax to the central state. In this situation the federal government had to scale down the proposed project, and, instead, to proceed with a piecemeal development of social insurance (Bundesblatt 1924, vol. II, p. 685). As a result, invalidity insurance was split off from old age and survivors’ insurance to ensure that a minimum federal mandate for legislation could be adopted. After six years of discussion, the constitutional amendment,

203

which empowered the federation to enact old age insurance, passed the obligatory referendum hurdle in 1925. The implementing bill, the so-called Lex Schulthess named after the liberal Federal Councillor Edmund Schulthess, was submitted to pre-parliamentary consultation in 1928. Owing to insufficient tax revenues, benefits were watered down compared to the first draft of 1919. Contributions as well as benefits were flat-rate and significantly lower than the benefits provided by the existing private and public occupational pensions funds10. Like the Lex Forrer, the bill was easily approved by the two houses of parliament. Social Democrats, Liberals, and a majority of the Catholic People’ Party voted for the bill (Sommer 1978: 151). However, liberal-conservative groups and right-wing Catholics launched a referendum against the Lex Schulthess in 1931. Based in French-speaking cantons, these groups strongly opposed the compulsory nature of the insurance, which was considered to be etatist, centralist and a step towards socialism (Rimli 1951: 276; Sigg 1978: 163; Binswanger 1986: 21). Even JeanMarie Musy openly supported the referendum (de Nicolo 1962: 136) and thus disavowed the collegial Federal Council. Though Musy basically supported old age pensions, he strongly favoured cantonal solutions and thus opposed any kind of centralist and bureaucratic approaches. Moreover, the encyclical Quadragesimo Anno rendered support to the opponents of the Lex Schulthess insofar as the proposed cantonal policy jurisdiction was in line with the principle of subsidiarity. In addition, Quadragesimo Anno provided a programmatic platform for a ‘third way’ between socialism and liberalism by justifying a corporatist make-up of society and a greater role for co-operations in policy-making. By referring to Quadragesimo Anno, Abbé André Savoy, the leader of the French speaking Christian-socials and prominent advocate of corporatist ideas (Weber 1989: 50), rigorously opposed public old age insurance. In his view, insurance was to be the exclusive matter of individual, private or corporatist initiative (Lepori 1994: 59). In the run-up to the plebiscite the French-

10 Given the long period of non-decisions in public pension policy, old age provision was pre-empted by private and occupational initiatives. Occupational pension funds date back to the late 1880s and experienced a substantial expansion in the 1920s, when approximately 150 new pension funds were established each year. In 1925, about 17.4 per cent of the labour force was already covered by occupational pension schemes. Moreover, about 800,000 persons had life insurance at that time. Hence, occupational pensions and private insurance were well established when the Lex Schulthess was submitted to parliament for deliberation (see Lengwiler 2003).

204

speaking Catholic newspaper Le Pays mobilized readers against the Lex Schulthess in the following way: “If you want stick to the corporatist ideal emphasized by the recent papal encyclical, then vote against the Lex Schulthess!”11 The Catholic camp was however divided. Whereas the French-speaking corporatists mobilized against the bill, the German-speaking and more pragmatic representatives of the Catholic Party supported the Lex Schulthess (Lepori 1994; Weber 1989). Beyond Catholic resistance, employees enrolled in occupational pension plans opposed the bill in order to avert ‘double insurance’ since the Lex Schulthess was not designed to replace occupational pensions (Lengwiler 2003). The Social Democrats supported the bill in order to salvage a solution of some kind, while the Communists joined the referendum committee because they considered the benefits on offer to be entirely inadequate. As a consequence, the Lex Schulthess was attacked by a curious alliance consisting of anti-centralist liberal and Catholic groups on the one hand and the far left on the other (Zimmermann 1948: 124). To provide a counter-proposal to the Lex Schulthess, Catholic-corporatist and right-wing groups led by Abbé Savoy launched a people’s initiative (Studer 1998: 175), which demanded the extension of means-tested benefits controlled by the cantons. This move forced the voters to choose between social insurance and social assistance. A majority favoured the latter and rejected the Lex Schulthess on 6 December 1931. Not surprisingly, the bill was rejected in the Catholic French-speaking regions where it was branded as “dangereuse loi d´inspiration germanique et socialiste” (Sommer 1978: 157–58). The defeat of a public pension scheme had two major consequences: first, the introduction of old age and survivors’ insurance was delayed until 194812, while invalidity insurance was postponed even longer, until 1960. Second, the defeat of the Lex Schulthess had long-lasting implications for the public/private divide of benefit provision. Social policy in the 1930s followed the trajectory outlined by the Catholic and liberal proponents of the people’s initiative. During the Great Depression, the federal government provided grants to social assistance programmes and subsidized charitable

11 Quoted from Lepori (1994: 61), my translation. 12 A second bill on public pensions that was passed by parliament in 1946 was also attacked by liberal and Catholic groups (Obinger 1998a: 167–170). However, about 80 per cent of voters approved the bill in 1947.

205

foundations such as Pro Senectute and Pro Infirmis. In addition, the lack of public pensions further stimulated the growth of privately organized pension schemes. The number of employees covered by occupational pension schemes increased from 258,000 in 1925 to 1,342,000 in 1966. From 1931 on, it therefore became increasingly evident that public old age insurance would never be able to replace the manifold forms of pension provision that preceded state provision. Hence, the referendum in 1931 was a critical step in the progress towards the Swiss multi-pillar approach in pension policy that emerged in the 1970s. Once again, the optional referendum had decisively preconfigured the patterns of social security schemes.

7.4.5 Family Policy: A Catholic Domain Since the late 1920s, the Catholic People’s Party increasingly advocated public measures to improve the situation of families (Wäger 1944; Gernet 1994: 176). Since neither the federal government nor the cantons provided income transfers to families before the Second World War13, the party proposed the introduction of family wages in line with the male-breadwinner model and demanded public income support to families with dependent children. Because the federal government did not respond to these claims, the party decided to launch a people’s initiative (the so-called Familieninitiative) in 1941. Facing this initiative, the federal government worked out a counterproposal in 1944, suggesting a constitutional amendment to empower the federal government with policy jurisdiction in family affairs. Hereupon the Catholic People’s Party withdrew its initiative and the government’s counter-proposal was approved by a majority of voters in 1945. As a result of this plebiscite, a new constitutional provision was adopted that empowered the federal government to provide family allowances and maternity benefits. During the post-war period, however, the federal government only introduced family allowances for peasants. The corresponding bill of 1952 was by itself mainly a legacy of World War II. In 1944, the federal government used its emergency powers to introduce family allowances for mountain farmers and agricultural employees in order to secure a sufficient supply with food during wartime and to stop migration into cities (Maurer 1980: 798; Tschudi 1989: 34).

13 The only exception from this rule applied to members of the federal civil service.

206

With the exception of family allowances for peasants, income support for families remained a cantonal responsibility to the present day. In November 2006, however, the Swiss approved a federal framework bill in a referendum that will lead to a greater harmonization of cantonal cash benefits offered to families. Furthermore, maternity insurance was only adopted in 2004 after three attempts to introduce maternity benefits failed in a referendum.

7.5

Conclusion

Religion has undoubtedly influenced welfare state building in Switzerland. First of all, the religious cleavage in Swiss society had a profound impact on state-building. Although the Swiss federation created by the Liberals was strictly laïcist, the Federal Constitution of 1848 instituted a division of power that granted the federal government only limited responsibilities and guaranteed cantons substantial policy autonomy as well as considerable influence on national decision-making. Catholics used the political autonomy guaranteed by federalism to build a territorially-bounded Catholic sub-society in their cantonal strongholds. Catholic anti-modernism stood in sharp contrast with Liberals’ efforts to create a secular state. In addition, direct democracy provided Catholics with considerable veto powers to defend cantonal policy autonomy and to oppose big government. To mitigate the veto powers of the Catholic-conservatives resulting from the optional referendum introduced in 1874, the party soon was integrated into the federal government. Social Democrats, in contrast, were politically marginalized by an increasingly diversified bourgeois bloc that controlled the federal government during almost the entire period of observation. Hence, and in contrast to many continental countries, there was no red-black coalition in Switzerland before 1943 that could have supported welfare state development. It was argued in this chapter that this pattern of coalition-building can be attributed to the multiple veto points enshrined in the country’s polity which itself was strongly shaped by the religious cleavage. Whereas federalism and the optional referendum represented core elements within the strategic repertoire of Catholics, the left did neither benefit from federalism nor from direct democracy. Federalism is by its very nature an obstacle against ideas of big government, while the left did not utilize the optional referendum as long as the Liberals were inclined to strengthen the powers of

207

the federal government, including social policy. The situation only changed when the bourgeois bloc started to impose retrenchment policies during the inter-war period. A black-red coalition also failed due to the sharp ideological conflicts between the left and the Catholic People’s Party. Furthermore, the confessional fragmentation of the labour movement and the resulting antagonism between red and black unions has undercut the coherence and power resources of wage earners. The conflict between both camps mainly stemmed from distinct views about the adequate political strategy for promoting and organising social policy. Whereas the Social Democratic unions celebrated the class struggle, the Christian unions championed corporatism as a sort of third way between capitalism and socialism. The dominance of centre-right parties and political deadlock caused by the institutional veto points associated with federalism and direct democracy are the most important factors accounting for the belated development of the Swiss welfare state and its liberal make-up. The veto power associated with the optional referendum did not only pave the way for a coalition between Liberals and Catholics, but also decisively influenced the timing of welfare state consolidation because of the strong status quo bias resulting from this type of referendum. Furthermore, the referendum has considerably preconfigured welfare state patterns ever since the idea of introducing a class-based social insurance was rejected by the people. Instead, the referendum favoured social policies where costs and benefits are symmetrically distributed. Hence, the referendum contributed to the introduction of universal insurance schemes and, in consequence, to very distinct patterns of stratification and status reproduction one finds in other continental welfare states. Finally, many social programs like unemployment and health insurance remained voluntary, whereas the level of benefits was low due to the traditionally limited fiscal resources of the federation, which additionally were constrained by direct democracy.

208

References Altermatt, Urs, 1972: Der Weg der Schweizer Katholiken ins Ghetto. Die Entstehungsgeschichte der nationalen Volksorganisation im Schweizer Katholizismus 1848–1919, Zürich/Köln. Altermatt, Urs, 1991: Katholizismus und Moderne. Zur Sozial- und Mentalitätsgeschichte der Schweizer Katholiken im 19. und 20. Jahrhundert, Zurich: Benziger, 2nd edition. Altermatt, Urs, ed., 1994: Schweizer Katholizismus zwischen den Weltkriegen 1920–1940, Freiburg: Universitätsverlag Freiburg. Armingeon, Klaus, 2001: Institutionalizing the Swiss Welfare State, in: West European Politics 24: 145–168. Banting, Ketih, 2005: Canada: Nation-building in a federal welfare state, in: Herbert Obinger, Stephan Leibfried, Francis G. Castles, eds., pp. 89–137. Binswanger, Peter, 1986: Geschichte der AHV, Zurich: Pro Senectute. Bolliger, Christian; Zürcher, Regula, 2004: Deblockierung durch Kooptation? Eine Fallstudie zur Aufnahme der Katholisch-Konservativen in die schweizerische Landesregierung 1891, in: Swiss Political Science Review 10: 59–92. Bütschi, Danielle; Cattacin, Sandro, 1993: The Third Sector in Switzerland: The Transformation of the Subsidiarity Principle, in: West European Politics 16: 362–79. Castles, Francis G., 1994: On Religion and Public Policy: Does Catholicism Make a Difference?, in: European Journal of Political Research 25: 19–40. Cattacin, Sandro; Famos, Cla Reto; Duttwiler, Michael; Mahning, Hans, 2003: Staat und Religion in der Schweiz. Anerkennungskämpfe, Anerkennungsformen, Eine Studie des Schweizerischen Forums für Migrations- und Bevölkerungsstudien (SFM), Bern. De Nicolo, Marco, 1962: Die Sozialpolitik des Schweizerischen Gewerkschaftsbundes (1880– 1960), Winterthur: P.G. Keller. Deutsch, Karl W., 1976: Die Schweiz als paradigmatischer Fall der Integration, Bern: Haupt. Funk, Friedrich Otto, 1925: Die eidgenössischen Volksabstimmungen von 1874 bis 1914, Bern: Gustav Grunau. Gernet, Hilmar, 1994: Der Schweizerische Katholische Volksverein im Spannungsfeld von Kirche und Partei 1930–1960, in Urs Altermatt, ed., pp. 167–181. Giovanoli, Fritz, 1932: Die eidgenössischen Volksinitiativen 1874–1931, Schweizerische Zeitschrift für Statistik und Volkswirtschaft 68: 404–429. Gould, Andrew C., 1999: Origins of Liberal Dominance. State, Church, and Party in Nineteenth-Century Europe, Ann Arbor: University of Michigan Press. Gruner, Erich, 1969: Die Parteien in der Schweiz, Bern: Francke. Gruner, Erich (ed.), 1988: Arbeiterschaft und Wirtschaft in der Schweiz, Zurich: Chronos Holenstein, Dieter, 1993: Die Christlichsozialen der Schweiz im Ersten Weltkrieg. Entwicklung der christlichsozialen Organisationen und ihre Stellung in der schweizerischen Arbeiterbewegung und der katholischen Sondergesellschaft 1914–1920, Freiburg: Universitätsverlag Freiburg. Immergut, Ellen M., 1992: Health Politics: Interests and Institutions in Western Europe, Cambridge.

209

Kersbergen, Kees van, 1995: Social Capitalism. A Study of Christian Democracy and the Welfare State, London/New York: Routledge. Kölz, Alfred, 1998: Der Weg der Schweiz zum modernen Bundesstaat. Historische Abhandlungen, Chur/Zurich: Rüegger. König, Mario, 1998: Politik und Gesellschaft im 20. Jahrhundert. Krisen, Konflikte, Reformen, in: Hettling, Manfred; König, Mario; Schaffner, Martin; Suter, Andreas, Tanner, Jakob, 1998: Eine kleine Geschichte der Schweiz. Der Bundesstaat und seine Traditionen, Frankfurt a.M.: Suhrkamp, pp. 21–90. Krumbiegel, Kurt, 1913: Die schweizerische Sozialversicherung insbesondere das Krankenund Unfallversicherungsgesetz vom 13. Juni 1911 verglichen mit der entsprechenden deutschen Gesetzgebung, Jena: G. Fischer Lengwiler, Martin, 2003: Das Drei-Säulen-Konzept und seine Grenzen: Private und berufliche Altersvorsorge in der Schweiz im 20. Jahrhundert, in: Zeitschrift für Unternehmensgeschichte 48: 29–47. Lepori, Antonio, 1994: La réception de Quadragesimo Anno en Suisse romande, in: Urs Altermatt, ed., Schweizer Katholizismus zwischen den Weltkriegen 1920–1940, pp. 57–68. Lijphart, Arend, 1999: Patterns of Democracy, New Haven: Yale University Press. Linder, Wolf, 2005: Schweizerische Demokratie, Bern: Haupt, 2nd edition. Manow, Philip, 2002: The Good, the Bad, and the Ugly. Esping-Andersens Wohlfahrtsstaatstypologie und die konfessionellen Grundlagen des westlichen Sozialstaats, in: Kölner Zeitschrift für Soziologie und Sozialpsychologie 54: 203–225. Maurer, Alfred, 1980: Länderbericht Schweiz, in: Peter A. Köhler, Hans F. Zacher, eds., Ein Jahrhundert Sozialversicherung in der Bundesrepublik Deutschland, Frankreich, Großbritannien, Österreich und der Schweiz, Berlin: Duncker & Humblot, pp. 731–833. Mooser, Josef, 1998: Eine neue Ordnung für die Schweiz: Die Bundesverfassung von 1848, in: Brigitte Studer, ed., 1998: Etappen des Bundesstaates. Staats- und Nationsbildung der Schweiz, 1848–1998, Zürich: Chronos, pp. 45–62. Moser, Julia; Obinger, Herbert, 2006: Schlaraffenland auf Erden? Auswirkungen von Volksentscheiden auf die Sozialpolitik, in: Uwe Wagschal, Markus Freitag, eds., Direkte Demokratie – Bestandsaufnahmen und Wirkungen im internationalen Vergleich, Hamburg/Münster: LIT(in press). Neidhart, Leonhard, 1970: Plebiszit und pluralitäre Demokratie. Eine Analyse der Funktion des schweizerischen Gesetzesreferendums, Bern: Francke. Nöthiger-Strahm, Christine, 1981: Der deutsch-schweizerische Protestantismus und der Landesstreik von 1918. Die Auseinandersetzung der Kirche mit der sozialen Frage zu Beginn des 20. Jahrhunderts, Bern: Lang. Obinger, Herbert, 1998a: Politische Institutionen und Sozialpolitik in der Schweiz, Frankurt a.M.: Lang. Obinger, Herbert, 1998b: Federalism, Direct Democracy, and Welfare State Development in Switzerland, in: Journal of Public Policy 18: 241–263. Obinger, Herbert; Leibfried, Stephan; Castles, Francis G., eds., 2005: Federalism and the Welfare State. New World and European Experiences, Cambridge: Cambridge University Press.

210

Pierson, Paul, 1995: Fragmented Welfare States: Federal Institutions and the Development of Social Policy, in: Governance 8: 449–478. Reichesberg, Naum 1906: Die Arbeitslosenversicherung in der Schweiz. Bern: Scheitlin, Spring & Cie. Rimli, Bruno, 1951: Sozialpolitische Ideen der Liberal-Konservativen in der Schweiz (1815– 1939). Zurich: Europa Verlag Zurich. Ruf, Hans, 1991: Spitzenverbände der schweizerischen Arbeitnehmerorganisationen, in: Robert Fluder; Heinz Ruf, Walter Schöni; Martin Wicki, eds., Gewerkschaften und Angestelltenverbände in der schweizerischen Privatwirtschaft, Zurich: Seismo, pp. 63–200. Rutishauser, Hans, 1935: Liberalismus und Sozialpolitik in der Schweiz. Univ. Diss. Lachen: Gutenberg. Saxer, Arnold, 1951: Die Entwicklung der Sozialversicherung in der direkten Demokratie, in: Walter Rohrbeck, ed., Aus der Privat- und Sozialversicherung des In- und Auslandes. Berlin: Duncker & Humblot: 222–251. Sigg, Oswald, 1978: Die eidgenössischen Volksinitiativen 1892–1939. Bern: Francke. Sommer, Jürg H., 1978: Das Ringen um soziale Sicherheit in der Schweiz. Chur: Rüegger. Spieler, Willy, 1994: Zur Marginalisierung der politischen Linken in der katholischen Kirche, in: Urs Altermatt, ed., Schweizer Katholizismus zwischen den Weltkriegen 1920–1940, pp. 253–278. Studer, Brigitte, 1998: Soziale Sicherheit für alle? Das Projekt Sozialstaat, in: Brigitte Studer, ed., 1998: Etappen des Bundesstaates. Staats- und Nationsbildung der Schweiz, 1848–1998, Zurich: Chronos, pp. 159–186. Tschudi, Hans Peter, 1989: Entstehung und Entwicklung der schweizerischen Sozialversicherungen, Basel: Helbling & Lichtenhahn. Wäger, Franz, 1944: Familie und Alter. Ein konservativer Beitrag zur sozialpolitischen Diskussion. Heft 3 der Schriftenreihe “Sammlung zum Aufbau“, hg. vom Generalsekretariat der Schweizerischen Konservativen Volkspartei, Bern. Weber, Quirin, 1989: Korporatismus statt Sozialismus. Die Idee der berufsständischen Ordnung im schweizerischen Katholizismus während der Zwischenkriegszeit, Freiburg: Universitätsverlag Freiburg. Wigger, Bernhard, 1997: Die Schweizerische Konservative Volkspartei 1903–1918. Politik zwischen Kulturkampf und Klassenkampf, Freiburg: Universitätsverlag Freiburg. Wili, Hans-Urs, 1991. Jux Populi? Vox Dei? 100 Jahre eidgenössische Volksinitiative auf Partialrevision der Bundesverfassung, in: Zeitschrift für Schweizerisches Recht 110: 485–519. Zacher, Georg, 1899: Die Arbeiter-Versicherung in der Schweiz. Berlin: Verlag der ArbeiterVersorgung. Ziegler, Alfred R., 1939: Die evangelisch-soziale Bewegung der Schweiz, Zürich: Verlag Schweiz. Verband evangelischer Arbeiter und Angestellter. Zimmermann, Hans, 1948: Sozialpolitische Ideen im schweizerischen Freisinn 1914–1945, Zürich: Juris Verlag.

211

8

The Church as Nation? The Role of Religion in the Development of the Swedish Welfare State

Karen M. Anderson

8.1

Introduction

Religion as an explanatory variable is conspicuously absent in most accounts of the historical development of the Scandinavian welfare states. Most explanations of the emergence of a “social democratic“ welfare regime in Scandinavia emphasize the power resources of class actors, especially the labor movement (Korpi 1978, EspingAndersen 1985, Stephens 1979), the influence of cross-class coalitions (Swenson 2002), or the impact of policy legacies and autonomous state actors (Heclo 1974; Weir and Skocpol 1985). A survey of the large literature on the formative period of the Swedish welfare state reveals few references to the role of religion or the Church of Sweden, and the same is true of the period after 1932 when the Social Democratic Party (SAP) was arguably the single most important actor in Swedish social policy development.1 One of the reasons religion has played such an insignificant role on Scandinavian politics is that there are few religious parties to push confessional issues. To be sure, the Christian Democratic Party in Norway has always been an important force, but Norway is the exception to the Scandinavian rule. Religious parties have never had much influence in Sweden, Denmark and Finland.2 The classic analyses of the development of party systems in Western Europe attribute the weakness of religious parties in Scandinavia to the absence of a religious cleavage. In Denmark and Sweden, the monarchs usurped church influence and wealth in the wake of the Reformation (Lipset and Rokkan 1967; cf. Kasperssen and Lindvall 2006).3 Scandinavian monarchs seized Church lands and wealth, incorporated the clergy into the state apparatus, and

1 The literature on the formative period, about 1880–1930 is mostly in Swedish. Key works are Berge (1995); Edebalk (1996); Montgomery (1951); and Höjer (1952). See Heclo (1974), Baldwin (1990) and Olsson (1993) for research in English. 2 The Christian Democratic Party in Sweden (kds, kristdemokratiska partiet) entered the Riksdag for the first time in 1991 and gets only 5–8% of the national vote. 3

Norway was part of Denmark at the time and Finland was part of Sweden.

212

assumed the welfare functions that the Church had previously performed, such as hospital care, and poor relief. This fusion of church and state reduced religiouslybased conflict, and located defenders of the established religion firmly within the state apparatus. As Morgan (this volume) and Manow (2004) argue, increasingly secular Scandinavian states could assume responsibility for welfare functions in the period of industrialization and democratization, when the labor movements emerged and the “social question” reached the political agenda. The essential point is that the Lutheran Church in Scandinavia did not oppose state involvement in social welfare. This chapter investigates the role of religion in the formative period of the Swedish welfare state. What was the role of the official state church in this early period of welfare state building in Sweden? Did an increasingly secular state assume responsibility for social risks like pensions, sickness, and occupational injury as it would later do in family policies? Did the fusion of church and state and the weakness of religious cleavages facilitate increased state involvement in social welfare? The account of the formative period of Swedish social insurance presented here provides much support for the Manow/Morgan thesis that low levels of religiously-based conflict in Sweden facilitated the emergence of activist welfare statism. The “secularization of political life” (Morgan 2003) in the 1800s and 1900s meant that liberal intellectuals, state administrators and Social Democrats competed to define the emerging social policy agenda in the late 1800s and early 1900s. Thus the character of early social insurance in Sweden was the outcome of political bargaining between Agrarians, Social Democrats, and Liberals. A closer look at the role of the Church of Sweden in this process, however, reveals a puzzle. The Church of Sweden’s initial response to growing state activity in the field of social policy was essentially negative: Church doctrine (at least as it was applied in Sweden) claimed that economic conditions such as income differences and poverty were the will of God and should be accepted as such. This view implied that the state had no business intervening to improve the material existence of citizens and wage earners. Despite the Church’s privileged position, it could not prevail on issues of social policy, and it vigorously opposed social democratic (and liberal) proposals for social reform until well into the 20th century. A coalition of Social Democrats and liberals pushed for the introduction of the earliest social insurance initiatives in the 1890s, as well as the important pension reform of 1913. Liberal and Social Democratic

213

views dominated until 1932 when the balance of political power shifted decisively toward the Social Democrats and their new ally, the Agrarian Party.4 When it became clear in the first few decades of the 1900s that social democracy would emerge as a powerful political force, the Church moderated its stance concerning social democracy and social reforms, but it was too late for the Church to try to regain the initiative. The social democratic-liberal consensus had prepared the ground for more farreaching social democratic social policies, and the Church of Sweden was mainly an onlooker in this historical process. In line with the analytical framework advanced in this volume, this chapter argues that Lutheranism shaped Scandinavian welfare state development because Lutheranism as an organized religion placed few obstacles in the way of state involvement in social welfare. However, this chapter highlights the potential for conflict between the Church of Sweden and emerging state social policy, even if this conflict did not develop into a strong church-state cleavage similar to the Netherlands or France. In contrast to the state churches in Norway and Denmark, the Swedish state church was orthodox and conservative. As Thorkildsen (1997) puts it, the Swedish (and Finnish) church was a ‘high church’ and the Danish and Norwegian churches were ‘low churches.’ The Church of Sweden did try to influence the direction of social policy, but failed. In other words, Manow and Morgan are correct that the secularization of political life that began in Sweden in the mid 1800s facilitated the development of a strong role for state social policy, but this occurred against the backdrop of a conservative, privileged state church’s unsuccessful attempts to influence the direction of social policy, especially poor relief. Bringing Lutheran values into the analysis complicates things even further. According to Thorkildsen (1997: 159), Lutheranism’s emphasis on the importance of daily work and a “priesthood of all believers” correspond to two values central to the Scandinavian welfare states: full employment and equality. In other words, Lutheran theology seems to provide the justification for, or at least did not contradict, universal social policies and high levels of employment, in addition to the “bread first, work later” justification of social assistance that Kahl

4 The SAP and Agrarian Party (Bondeförbundet) agreed on the so-called “Cow Trade” (kohandeln) in 1933. The SAP supported agricultural price supports in return for the Agrarians’ support of SAP economic policies. This deal inaugerated the “Red-Green” alliance between the SAP and Agrarians that lasted (with some modifications during WWII) until 1957. The alliance broke up over the issue of supplementary pensions. See Heclo 1974.

214

(this volume) stresses as a characteristic of Lutheranism. Why did the Church of Sweden then stick to its orthodox interpretation of Lutheran theology in social policy matters? The chapter argues that the Church of Sweden was so close to the state and so dependent on state resources and support that it settled into a period of stagnation that reinforced its conservativism. The Church of Sweden, because of its privileged position, showed very few signs of renewal or flexibility as the “social question” emerged in the second half of the 1800s. To put it more bluntly, the Church of Sweden became organizationally complacent and showed few signs of trying to adjust to changing external conditions, such as the challenges posed by industrialization and the emergence of revivalism (Finke and Stark 2005). So the Church stuck to its old conservative Lutheran doctrines which legitimized inequality (and alienated workers and the middle class) and the doctrinal supremacy of Lutheranism. This environment contrasts sharply with the experience of Catholic churches and Catholic parties during the same period. The Catholic Church and the parties that defended it faced much more hostile environments than the Church of Sweden and thus faced stronger pressure to react. They did this by displaying more openness to the world around them, and they were much more likely to seek new sources of support. This is why the Catholic Church actively tried to expand its support, among other things by mobilizing workers in Catholic trade unions and political parties (see Lynch, Morgan, and van Kersbergen and Manow, this volume). The chapter begins with a survey of the religious and political landscape during the 1800s when the first elements of the welfare state were introduced. The subsequent sections address the role of the Church of Sweden and other political forces in early social welfare legislation. I conclude with a short discussion of the role of the Lutheran state church in Scandinavian (Swedish) welfare state development in light of the arguments that frame the rest of the chapters in this volume.

215

8.2

Politics and Religion during the Formative Period of the Swedish Welfare State

Politics Political and economic development occurred relatively late in Sweden. Sweden was one of the last European countries to industrialize (in the 1870s), parliamentarism was established only in 1917, and the struggle for universal suffrage was delayed; all men got the right to vote in 1909 and women in 1918. Proportional representation was introduced in the Second Chamber (the Lower House) in 1907. In 1865, parliament was composed of four houses, each corresponding to an estate: clergy, burghers, nobility, and peasantry. In 1866, a bicameral parliament replaced the estate parliament, but the reform more or less preserved the existing distribution of power in parliament.1 Some groups did lose power, however, including the clergy. Louis De Geer, the architect of the reform, is often called a liberal, but the reform consolidated the power of conservative forces and delayed the emergence of an organized liberal movement. According to Rustow (1955: 25), the reform “laid the foundations for an alliance between bureaucracy, landed wealth, and industry.“ The old four estate parliament had hindered the formation of political parties, so the 1865 reform facilitated the emergence of modern political parties, although they were slow to develop and initially weakly organized. The first organized political party was the Lantmannaparti (Ruralist Party) in the lower house in 1868. After the 1865 reform, land-owning peasants were the largest grouping in the Lower House so there were gains to be made by trying to behave as a cohesive block. In 1888 the Ruralists Party split into a free trade and protectionist wing only to merge again in 1895. Thereafter, the party represented a more conservative ideology,2 but the rural interests of the party became much less pronounced when the party merged with one of the forerunners of the Conservative Party in 1912.3 Liberal ideas were very influential in Sweden in the 1800s, and many members of parliament were acknowledged liber-

1 Only 20% of males could vote in elections to the lower house because of property requirements. 2 The Ruralist Party was not an agrarian party, since it represented landholders and not agricultural laborers. The Ruralist Party is one of the forerunners of the current Conservative Party (moderata samlingspartiet). 3

This new party was called Lantmanna- och borgarepartiet,

216

als even before the Liberal Party (liberala samlingspartiet) was finally formed in 1900.4 The Social Democrats formed their own party organization prior to this, in 1889. With no political party representing them, rural interests formed two parties. In 1913, the Farmer’s Party (Bondeförbundet) was formed, appealing mainly to the interests of smallholders. In 1915 the more conservative and pro-defense National Association of Farmers (Jordbrukarnas Riksförbundet) was established. The two parties cooperated closely in Parliament; getting 12 seats in the 1917 election and 29 in the 1920 election. In 1921 they merged under the name Farmers’ Party.5 Nationally organized political parties grew in tandem with the struggle to extend the franchise, and this common mission was a powerful force uniting the Social Democrats and Liberals. For the Social Democrats, revolutionary ideology was subordinated to the campaign for universal suffrage. The more or less simultaneous emergence of organized liberalism and socialism in the context of the campaign for universal suffrage was to have important implications for the first Swedish social insurance policies. As Rustow puts it, “Sweden is … perhaps the only country where political liberalism and socialism emerged simultaneously (1955: 43).” The Liberals were willing to cooperate with the SAP, not only to extend the suffrage but also to enact social policy legislation. This early alliance between the Social Democrats and Liberals decisively shaped social policy from the turn of the century until the 1930s (Höjer 1952: 38).6 A Liberal-Conservative coalition was out of the question because of bitter differences between the two over defense spending. Rustow (1955: 44) characterizes the period 1890 to 1920 as the “transition from oligarchy to democracy.” Prior to 1920 less than half of males could vote, parties were weakly organized, and the principle of government responsibility to parliament had

4

In 1902 Frisinnade Landsföreningen was formed, the national organization for the Liberal Party

5 The Upper House was indirectly elected by the County Councils. Thus all references to national elections are for the Lower House. 6 By 1902 the Liberals were the largest party in the lower chamber. 1914 Liberals split with dissenters joining the conservatives. At the time the Liberals were internally divided between free traders who also thought religion and alcohol were private matters, and members of the free churches who were pacifist and temperate (drank no alcohol). The campaign for universal suffrage had kept the two factions together. Before the split the Liberal party was losing members to the SAP and Agrarians. The parties joined again in 1934 under the name Folkpartiet.

217

not yet been established. In parliament, parties and individual members represented the interests of three groups: state bureaucrats, land-owning peasants, and industrialists. Until the turn of the century, parties played little role in social policy. Initiatives came from individual members of parliament, and these were most often Liberals (Höjer 1952). Liberals more or less dominated the few parliamentary ventures into the sphere of social policy. By the early 1900s, the Liberals were the largest party in the Second Chamber (Lower House) of Parliament, and from 1920–1932 the Liberals were the pivotal party. Government turnover was frequent, but the Liberals always participated despite losing ground to the Conservatives after 1917. To sum up, parliamentary democracy and universal suffrage were slow to arrive in Sweden. Liberalism was the dominant political ideology of the 1800s, but socialist ideas began to challenge liberalism in the late 1800s. Despite differences in ideology, Liberals and Social Democrats were united in their common struggle to achieve universal suffrage and to introduce social legislation. Soon after the introduction of universal male suffrage, the Social Democratic Party emerged as the largest political party in the directly elected Second Chamber. But government formation and policy achievements were hampered by the fragmentation of the parties (which resulted in unstable minority governments), as well as the veto power of the First Chamber. In the early 1930s, a sea-change occurred. Farmers fled the Conservatives for either of the two new Farmers’ Parties, leaving the Conservatives as the representatives of business; and the Farmers as representatives of more popular, rural interests. In the early 1930s, the Social Democratic-Liberal alliance broke up over the issue of how to fight to the Depression, and the 1933 “Cow Trade“ between the SAP and Farmers’ Party marked the end of a long period of liberal dominance (Rustow 1955: 3; see also Esping-Andersen 1985). This paved the way for the 1933 “Cow Trade” between the SAP and Farmers’ Party, in which agricultural price supports were traded for unemployment benefits and active labor market policies for urban laborers.7 In

7 The Social Democratic-Farmers’ Party crisis agreement also marked the beginning of Social Democratic ‘hegemony.’ From 1932 through 1976 every Swedish government was led by Social

218

1934, competing Liberal factions (previously-split over free trade and temperance) fused into the new Liberal Party (Folkpartiet). In 1943, various left splinter parties merged into the Communist Party (later, Vänsterpartiet-Kommunisterna, VPK). (Rustow 1955: 3; Hadenius et al. 1991). This ‘classic’ Swedish five-party system remained stable until the late 1980s. Religion Sweden’s history as an independent state stems from the peasant rebellion led by Gustav Vasa in 1521, and Swedish political development has been marked both by early and effective state-building, combined with late industrialization and democratization (Heclo & Madsen 1987). Indeed, the roots of Sweden’s welfare state date to the expropriation of Church lands and the establishment of a State Church, which resulted in the State usurping the Church in care for the poor and the aged, and in the provision of hospital services. 8 Evangelical-Lutheranism was the official state church of Sweden from 1593 to 2000. The Reformation came to Sweden when King Gustav Vasa confiscated church property and assets, and incorporated the Church into the emerging state apparatus. The process of replacing Catholicism with Lutheranism was slow and uneven, but complete by 1593. In the 1600s, religious freedom was not tolerated and “identification between the state and the church” was “total” (Gustafsson 2003: 51; see also Blűckert 2000). This fusion lasted well into the 1800s. Dissenters could be exiled, and this law survived until the second half of the 1800s. Only since 1858 were citizens permitted to gather for religious services without a priest (since 1726, see Ekström 1999). In the 1870s citizens could leave the Church of Sweden, but only to join another Christian

Democrats (other than a period of three months in 1936), and from 1951 to 1969. During this period, the SAP achieved a majority in the Second Chamber only twice. Instead, their control of the government was made possible by coalitions with the Farmers’ Party (1936–9, 1951–7), and the passive support of the communists, as well as all-party support during the war years. 8 A legacy of these struggles was the Swedish tradition of administrative independence from both Monarch and parliament in the form of independent agencies (ämbetsverk) governed by administrative boards (styrelser), often with corporatist representation (Heclo & Madsen 1987; Anton 1980).

219

church. Full religious freedom was not achieved until the 1950s, and even then, few Swedes left the church.9 In the 19th century, opposition to the privileged status of the Church of Sweden grew fierce. The revival movement (väckelserörelsen) had begun in the 1700s within the Church of Sweden, and pressure was building for more religious freedom. Parliament tried to reform the laws on religious freedom throughout the 1800s, but the State Church blocked many of them. Many seeking religious freedom emigrated to North America, while others joined with the emerging liberal movement to agitate for religious freedom. The 1864 freedom of association reform spurred the growth of the “free churches,“ (frikyrkorna) protestant churches that had broken away from Lutheranism, including Baptists, Methodists and other evangelical groupings. The free churches attracted members from all strata of society, especially the industrial and rural working classes. The 1864 reform also facilitated the emergence of other organized groups in civil society, such as labor unions, the temperance movement and consumer organizations (collectively known as the popular movements). The free churches challenged the privileged role of the Church of Sweden, as did the labor movement, which was officially anti-religious. Leaders of the free churches came mainly from among the agricultural classes and the urban upper classes. The membership of the free churches did include large numbers of industrial workers, however. According to Palm (1982), the free churches began as a rural movement, a protest against the agricultural conditions of the 1800s. As the industrial working class increased in the last decades of the 19th century, many of these workers joined the free churches. The churches’ stance toward the nascent socialist labor movement was ambivalent, however. The leadership (see above) opposed industrial action and socialist goals, but did not take this so far as to forbid church members from joining socialist labor unions. On the other hand, the free churches, liberals and social democrats cooperated in the campaign for universal suffrage and in fighting for social reforms. The free churches devoted some energy to establishing Christian labor unions that would not be affiliated with any political party. The Swedish Workers’ Association

9

This section is based on Gustafsson (2003).

220

was established in 1889, a year after the Social Democratic Party was formed, with the explicit aim of providing an alternative to socialist labor unions and the SAP (Palm 1982). The SWA was short-lived, despite full backing of the leaders of the free churches. Palm (1982) reports that most workers who belonged to the free churches were torn between their religious affiliation and their status as a worker who might join a union. According to Ekström (1999: 21) there was widespread recognition that the Church of Sweden was “in crisis“ at the turn of the century. Church attendance had fallen, and fewer church members took communion. Moreover, Ekström argues that the Church itself was partly to blame for this development: the rise of the free churches was a clear sign that the Church of Sweden was responding inadequately to the “religious demands of a new era.” This brief review of political and religious conditions in the 1800s demonstrates that the Church of Sweden was not an active participant in shaping the social and political agenda of the 19th century. The Church’s stance was essentially reactive: the Church managed to slow the process of reforms that increased religious freedom, but it had no organized political arm in the form of a political party. To be sure, many conservative members of parliament (including some Church officials) represented the Church’s views. Most importantly, the Church’s status was being challenged by powerful liberal, social democratic, and “free church’s” forces.

8.2.1 The Church of Sweden and Social Policy Before 1900 Since the medieval period, the Church of Sweden was heavily involved in poor relief. The 1847 and 1852 Poor Laws built on existing arrangements and clarified the responsibility of cities and parishes to provide poor relief (see Höjer 1952). The Church’s views concerning poor relief were based on the idea that if there were rich and poor, this was God’s will. Church teachings explicitly reinforced the existing social order and offered no rationale for trying to improve the material conditions of the poor. As Höjer (1952: 23) puts it, “the thought that it might be a Christian duty to work for better social conditions was completely foreign to the men of the Church of Sweden.”

221

The Church of Sweden’s general response to demands for social policy in the 1800s was that “it was a divine law that wealth and poverty shall coexist on earth, and the poor man is comforted with the fact that it will be easier for him to enter heaven than it is for the rich man (Höjer 1952: 22).” For the Church, charity was the duty of the Christian, but this did not entail any rights on the part of the poor. The poor were viewed officially as persons who were not willing to work, wasteful or imprudent, and in any case poverty was God’s will.10 The local government reform (kommunallagen) of 1862 had important implications for church-state relations. The new law legally separated the church parishes from local administrative units even though both shared the same geographic space, and separated decision-making between them (parishes decided on church-related issues, local governments on all other issues). At the time, the central tasks of the Church included education and day care.11 Even though poor relief had traditionally been administered by the Church, the reform placed poor relief explicitly under secular local authority control.12 Poor relief was now a matter for the secular municipality. Secularization had little immediate impact: poor relief was still a charity based on a religious rationale. In the long term, however, the change marked a shift toward secular (public) responsibility for means-tested assistance (Höjer 1952). The secularization attracted opposition in religious circles (Duhne 1969) and the Church later expressed its opposition to publicly administered poor relief. The second half of the 1800s was marked by economic hardship and mass emigration to North America. Poor relief was reformed again in 1871 and the result was a setback compared to the 1847 reform. The poor lost their right to appeal municipal decisions. The charity nature of poor relief was emphasized in the new law (Höjer 1952). Two developments in the 1880s prompted the Church to take a more active interest in poor relief. First, the rise of social democracy and its calls for more social reform raised the specter of even more state intervention in social issues, and the Church’s

10 Thus the Church of Sweden’s interpretation of Lutheran theology differed from other Lutheran churches in Europe. See Kahl, this volume. 11 School has been compulsory since 1842. 12 The same reform created a new administrative structure at the provincial (län) level: the landsting. From this point on local hospitals would be organized under the landsting.

222

leaders saw this as a threat. Second, the economic downturn exposed the limits of poor relief as an instrument for alleviating poverty. It soon became clear that the existing system of poor relief was inadequate for dealing with emerging social problems. Church leaders responded by changing their position on poor relief, at least rhetorically: instead of viewing secular poor relief as a threat to the Church’s teachings, Church leaders now publicly advocated supplementing state poor relief with Church-sponsored measures. The idea was simply to increase the role of religious values in the state-run system, without replacing it entirely with a Church-run system. Church leaders discussed providing religious training to the social workers involved in administering poor relief who might work in poor relief (Höjer 1952). Nothing came of the Church’s few concrete proposals for strengthening the role of the state church in poor relief. Moreover, the Church’s view of public poor relief remained skeptical: even though poor relief was stingy and means-tested, the Church insisted it should be provided on the basis of love and not “entitlement.” Church leaders continued to argue that poor relief severed the link between the rich and poor because public benefits broke the link between charity and gratitude that should unite rich and poor (Duhne 1969: 165). And the Church still emphasized that moral failing was the source of all ills, including poverty. This view of publicly organized poor relief prevailed until the turn of the century. The Church continued to criticize public poor relief because it created a legal entitlement and left little scope for “care of the soul” and moral education. Only in the first decades of the 20th century did the Church begin to advocate a more active Church role in the administration of state poor relief.

8.2.2 Early Social Insurance 1880s–1900 The rapid industrialization of the Swedish economy after 1870 was the backdrop to the first serious calls for social insurance legislation. Between 1870 and 1900, the number of persons employed in industry nearly doubled from 15% to 29%. The number engaged in agriculture decreased in the same period from 72 to 55% (Höjer 1952: 35).

223

Sweden, like the rest of Scandinavia, was home to a strong Liberal movement in the 1800s. Industrialization and freedom of occupation (the guild system was abolished in 1846) encouraged this development (Kuhnle 1981: 135). Economic liberalism stressed individual self-reliance and skepticism toward state intervention. Liberalism was strongest in Denmark, but Sweden was a close second. During this period, the first mutual aid societies were formed, particularly sickness and burial funds, beginning about 1820 (Johansson 2003). Most scholars of Swedish welfare state development cite 1884 as the decisive year for the emergence of a genuine debate about social policy. In 1884, a Liberal member of the lower house, S.A. Hedin, introduced a bill requesting an official commission of inquiry in the area of “workers’ insurance.” Hedin’s motion was inspired by Bismarck’s reforms in Germany. The motion described developments in other countries and requested an investigation and proposal for Sweden. The Agrarians in parliament supported the bill on the condition that it include all workers not just industrial workers. According to Olsson, Hedin’s motion reflected his ties to free churches and the temperance movements (Olsson 1993: 46).

Hedin’s motion was approved by

parliament and marked the beginning of a period of discussion about the role of the state in social insurance. Parliament appointed a committee, and on the basis of the committee proposals, the 1891 sickness fund law was adopted. The committee rejected employer financing as too costly and argued instead for minimal state subsidies. The resulting law was weak because it only required regulation of existing funds’ activities (see Johansson 2003). A new parliamentary committee was appointed in 1893 to investigate the issue of old age pensions and accident insurance. The committee proposed that financing be shared by employers and employees, but parliament rejected bills in 1895 and 1898 based on the committee’s work, largely because of the opposition of self-employed farmers and those representing farm laborers. Thus there was no consensus on social insurance legislation beyond the very weak sickness fund legislation adopted in 1891. An exception to this deadlock was the 1886 legislation requiring employers in the railroad industry to provide compensation for work-related accidents. This was the basis for more comprehensive legislation in 1901 that extended the provisions to employers in specific sectors. Employers could insure themselves in one of the newly

224

created national insurance agencies. Conservative opposition to the legislation was strong, and it was based, at least rhetorically, on religious principles. Höjer reports that a conservative member of parliament characterized the law as “undoubtedly socialist” and “paving the way for comprehensive legislation that would depart from the old adage, ‘help yourself and God will help you.’ It is the duty of each and every one to provide for himself and to not leave this to others.“ (Höjer 1952: 42–43). The only Social Democratic member of parliament, Hjalmar Branting, voted against the legislation, claiming it was worse than nothing. In sum, the first decades of social policy activity in parliament produced dismal results, at least from Social Democratic and liberal perspectives. What was the role of the Church of Sweden in the last decades of the 19th century when social policy questions reached the political agenda? Höjer (1952: 50) observes that the Church’s view on social policy questions had changed little during the 1800s. Höjer summarizes the Church’s position with the following quotation from a prominent church official: “‘human society, with its class differences, its wealth and its destitution, are given by God.’” According to Archbishop Rueterdahl, (cited in Höjer 1952: 50) “A Christian submits with humility to the condition that has been decided for him.” Given this perspective, it was Church policy to oppose social reforms that aimed to ameliorate poverty and other social ills. A citation from Väktaren (The Sentinel), the magazine for state and church, illustrates a common view among Church officials, that socialism was basically the same as social reformism. Moreover, socialism was the “arch enemy of the church, the state and every individual, peaceful citizen and should be fought by all available means (51).” As Höjer puts it, the church was the guardian of spiritual values while the social reformers were “materialists (51).” The Church’s view began to change somewhat in the 1890s. It was by now clear that the labor movement was emerging as an important social actor and the church’s vocal opposition to measures that would improve workers’ material existence aroused suspicion. The Church’s opposition to social reforms had already caused many workers to turn away from the Church (Höjer 1952: 51). Church opposition to social reforms mainly concerned measures to improve the economic conditions of lower classes. The Church was less opposed to measures aimed at improving social services, such as foster children care.

225

8.3

Social Democrats, Liberals and Agrarian Competition in Social Policy: 1900–1932

The 1913 pension reform was a turning point in Swedish social policy development because it introduced a universal right to income support in old age. The reform was adopted under the Liberal Prime Minister Karl Staaff, coming nearly three decades after Hedin’s call for workers’ pension insurance in 1885. Baldwin (1990) argues that agrarian interests blocked attempts to introduce pensions between 1895 and 1913 because proposals included only industrial workers and not the agrarian selfemployed or rural laborers. Agrarian interests opposed measures that would finance pensions by employer contributions and that excluded rural laborers. Both agrarian employers and landless laborers would lose in such a system. When the Social Democrats and Liberals realized that “political power resided in the countryside” (Baldwin 1990: 83) they changed their proposals to appeal to agrarian interests as well as the working class. The resulting legislation consisted of income-related pensions financed by individual contributions as well as income-tested supplements for those with inadequate earnings-related benefits. One of the primary motivations for the 1913 pension law was to reduce reliance on poor relief.13 The 2,400 municipalities administered poor relief, which was highly punitive and stigmatizing. Earlier low rates of infant mortality and high rates of emigration resulted in a high old age dependency ratio in Sweden at the turn of the century, which was twice as high as in England and Germany. Poor relief was financed by local taxes and implementation varied greatly. Income tested supplements in the new pension legislation were specifically designed to reduce reliance on poor relief, thereby reducing costs for local government, and to eliminate the inhumane elements of poor relief. The 1913 pension law introduced two principles that would later become cornerstones of Swedish social policy: universalism and significant tax-financing. The weakness of the Social Democrats, the strength of agrarian interests, and the Liberals embrace of universalism explain the shift in emphasis from workers’ insurance to universalism. Farmers preferred full tax financing for obvious reasons, but the state

13 This section is based on Edebalk (2000; 2003).

226

of public finances precluded this option, so farmers had to settle for partial tax financing. To sum up, early social insurance legislation bore the imprint of Liberal, Agrarian and Social Democratic interests. Between 1880 and 1910, farmers were still the largest group in Swedish society, despite the rapid growth of the industrial working class. The Church of Sweden had little influence on this legislation.

8.4

Secular Ideological Influences on Early Swedish Social Policy

If the Church of Sweden did not take the lead in shaping the debate on social reforms, who did? As noted, liberalism was a powerful political force in Sweden in the 19th century and liberal ideas were perhaps the strongest source of inspiration for social policy ideas until a coherent social democratic ideology emerged in the early 1900s. Liberals were not directly opposed to the Church of Sweden but criticized the Church as an organ for reactionary groups. As noted, the free church movement was closely associated with Liberalism: both were united in their struggle for universal suffrage and religious freedom and both had strong roots in the urban bourgeoisie (Rosenberg 1948).

8.5

Liberal Intellectuals

Non-religious liberal groups such as the organization Verdandi founded in 1881 were important sources of ideas about social reform. Verdandi published Swedish translations of John Stuart Mill, and Karl Staaff, a prominent liberal member of the lower chamber of parliament and later Prime Minister was a member.

227

Economists also played an important role among liberal intellectuals. Gustav Cassel14 and Eli Heckscher were among the more prominent, and their writings and speeches had a significant impact on liberal thinking on social reform. Olsson (1993: 54) describes the period as follows: In particular the 1880s marked a breakthrough of a new era in the intellectual life of Sweden: the cultural and religious orthodoxy under united state-church auspices was challenged by a heterogeneous European-influenced counter-culture which included natural scientists, artists and social philosophers.

The CSA, the National Association of Social Work, founded in 1902, also came to play an important role in shaping the emerging social-liberal social policy discourse. The CSA was politically neutral but firmly liberal in its outlook, promoting the values of voluntarism, preventive self-help, and individual responsibility. The CSA had strong ties to both the liberal party, the temperance movement, and various churches, and it sponsored research, conferences, training and philanthropic activities that promoted the role of civil society (rather than the state) in solving social problems. CSA members came mainly from the urban upper middle class, and male CSA members were later an important source of talent for the emerging welfare bureaucracies. The CSA was sharply critical of existing poor relief arrangements and devoted much of its early activities at working toward improving poor relief.15 As Qvarsell puts it, the CSA was a meeting place for those engaged in the increasingly controversial social questions of the time. And ironically enough, the CSA, though a philanthropic organization, actively promoted a stronger central state role in poor relief as well as a stronger state role in other areas of social policy. This paradox must be seen in the context of the state’s minimal and passive role in social policy questions at the turn of the century. The CSA reached its peak in influence in the first three decades of the 20th century and was an important actor pushing for the establishment of central state institutions in the field of social welfare. The National Board of Social Welfare (Socialstyrelsen) was set up in 1912 and the Ministry of Social Affairs in 1921, at least partly due to CSA influence and activities (Qvarsell 2003). The CSA played a major role in shaping the 1918 Law on Poor Relief. (Edebalk 2003).

14 Cassel published a book, Socialpolitik (social policy) in 1902 that had a significant impact on public debate. 15 On the CSA see Olsson (1993), Qvarsell (2003).

228

8.6

Social Democrats, the Church of Sweden, and Early Social Policy Ideas

When the Social Democratic Party (SAP) was established in 1889, it did not have a clear social policy position. Early SAP social policy was overshadowed by the temperance issue: more people belonged to the temperance organizations than to unions before the 1909 general strike, and many unionists were molded by their experience in the temperance movements. Moreover, as previous sections demonstrate, the SAP’s chief objective in its early years was the attainment of universal suffrage, and all other ambitions were subordinated to this one. The Social Democratic Party was fiercely anti-clerical from its inception. The Church of Sweden was considered to represent conservative forces in society and besides, the Church strongly opposed any social reforms that would alter the existing social order. Thus Church doctrine and social democratic ideology were diametrically opposed.16 It is not surprising that the Social Democrats allied so quickly with Liberals in pursuit of social reforms and universal suffrage. Tingsten (1973) divides Social Democratic – Church relations into two rough periods. roughly two periods. From the 1880s until roughly 1910 the Social Democratic Party vigorously opposed the fusion of Church and state, especially Church control over education. The Church was equally hostile to the Social Democrats, engaging in “direct antisocialist agitation” (Tingsten 1973: 587). The Social Democrats were not only anti-clerical in these early years, they were anti-religious, criticizing the free churches. By the turn of the century, the Social Democrats had moderated their stance, voicing their respect for religion as a private matter, but still vigorously opposing the Church of Sweden. The SAP could not afford to offend workers with religious sympathies or to alienate the Liberals (with their ties to the free churches) with whom the SAP often cooperated during elections (Tingsten 1973: 597). In other words, the Social Democratic Party’s calls for the separation of Church and state seemed to fall on deaf ears, so the party began to tone down its anti-clerical rhetoric.17

16 On the relationship between the Social Democratic party and the Church of Sweden see B. Gustafsson (1953). 17 The expulsion from the party of the fiercely anti-religious “Young Socialists” in 1908 also contributed to the SAP’s moderation.

229

The Church of Sweden also began to adopt a more moderate position regarding social democracy and social reform around the turn of the century. As Tingsten (1973: 597–598) puts it, “Step by step, hostility diminished on both sides; the sharp attacks became fewer, expressions in favor of a certain mutual acknowledgement (if with great reservations) more numerous.” To be sure, the Church still opposed social democracy because it encouraged the poor to be dissatisfied with their fate and to seek improvement in their material status. In the first decade of the 20th century, prominent social democrats began to articulate the party’s new-found religious moderation. Instead of opposing the state church and religion, SAP leaders now criticized the Church of Sweden for its conservatism, and argued that fundamental Christian values were compatible, even congruent, with socialism. By the 1920s and 1930s, the Church of Sweden took on a much more politically neutral stance and became much more active in social reform. Tingsten characterizes the mutual acceptance of Social Democracy and the Church of Sweden as a long-term, interactive process. By the 1930s, the Social Democrats were the largest political party and thus less threatened by the Church. The party further toned down its rhetoric and policies critical of the Church starting in the 1930s.

8.7

Discussion and Conclusion

The preceding sections support the contention that the Church of Sweden played a passive if not oppositional role in political struggles over early social policy initiatives in Sweden. The Church opposed the secularization and expansion of poor relief, it played no role in the debates in parliament about the introduction of workers’ insurances in the 1880s and 1890s, and it had little influence on the more substantial social insurance reforms passed between 1900 and 1930. Why did the Church have so little influence on early social policy reforms? Given its privileged status as a state church, the Church of Sweden should have been in an ideal position to influence early social policies. For example, the Church might have produced experts/bureaucrats who could use their influence to push specific policy ideas. This did not happen; instead, the Church preached a conservative conception of social policy in which the poor were expected to accept their miserable lot. The

230

politicians and bureaucrats who shaped early social policy debates had little to do with the Church of Sweden; they were liberals and social democrats. Moreover, Church opposition to emerging state social policy had no political outlet in the form of a political party or some other societal platform. One of the keys to understanding the limited role of the Church of Sweden in early social policy legislation was the 1862 local government reform that transferred responsibility for poor relief from the Church parishes to the secular municipalities and decreased or abolished other aspects of the Church’s local influence. This reform set a precedent for state involvement in social welfare, even if it took more than 50 years for the state’s role to expand (in poor relief). Limited religious freedom and the emergence of the free churches also challenged the dominance and influence of the Church of Sweden. The Church of Sweden never really recovered from this loss of influence, and it made the fatal mistake of clinging fiercely to traditional Church doctrine about the causes and consequences of poverty and other social ills at a time when large segments of society were looking to the emerging social democratic and liberal movements to produce political solutions. Church leaders changed their strategy around 1900, but their efforts to regain the social policy initiative from liberals and social democrats were inadequate and too late. Whereas the Liberal Party had a strong link to the free churches, the Social Democrats were able to weather attempts by the Liberals and free churches to establish a Christian labor movement. The SAP incorporated religion into its party apparatus with the establishment of the “Broderskapsrörelse” (Social Democratic Brotherhood) in the 1930s. This party organization was open to SAP members of all ecumenical faiths and had the effect of de-politicizing religion within the party. The Social Democratic Brotherhood never dominated the party, but at least it meant that the religious cleavage in Swedish society did not reinforce the socialist-non-socialist cleavage. If the Church did not play a leading role in shaping the social policy debate, who did? In line with much previous research (Olsson 2003; Baldwin 1990; Edebalk 1996), I have stressed the role of liberalism, agrarian interests, and social democracy in shaping welfare state institutions before 1930. Social-liberal social policy initiatives were no doubt powerfully influenced by the ongoing secularization of Swedish society that began about 1880. In public debates about the role of social policy, liberal

231

economists and social liberal organizations like the CSA played an important role, and it is fair to say that they, and not the Church of Sweden, shaped the terms of the debate. It is probably not incorrect to argue that as the Church of Sweden continued to promote its orthodox Lutheran vision concerning the causes and consequences of poverty and other social ills, but few in society listened. Thus Church-state fusion was accompanied by the gradual secularization of political life, and the transfer of important welfare functions from the church to the state (poor relief) strengthened the secularization process.18 However, the Church of Sweden was not a benevolent participant in this process. The Church tried to influence the course of social policy in the formative period of the welfare state, but failed. Liberals, Social Democrats and Agrarians competed to control the content of early social policy.19 All three groups advocated state involvement in social insurance, but they differed on how benefits should be financed and whether social insurance should be universal or available only to workers.20 These crucial distinctions cannot be explained by religious factors. As I have argued, the role of the Church in early social policy was oppositional, and the working class actively opposed Church doctrine concerning the causes of social problems. If anything, the Church’s opposition to secular social welfare probably increased the attractiveness of state welfare to the working class. The Church of Sweden preached that individuals should accept the fate ordained by God until the first decade of the 20th century. And the dismal state of public social policy until about 1910 meant that even Swedish liberals moderated their calls for individual responsibility. Existing social policies (mainly poor relief) were so meager at the turn of the century that most liberals advocated increasing the role of the state in social policy. They continued to espouse individual responsibility, but in combination with an expanded state role. To sum up, an increasingly secular state assumed responsibility for social risks like pensions, sickness, and occupational injury (as it would later do in family policies),

18 It should be noted that the municipalities and not the central state, administered poor relief. 19 Predictably, Conservatives were opposed to increased state involvement in social insurance. 20 Liberals were also divided on the issue of individual responsibility versus universal entitlement.

232

but this seems to have been little influenced by the Church of Sweden. The “religious political space” in Sweden was occupied by a conservative state church that opposed public/secular social policies. The privileged status of this state church meant that its position was protected, and the vast majority of the population remained members. However, the Church’s privileged status as a state church also meant that it faced little pressure to adjust its teachings to new social and economic circumstances. So the Church of Sweden retained a conservative, traditional set of teachings that had little appeal to the emerging working classes. The Church of Sweden’s official state church status also did not mean that political actors would actually listen or heed its teachings, so alternative sources of social policy ideas emerged. In other words, church-state fusion left a vacuum that could be filled by secular social democratic and liberal forces.

Bibliography Anton, Thomas. 1980. Administered Politics. Elite Political Culture in Sweden. Dordrecht: Kluwer Academic Publishers. Baldwin, Peter. 1990. The Politics of Social Solidarity. Class Bases of the European Welfare State 1875–1975. Cambridge: Cambridge University Press. Berge, Anders. 1995. Medborgarrätt och egenansvar. De sociala försäkringarna i Sverige 1901–1935. Lund: Arkiv. Blückert, Kjell. 2000. The Church as Nation. A Study in Ecclesiology and Nationhood. Frankfurt am Main: Peter Lang. Cassel, Gustav. 1908. Socialpolitik. Stockholm: Hugo Gebers förlag. Duhne, Christian. 1969. Svenska kyrka och socialdemokratin. Till belysning av förhållandet dem emellan under brytningstiden. Stockholm: Verbum. Edebalk, Per-Gunnar. 1996. Välfärdsstaten träder fram. Svensk socialforsakring 1884 – 1955. Lund: Arkiv. Edebalk, Per-Gunnar. 2000. Emergence of a Welfare State – Social Insurance in Sweden in the 1910s. Journal of Social Policy, 29, 4, 537–551. Edebalk, Per-Gunnar. 2003. Folkpension och åldringsvård – om svensk socialpolitik 1903– 1950. Socialvetenskaplig tidskrift nr 2–3 pages 131–150 Ekström, Sören. 1999. Svenska kyrkan i utveckling. Historia, identitet, verksamhet, och organisation. Stockholm: Verbum. Esping-Andersen, Gøsta. 1985. Politics Against Markets. Princeton: Princeton University Press. Finke, Roger and Rodney Stark. 2005. The Churching of America, 1776–2005. New Brunswick, N.J.: Rutgers University Press.

233

Gustafsson, Berndt. 1953. Socialdemokratin och kyrkan 1881–1890. Lund: Håkan Ohlssons boktryckeri. Gustafsson, Göran. 2003. “Church-State Separation Swedish-Style,“ West European Politics. 26, 1, 51–72. Heclo, Hugh. 1974. Modern Social Politics in Britain and Sweden. New Haven: Yale University Press. Heclo, Hugh and Henrik Madsen. 1987. Policy and Politics in Sweden. Principled Pragmatism. Philadelphia: Temple University Press. Höjer, Karl J. 1952. Svensk socialpolitisk historia. Stockholm: P.A. Norstedt och söners förlag. Johansson, Peter. 2003. Fast i det förflutna. Institutioner och intressen i svensk sjukförsäkringspolitik 1891–1931. Lund: Arkiv. Kaspersen, Lars Bo and Johannes Lindvall. nd. “Why No Political Religion? Denmark and Sweden in Comparative Perspective.” Unpublished manuscript. Korpi, Walter. 1978. The Working Class in Welfare Capitalism. Work, Unions, and Politics in Sweden. London: Routledge & Kegan Paul. Kuhnle, Stein. 1981. “The Growth of Social Insurance Programs in Scandinaiva: Outside Influences and Internal Forces,“ in Peter Flora and Arnold J. Heidenheimer, The Development of Welfare states in Europe and America. New Brunswick, N.J.: Transaction Books. Lipset, Seymour Martin and Stein Rokkan. 1967. Party Systems and Voter Alignments. New York: Free Press. Manow, Philip. 2004. “‘The Good, the Bad and the Ugly,’ Esping-Andersen’s regime typology and the religious roots of the western welfare state.“ Paper prepared for workshop on “Religion and the Western Welfare State, Cologne, 30 April-1 May 2004. Morgan, Kimberly J. 2003. “The Politics of Mothers’ Employment.” World Politics 55, 259–289. Montgomery, Arthur. 1951. Svensk socialpolitik under 1800-talet. Stockholm: Kooperativa förbundet. Olsson, Sven. 1993. Social Policy and Welfare State in Sweden. Lund: Arkiv. Palm, Irving. 1982. Frikyrkorna, arbetarfrågan och klasskampen: frikyrkorörelsens hållning till arbetarnas fackliga och politiska kamp åren kring sekelskiftet. Stockholm: Almqvist & Wiksell international Qvarsell, Roger. 2003. “CSA och socialpolitiken kring sekelskiftet 1900,“ Socialvetenskaplig tidskrift 2–3. Rosenberg, Sven-Åke. 1948. Kyrkan och arbetarrörelsen. Lund: Gleerups förlag. Rustow, Dankwart A. 1955. The Politics of Compromise. A Study of Parties and Cabinet Government in Sweden. Princeton: Princeton University Press. Stephens, John D. 1979. The Transition from Capitalism to Socialism. Urbana: University of Illinois Press. Swenson, Peter. 2002. Capitalists Against Markets. Oxford: Oxford University Press. Thorkildsen, Dag. 1997. “Religious Identity and Nordic Identity,” in Øystein Sørensen and Bo Stråth (eds.), The Cultural Construction of Norden. Oslo: Scandinavian University Press, pp. 138–160.

234

Tingsten, Herbert. 1973. The Swedish Social Democrats. Their Ideological Development. Totowa New Jersey: Bedminster Press. Translation of Den svenska socialdemokratins idéutveckling, published in 1941. Weir, Margaret and Theda Skocpol. 1985. “State Structures and the Possibilities for “Keynesian“ Responses to the Great Depression in Sweden, Britain, and the United States,“ in Peter B. Evans, Dietrich Rueschemeyer, and Theda Skocpol, Bringing the State Back in. Cambridge: Cambridge University Press.

235

9

Religious Conservatives in U.S. Welfare State Politics*

Jill Quadagno and Deana Rohlinger

In the late 1970s a political movement arose in the United States whose origins lay neither in the class conflicts that led to the creation of national labor unions in the 1930s nor the “new social movements” of the 1960s – feminism, peace and the environment. Rather it originated among religious conservatives who were suspicious of government bureaucrats, fearful of losing control of their local subcultures, opposed to the Equal Rights Amendment, gay rights, and abortion and determined to insert faith into politics. At the core of the religious conservative movement were evangelical Christians, Protestants of various denominations who believe that the Bible is the inerrant word of God, that there is one true faith that provides an overarching meaning to life, and that it is the Christian duty of believers to spread the gospel (Keddie 1998; Wald, Owen and Hill 1988; Regnerus, Sikkink and Smith 1999). When evangelicals first appeared on the political scene in the 1970s, they had little experience with politics and low voter turnout but great potential to become a grassroots force for conservative principles. They represented over 20 percent of the population and had a massive infrastructure of Christian schools, bookstores, specialized magazines and newspapers as well as a vast communications network of Christian radio and television programs and stations. Televangelist preachers reached millions of viewers weekly and had a fund raising capacity that dwarfed that of political parties or individual candidates (Diamond 1995; Hardisty 1999; Oldfield 1996; Wilcox 2000). As evangelicals found themselves increasingly unable to isolate their families from what they perceived as disturbing secular trends, they began to mobilize this infrastructure for political purposes. They found a political ally in the Catholic Church, which was dedicated to overturning Roe v Wade, the 1973 Supreme Court decision that established a woman’s unlimited constitutional right to an abortion and among Mormons and other Christian groups that decried the erosion of moral standards

* We thank Kimberly Morgan, Catherine Bozendahl, Kees Van Kersbergen and Philip Manow for helpful comments on an earlier draft of this manuscript.

236

wrought by the social revolutions of the 1960s (Reichley 1986). Working through churches and the Christian media network, religious conservatives built an unparalleled national grassroots movement dedicated to promoting traditional “family values” (Moen 1992; Williams 2002; Clifton 2004). Their political objectives meshed with anti-communist fiscal conservatives, the traditional constituency of the Republican party, around social welfare policy. While traditional Republicans were mainly interested in reducing welfare spending, religious conservatives were more concerned with moral values and behavior. They castigated Medicaid-funded abortion services as state sponsored murder (Rohlinger 2002), spurned Aid to Families with Dependent Children (the federal-state benefit for poor mothers and children) for subsidizing “welfare queens” and “illegal aliens” (Diamond 1995; Hardisty 1999), damned a plan for national health insurance for encouraging abortion and euthanasia, and denigrated Social Security for eroding selfreliance and intergenerational connections and undermining the traditional family (Rohlinger and Quadagno 2006). The idea that a religious group or religious beliefs can influence welfare state politics challenges the traditional argument that the welfare state is a product of class politics and that the main purpose of welfare benefits is to ameliorate class inequality (Palme 1990; Korpi1989). Welfare state theories have just begun to recognize that religion can be as powerful a force as class (see van Kersbergen and Manow in this volume). But even for the analytical framework developed in this volume, which emphasizes the interplay between societal cleavage structures (frozen in party systems) and electoral rules, the US-American case still poses a serious puzzle, since the USA has a majoritarian electoral system and therefore a two-party system. However, the religious cleavage still seems to be highly relevant for US politics in general and for US social policies in particular. In this chapter we examine arguments about “American exceptionalism” in the context of two seemingly contradictory phenomena – a Constitution that mandates the separation of church and state versus intense personal religious beliefs that have frequently been the source of political mobilization. We analyze the main forces that brought conservative Christians into Republican party politics in the last quarter of the twentieth century. We assess the tactics and strategies they adopted to make their

237

voices heard, examine how their concerns were manifested in policy preferences and evaluate whether their demands were reflected in policy decisions. We argue that the religious conservative movement derives its strength from the organizational resources inherent in churches and from its ability to draw on a coherent theology to frame policy debates in moral terms. However, its success in the political arena also depends on an often contentious alliance with traditional Republicans who don’t share the emphasis on moral values and who are uneasy inserting religious objectives into policy matters. As a result, the Republican party has experienced periods of intense interparty party competition that threatens to undermine the coalition between religious and fiscal conservatives.

9.1

American Religious Exceptionalism

The United States has often been described as exceptional in terms of its class politics and unique experience of democratization. Unlike in Europe, where the struggle for the franchise emerged out of a feudal heritage and the experience of serfdom, in the United States, voting rights were nearly universal by the 1880s (at least for adult white males). Further, while the European labor movement regarded the modern state as the source of autonomy, the labor movement in the United States viewed the state as an instrument of coercion, used to break strikes and prevent unionization (Sombart 1906). Suspicion of state power made the American trade unions wary of government welfare programs and led to a preference for union-controlled benefits negotiated through collective bargaining (Stevens 1988). While the class exceptionalism of U.S. politics has been a prominent topic in welfare state debates, less readily recognized is American religious exceptionalism. In terms of church-state relations, the United States is a secular nation. The Puritans came to America in revolt against the English church hierarchy (Morone 2003). They constructed their society around a crusading religious spirit. Their search for God organized all pre-liberal institutions. The founding principle of the nation, as stated in the first amendment to the Constitution, mandates a separation of church and state: “Congress shall make no law respecting an establishment of religion or prohibiting the free exercise thereof.” There is no official state religion or religious party in the

238

United States and the historical tradition is one of religious pluralism. Yet the irony of U.S. politics is that the lack of an established church has given religion a vitality that infuses other social contexts and that erupts periodically into morally-motivated crusades (Tiryakian 1993). For centuries this religious fervor has endured. Alexis de Tocqueville wrote in 1835, “Religion never intervenes directly in the government of American society, but it should be considered the first of their political institutions…Christianity reigns, without any obstacles, the universal faith.” That statement holds true today. Compared to people in other countries, Americans are more likely to belong to a church, more likely to attend religious services, more likely to believe in God and more likely to claim that religion is important in their lives (Lipset 1996; Manza and Brooks 1997). As one recent international survey reports, 55 percent of Americans say religion is very important in their lives compared to 16 percent of the British, 14 percent of the French and 13 percent of Germans; 44 percent of Americans attend church at least once a week compared to 27 of the British, 21 percent of the French, four percent of Swedes and three percent of Japanese; 94 percent of American believe in God compared to 46 percent in Sweden, 61 percent in Britain and 56 percent in France. Finally, 76 percent of Americans believe in life after death compared to 39 percent in France, 45 percent in Britain and 38 percent in Germany (Inglehart and Norris 2004). Further, although the United States has never had a church-based political party equivalent to Europe’s Christian Democratic parties, many political movements have been religious-based. In the 1830s evangelical Protestantism gave rise to two political movements – the American Temperance Society, which at its peak had a national network of more than 8,000 auxiliaries and 1.5 million members, and the American Anti-Slavery Society, which claimed 1,838 auxiliaries across the North and 120,000 members. These organizations sponsored a network of church schools, published religious tracts and established a national infrastructure of religious resources in every state and territory (Young 2002). Only political parties and fraternal organizations could boast similar institutional penetration but they were unorganized and decentralized in comparison. In the early twentieth century, the Anti-Saloon League led the fight for prohibition (Burke 1908; Szymanski 2003). The League was conceived by the pastor of a Congre-

239

gational church in Chicago and organized and managed by a synod of churches. The majority of League leaders were trained ministers who used the pulpit to promote prohibition, appeal for financial support and provide information on candidates’ positions preceding elections. As one observer noted, “the Anti-Saloon League is merely an agent of the churches” (Chalfant 1923: 280). At its peak, the League operated in 43 states and had over 1,000 paid employees (Jackson 1908). The civil rights movement of the 1960s was rooted in the black church (McAdam 1999; Morris 1984). As Morris (1999: 524) explains, “The Black Church, which had a mass base and served as the main repository of Black culture, proved to be capable of generating, sustaining and culturally energizing large volumes of protest.” Its leadership arm, the Southern Christian Leadership Conference (SCLC), coordinated nonviolent direct action through churches across the South and worked with student groups to organize sit-ins and demonstrations (Calhoun-Brown 2000; McAdam 1999; Morris 1984). The SCLC was led by a young black minister, the Reverend Martin Luther King Jr. who mobilized support for the movement by drawing on familiar Christian themes such as forgiveness and charity coupled with appeals to Gandhian principles (McAdam 1996). Religion has also had a significant effect on voting behavior in the United States, often more so than class position. As Manza and Brooks (1997: 73) conclude in their study of U.S. voting behavior from 1960 to 1992, “the religious cleavage appears to be nearly twice the magnitude of the more widely debated class cleavage.” From the New Deal until the 1960s, Catholics, Jews and evangelical Protestants formed the core constituency of the Democratic party while mainstream Protestants were solid Republicans. After 1960 liberal Protestants drifted toward the Democrats while evangelicals began moving toward the Republican party. In the last quarter of the twentieth century, a new religious cleavage emerged, based not on denominational membership, but on level of commitment to orthodox religious beliefs. Increasingly, doctrinal conservatives identified with the Republican party, forming an influential conservative coalition that has pushed the party to the right and, according to some critics, converted the Republican party into a theocracy (Layman 1997; Phillips 2006). As Morone (2003: 26) argues, liberal political history underestimates the “roaring moral fervor at the soul of American politics.”

240

9.2

Religion and Social Welfare

Religious groups have been involved in social welfare in complex and varied ways. In the 19th century, Protestant sects sponsored a diverse network of charities and clubs that provided health care and social welfare services, largely independent of public funds (Morgan 2006). The Progressive reform movement in the 1910s was led by Protestant social reformers, who believed that public power was needed to protect the common welfare. Progressive organizations sponsored state laws for minimum wages, old age pensions, insurance against industrial accidents, and limits on the working hours of women and children. Catholic Progressives supported a Program of Social Reconstruction, based on the 1891 papal encyclical Rerum Novaru, which championed a living wage and state-level insurance for illness, disability, unemployment and old age (Derickson 2005). Yet the Catholic Church opposed the Sheppard-Towner Act of 1921, which funded health education programs for pregnant women and new mothers, for fear it would lead to a federal effort to promote birth control (Morgan 2006). The progressive spirit waned in the 1920s but was revived by the Great Depression of the 1930s. The Federal Council of Churches, the voice of liberal Protestantism, enthusiastically supported the New Deal, because it “embodied many of the social ideals of the churches” as did the Catholic Church (Quoted in Fones-Wolf 1994: 219). American cardinals openly expressed dissatisfaction with capitalism and many Catholic magazines repudiated the free enterprise system, couching support for New Deal programs in the language of ‘social solidarity’ with the poor (Fones-Wolf 1994: 220). In the same period, however, the Christian Conservative Crusade, led by right-wing radio evangelist Father Charles Coughlin, opposed the welfare state and other liberal ideals (Lienesch 1982: 408). The post-World War II era witnessed a revival of religious fervor. Church membership grew rapidly, religious books led best seller lists and evangelical rallies drew thousands. Many religious groups were intensely anti-communist because of communism’s expressed antipathy toward organized religion. When President Harry Truman proposed a plan for national health insurance in 1946, he faced opposition from several religious groups. The Catholic Church, which ran 775 hospitals across the nation, was against the Truman plan on the grounds that it could lead to “a sti-

241

fling of initiatives, a coercion of independence and a concentration of power,” according to Father Alphonse Schwitalla, President of the Catholic Hospital Association.1 Two evangelical organizations, the Christian Anti-Communism Crusade and the Church League of America, also fought against Truman’s plan, which they decried as a communist-inspired plot (Wilcox 2000). Throughout the 1950s the Catholic Church maintained a low political profile on social welfare issues, devoting most of its energies to lobbying for state aid for Catholic schools and hospitals (Reichley 1986: 32). In 1959, however, the Catholic Church cautioned that a bill to expand day care services for AFDC recipients could undermine the traditional family. According to the Monseigneur Raymond Gallagher, Secretary of the National Conference of Catholic Charities: We believe it is un-American to place such large numbers of children under governmental care in order to free so many mothers to enter the labor market … I would think that the nature of our society is such that it would first have made every effort possible to train men to fill the job opportunities that are available. Putting it another way, who is to say that the United States can endure a blow to family life of this proportion where indiscriminate recruiting of mothers of small children for industry continues without giving evidence of having thought through to the very ultimate the effect that this has on family life generally.2

More fundamentally, expanded federal funding could interfere with the principle of subsidiarity. Catholic Charities agencies operated over 50 day care centers and did not want the government to intervene in how these facilities were run.3 To mitigate these concerns, Congress allowed religious organizations to run the expanded day care programs. By 1965 17 states operated day care centers through public welfare

1 Testimony of Father Alphonse Schwitalla, President of the Catholic Hospital Association, National Health Program, Hearings before the Subcommittee of the Committee on Labor and Public Welfare, U.S. Senate, 80th Cong, 1st Sess on S. 545 and S. 1320. Part 1, May 21, 1947: 289. 2 Letter from Monseigneur Raymond Gallagher to Mrs. Katherine B. Oettinger, Chief, Children’s Bureau, Department of Health, Education and Welfare, April 27, 1962. National Archives, Record Group 102, Records of the Children’s Bureau, Central File 1958–62, Box 832, File: June 1960–1962, Day Care of Children of Working Mothers. 3 Memorandum Re: Day Care from Katherine Oettinger, February 1, 1965. National Archives, Record Group 102, Records of the Children’s Bureau, Central File 1958–62, Box 1005, File: July 1963–1968.

242

agencies, 33 provided family day care and 31 purchased day care from existing centers, many in churches.4 Churches were also involved in President Lyndon Johnson’s War on Poverty, created by the Economic Opportunity Act of 1964. In the South, anti-poverty funds went to local civil rights organizations, many of which were closely linked to black churches. The most ambitious community action program was Systematic Training and Redevelopment, or Operation Star, a demonstration project initiated by two Catholic priests and run by the Catholic church. Operation Star established an alternative social welfare system in Mississippi to train the hardcore unemployed and the functionally illiterate. With offices, staff, access to nationwide news media and an established bureaucratic hierarchy, the Catholic Church was the only organization with the capacity to circumvent Mississippi’s segregated educational system (Quadagno 1994). Thus, from the 1910s to the 1960s religious groups ran social services, often with support from states and the federal government. They also intervened both for and against government welfare benefits but these activities were not coordinated and were only marginally influential in policy decisions.

9.3

The Rise of the Christian Right

Although the first suggestion of a religious political movement occurred in the 1968 election when white southerners, mostly Baptists, defected from the Democratic party, evangelicals did not became a politically visible voting bloc until the 1976 election when one of their own, Jimmy Carter, a born-again southern Baptist, ran for the Democratic party nomination. Evangelicals provided Carter’s margin of victory over his opponent, President Gerald Ford, but their support was short-lived as it became apparent that Carter was more liberal than he had appeared during his campaign (Reichley 1986: 25; Manza and Brooks 1997: 61). They were especially angered over Carter’s decision to deny tax exemptions to white Christian academies that were springing up throughout the South in the wake of school desegregation efforts.

4 Letter from Betty Jane White, Secretary in Children’s Program, Board of Christian Education to Gertrude Hoffman, Oct. 23, 1964. National Archives, Record Group 102, Records of the Children’s Bureau, Central File 1963–66, Box 1005, File: July 1963–1968.

243

The political potential of evangelicals was noted by the New Right, a self-selected term to distinguish its adherents from the “slightly effete conservative leadership of the East Coast” (Crawford 1980: 7). The New Right consisted of a loose knit coalition of large multipurpose groups like the American Conservative Union and the Conservative Caucus as well as think tanks like the Heritage Foundation, periodicals and publishing houses, tax-exempt legal foundations and a vast array of single issue groups. The New Right goal was to mobilize conservative opinion around free market economics, anti-communism and national defense and turn that opinion into electoral clout (Bruce 1990: 479). Well-funded and well-organized but lacking a grassroots force, New Right leaders began wooing evangelicals (Lienesch 1982). Evangelicals were most concerned with protecting Christian schools from Internal Revenue Service investigations over the issue of racial balance but were also vehement in their opposition to communism (Oldfield 1996: 71). They believed the United States had an important role in God’s plan for worldwide evangelism and saw communism as a threat to the nation and to evangelicalism (Oldfield 1966: 57). New Right leaders introduced evangelicals to modern direct mail fund raising, media manipulation and grassroots campaign tactics, helping the conservative Christian movement to grow in political sophistication and influence. Still unclear was exactly how the movement would fit into partisan politics (Reichley 1986: 26). The answer came in the wake of Roe v. Wade, the 1973 Supreme Court decision that legalized abortion.

9.4

Religious Conservatives in US Welfare Politics

9.4.1 Banning Medicaid Funding of Abortion The Roe decision legalizing abortion greatly altered the number and type of religious groups involved in political action (Blanchard 1995). Before Roe the Catholic Church was cautious about when and how it weighed in on political debates, fearing the public might view any intervention as a violation of the separation of church and state. For example, while the National Catholic Welfare Council, an organization of bishops, issued statements on racial equality and justice for farm workers in the 1960s and came out against the Vietnam War in 1973, the church as an institution did not

244

actively pursue these social issues. Abortion was different, because it represented what the Catholic Church defined as a fundamental moral issue. As Cardinal Krol explained in testimony before the Subcommittee on Constitutional Amendments, the right to life was “a basic human right which must undergird any civilized society.”5 Other religious groups that shared this position included the Greek Orthodox church, the American Association of Evangelicals, Orthodox Presbyterians, the American Association of Christian Schools, and Baptists for Life, a grassroots organization formed in 1973, whose members included Southern Baptists, American Baptists, the General Assembly of Regular Baptists, the Independent Baptist Churches and the National Baptists Convention.6 For the Mormon Church, abortion was “one of the most revolting and sinful practices in this day when we are witnessing the frightening evidence of permissiveness leading to sexual immorality.” The Missouri Synod of the Lutheran Church testified that “without the right to life, all other rights are meaningless.”7 Another Christian organization, the American Life League (ALL), advocated a constitutional amendment outlawing abortion without exception even if a pregnancy threatened a woman’s life (Blanchard 1995; Diamond 1995; Maxwell 2002). Following Roe the Catholic Church officially sanctioned civil disobedience and embarked on a sustained effort to change the law (Burns 1992). In 1975, the National Conference of Catholic Bishops adopted a sophisticated blueprint for political action and spent a massive amount of money to mobilize the church’s infrastructure against abortion rights (Tribe 1990). In each state bishops’ representatives formed committees to coordinate political efforts in the diocese. In each parish, pro-life groups worked toward a constitutional amendment to overturn Roe. So as not to jeopardize its taxexempt status, the Catholic Church created a separate, multi-denominational organization, the National Right to Life Committee, to support pro-life groups behind the scenes. The NRLC sought to overturn Roe but also pursued a short-term strategy to

5 U.S. Senate, Statement of John Cardinal Krol, Archbishop of Philadelphia before the Subcommittee on Constitutional Amendments, Judiciary Committee, March 7, 1974, p. 153. 6 House of Representatives, Hearings before the Subcommittee on Civil and Constitutional Rights, H521–2, Proposed constitutional Amendments on Abortion, Part 1, Feb. 4, 1976. 7 U.S. Senate, Statement of David McKay before the Subcommittee on Constitutional Amendments, Judiciary Committee, March 7, 1974, p. 318.

245

erect whatever barriers might be legislatively possible to limit abortion access (Rohlinger 2002; Rohlinger and Meyer 2005). The first pro-life victory came in 1976 with a ban on Medicaid funding of abortion. Medicaid, the joint federal/state program of health insurance for the poor, paid for nearly one third of all legal abortions performed in the U.S., mostly on AFDC recipients. Pro-lifers viewed Medicaid-funded abortions as government-subsidized murder. If they could not outlaw abortion entirely, they could try to stop the federal government from paying for it. The Medicaid funding ban was introduced in 1974 in the House of Representatives by Rep. Dewey Bartlett (R-OK) as an amendment to that year’s appropriations bill for the Departments of Labor and Health, Education and Welfare (HEW). The appropriations bill contained the federal share of funds to the states for AFDC and Medicaid. Following Roe, several states had passed measures banning the use of state Medicaid funds for abortion, but fifteen states still paid for abortions. The Bartlett amendment specified that no federal funds could be use to pay for abortion except when necessary to save the life of the mother. As Bartlett, explained, tax dollars should not be used to pay for a procedure that many people vehemently opposed. Rather there needed to be “better more meaningful care for the poor, more meaningful assistance to the unwed mother, and a public consensus to improve the lot of all unwanted, including unwanted children.”8 When the bill reached the Senate, opponents argued that an appropriations bill was not the proper place to consider such an important measure. The committee’s purpose was to appropriate funds, not to change the authorization of any government programs. Further, cutting off Medicaid funds for abortion would discriminate against poor women: The most fundamental purpose of the Labor-HEW appropriation bill is the equalization of access to education and health services and the elimination of practices which, intentionally or not, discriminate against individuals on the basis of income.9

8

Congressional Record, Sept. 17, 1974: 31454.

9

Congressional Record, Sept. 17, 1974: 31456.

246

That view prevailed and the Bartlett amendment was defeated in the Senate by a vote of 34–50. In 1976 pro-life forces gained a new champion, Rep. Henry Hyde (R-IL) who again tacked a strict Medicaid abortion funding ban – with no exceptions – onto that year’s Department of Labor/HEW appropriations bill. The Hyde amendment passed in the House, 207–167 over the objections of the congressional black caucus whose leader declared the vote marked “the process of establishing a second post-Reconstruction period in this country” (quoted in Tribe 1990: 155). After a provision was inserted allowing abortion when the mother’s life was endangered, the bill also passed in the Senate. The legislation affected all poor women, but its impact fell most heavily on African American women who represented a disproportionate share of AFDC recipients and on whom a disproportionate number of abortions were performed. Although the Medicaid abortion ban was a minor provision in terms of the amount of funds involved, it was the first welfare benefit cut since the Social Security Act was enacted in 1935. More importantly, the Hyde amendment represented a retrenchment in democratic rights: it rescinded a poor woman’s constitutional right, granted by the Supreme Court, to choose whether to give birth to a child. Because the Medicaid abortion ban was included in an appropriations bill, it would have to be re-enacted each year, guaranteeing a continuing political struggle. For the pro-life movement, a more permanent solution had to be found. Many of the state statutes limiting abortion rights had been declared unconstitutional, but in 1977 the Supreme Court ruled in the case of Maher v. Roe that states could deny funding for non-therapeutic abortions, a decision that appeared applicable to federal funding as well. Thus, the question of whether these state restrictions were constitutional was resolved in favor of the pro-life movement. Another victory occurred in 1980 when the narrow restriction “severe or long-lasting physical health damage as determined by two physicians” was dropped from that year’s federal appropriations bill (Tribe 1990: 156).

247

9.4.2 The Rise of the Christian Conservative Movement in the Republican Party The Hyde amendment not only signified a test of the political might of the pro-life movement but more generally the emergence of a socially conservative mood centered around a backlash against big government, resentment toward welfare recipients and concern that society had become lax about moral issues and sex. Yet a cohesive, conservative religious-based movement had not yet blossomed. In the 1980 presidential election the Republican Party openly wooed the religious right. The Republican platform supported school prayer, family values and a constitutional amendment to restore the “right to life for unborn children.”10 The Republican presidential candidate, Ronald Reagan, attended a national affairs briefing sponsored by the Religious Roundtable and promised that, in his administration, “traditional moral values” would be “reflected in public policy” (Reichley 1981–82: 543). Reagan was rewarded for his efforts with the endorsement of the Moral Majority, an evangelical organization created by the Reverend Jerry Falwell in 1979 to defend Christian schools. Led by pastors in the Baptist Bible Fellowship, the Moral Majority played a visible role in Reagan’s campaign, organizing voter registration drives, issuing moral “report cards” and attacking liberal members of Congress. Reagan soundly defeated his rival, President Jimmy Carter, and the Democrats lost control of the Senate for the first time since the 1940s (Oldfield 1996). Uncontrollable inflation, high unemployment and a hostage crisis in Iran were primarily responsible for Carter’s defeat, but the defection of evangelicals and Catholics from the Democratic party was also a factor. When Reagan ran for reelection in 1984, he won 80 percent of the evangelical vote as well as 55 percent of the Catholic vote. Throughout the 1980s the pro-life movement advanced its agenda with some success (Diamond 1995). The Reagan administration instituted a “gag rule,” which forbid personnel at federally-funded clinics from discussing abortion with clients, and the “Mexico City Policy” (which prohibited nongovernmental organizations receiving federal funding from providing or promoting abortion services). Reagan also banned the use of fetal tissue in medical research, stymied the importation of emergency con-

10 Republican Platform, 1980 (Washington, D.C.: Republican National Committee, 1980), pp. 10–15.

248

traception into the U.S. and prohibited privately funded abortions in overseas military hospitals (Rohlinger 2004). In the 1988 presidential election, the Pentacostal televangelist preacher, Pat Robertson, decided to seek the Republican party presidential nomination. During his campaign, Robertson trained activists in every precinct in the country, creating a grassroots evangelical network to raise funds and recruit volunteers (Bruce 1990: 484; Shibley 1998: 80). In response his opponent, Vice President George Bush cultivated conservative Christians, meeting individually with evangelical leaders, appointing a religious liaison and discussing his own faith in public settings (Oldfield 1996: 144– 45). The Bush campaign organization also identified the 215 largest southern churches, located a supporter in each church and distributed a special video directed at evangelicals to be played in churches. Robertson lost badly in the primaries, but his attempt to capture the party had longlasting consequences. During his campaign, evangelicals gained control of many GOP state and local organizations, embedding religious conservatives deeply into the Republican party machinery. Further, the Republican Party adopted many of the Moral Majority’s campaign issues and invited conservative Christian leaders to speak at the Republican National Convention. As Robertson recognized: Could it be that the reason for my candidacy has been fulfilled in the activation of tens of thousands of evangelical Christians into government?…for the first time in recent history, patriotic, pro-family Christians learned the simple techniques of effective party organizing and successful campaigning. Their presence as an active force in American politics may result ultimately in at least one of America’s major political parties taking on a profoundly Christian outlook in its platforms and party structure (Quoted in Oldfield 1996: 183). However, many Republican party regulars found disturbing their evangelical allies’ confrontational tactics and insistence that candidates adopt uncompromising positions in exchange for political support (Leege 1992). In some states, extreme Christian Right candidates so alienated moderate Republicans that they supported the Democratic candidate. Thus, interparty competition undermined Republicans electoral chances (Hacker and Pierson 2005).

249

Bush won the election with 70 to 80 percent of the evangelical vote, and it seemed clear that the fortunes of the Republican party were now inextricably linked to the Christian conservative movement. Although the Moral Majority collapsed in 1989 following Robertson’s defeat and a scandal involving a televangelist preacher, the Moral Majority’s successor, the Christian Coalition, shaped this nascent religious conservative movement into a cohesive political force. The Christian Coalition was founded in 1989 by Robertson who turned the organization over to Ralph Reed (Shibley 1998). Between 1990 and 1995 it grew from 25,000 members to 1.6 million members with 1,600 local chapters, a national lobbying staff headquartered in Virginia and 50 independently incorporated state organizations (Rozell and Wilcox 1996; Oldfield 1996). Under Reed’s guidance, the Christian Coalition became a sophisticated political machine, dedicated to training Christians for political action, keeping them informed on legislative issues and promoting a conservative policy agenda (Wilcox 2000). The chapters organized grassroots activities and sponsored political candidates. Starting with lower level offices such as school boards, city councils and county election, the Christian Coalition created a pool of candidates qualified to run for higher offices in future races (Oldfield 1996; Usher 2000). Whereas in the past openly evangelical candidates had never been able to muster more than 35 or 40 percent of the vote against GOP regulars in primary races, under Reed’s direction they learned to eschew religious language that made them appear intolerant and use secular arguments to support the same position (Guth 1995: 9; Rozell and Wilcox 1996). Reed also took advantage of differences in state parties delegate selection processes to capture the Republican party machinery. Unlike the Democratic party which mandates that all potential delegates get prior approval from a campaign, Republicans have no such mandate. Of the five methods of selecting Republican delegates, three are permeable to grassroots activists. The first of these is where state parties select delegates directly in party caucuses. In the second process, state parties use precinctlevel caucuses to choose delegates to district conventions, which then choose delegates to state conventions where national delegates are selected. Finally, state parties may choose delegates in one stage at a state party convention. In each of these processes, caucus, caucus/convention and convention, grassroots activists can control party meetings with a small number of participants (Usher 2000). By mobilizing

250

Christians to attend precinct party caucus meetings, the Christian Coalition succeeded in taking control of state delegates to the national convention in states with permeable procedures. In these states far right delegates could ascend to positions of power without making concessions to more moderate groups. The Christian Coalition also formed alliances with other like-minded groups like the anti-feminist Concerned Women for America (CWA), founded in 1979 by stay-athome Christian mothers and home schoolers (Rozell and Wilcox 1996). Built through women’s prayer groups and bible study groups, CWA grew quickly and became involved in litigation and national politics (Rohlinger 2002, 2004). Another ally was Focus on the Family, the national umbrella organization for 35 state-affiliated family policy councils. Led by James Dobson, a radio minister whose program was broadcast on more than 4,000 stations, Focus had a staff of more than 1,300 people, a $78 million budget and a mailing list of 2.5 million (Wilcox 2000: 64) Both organizations held conferences, workshops and conventions, graded candidates based on their voting records, sponsored voter registration drives in churches and provided activists for state caucuses and conventions. Reed also created the Catholic Alliance to strengthen political ties with conservative Catholics (Oldfield 1996; Rohlinger 2004). The 1992 presidential election was a watershed in the ascendance of religious conservatives. At the Republican convention more than three hundred Christian delegates helped shape the party platform (Regnerus, Sikkink and Smith 1999). Christian Right leaders also were given prominent billing on the program with Jerry Falwell declaring a religious war “for the soul of America” and Pat Robertson condemning the Democratic party candidate, Bill Clinton, for his “radical plan to destroy the traditional family” (Quoted in Oldfield 1996: 89). Yet Bush only received 62 percent of the white evangelical vote, and turnout among evangelicals was low overall. He had broken his promise not to raise taxes and had invited gay rights leaders to attend the signing of the Hate Crimes Bill. The Democratic party’s candidate, Bill Clinton, won by a narrow margin (Oldfield 1996).

251

9.4.3 Abortion Politics and Health Security Clinton’s victory was a stunning blow for the pro-life movement. Pro-lifers had won a victory with the 1989 Supreme Court decision in the case of William Webster v. Reproductive Health Services. Webster upheld a restrictive Missouri law that prohibited the use of public funds for abortions except to save a woman’s life, required that a fetal viability test be performed if a woman was at least 20 weeks pregnant, and stated that human life begins at conception. As Webster opened the door to further restrictions on abortion, religious groups immediately introduced state legislation to test the new limits (Tribe 1990; O’Connor, Orloff and Shaver 1999). These gains were threatened by Clinton who backed a proposal to repeal the Hyde amendment entirely and supported the Freedom of Choice Act, which prohibited restrictions on third-trimester abortions and 24-hour waiting periods. The Clinton administration also overturned parental consent requirements and signed the Freedom to Access Abortion Clinic Entrances into law. When Clinton sought to fulfill his campaign promise for universal health care, however, his pro-choice stance on abortion became an impediment. Clinton’s Health Security plan would have guaranteed universal coverage through a complex arrangement that included a National Health Board and regional purchasing alliances. It also would have given the large Health Maintenance Organizations a central role in health care financing (Quadagno 2004; 2005). A key controversy concerned the inclusion of abortion under “pregnancy-related” services. Health Security would require all health plans to cover abortion, effectively nullifying the Hyde Amendment and the laws of 37 states that restricted state funding of abortion. It would also cap rate increases in health insurance premiums, a provision that pro-life forces feared could lead to health care rationing or the withholding of care or even infanticide and euthanasia. As Wanda Franz (1993a: 3), the President of NRLC at the time, warned: Children would grow up finding out that their siblings were killed and that they themselves could have been killed, too, because the government’s health plan offered their mothers abortion as an “option.” Mothers would find themselves under pressure to abort their children because the bureaucrats administrating (it) will be under pressure to save money by avoiding more costly services for birth.

252

Health Security posed a conflict for the Catholic Church. In 1981 the American Catholic Bishops had published a pastoral letter citing Pope John XXII’s encyclical Pacem in Terris which listed “medical care among those basic human rights which flow from the sanctity and dignity of human life … . Catholics “had a responsibility to contribute to the development of a just and humane national health policy.”11 The letter had warned, however, that Catholic institutions should not permit practices that violated Catholic teaching, particularly regarding abortion. Since Health Security included abortion services, the bishops initiated a letter-writing campaign opposing that feature of the bill. In testifying on behalf of the Catholic Bishops of the United States, Helen Alvare explained: We believe that the fundamental right to health care is grounded in the right to life, a right of all human beings regardless of age, sex or condition of dependency. True healthcare reform, therefore, could never promote attacks on life itself. Tragically, the administration’s plan, by including abortion as a mandated benefit, undermines(s) health reform at its root … abortion advocates … have not hesitated to burden the entire momentum for health care reform with their agenda.12

The Southern Baptist Convention also vehemently opposed including abortion in health care reform. As the Convention’s Director for biomedical and life issues, Ben Mitchell, warned: (B)ecause of our historically attested and committed opposition to public funding for elective abortion, the Southern Baptist Convention is prepared to marshal as many of our resources as necessary to oppose vigorously any health plan that includes coverage … This is such a critical concern for us that we are prepared to oppose any reform which is otherwise excellent if it includes coverage for elective abortions.13

Ralph Reed (1994: 29), head of the Christian Coalition, linked abortion to broader conservative issues, claiming that “The Clinton plan is really a Trojan horse for a not-

11 “Health and Health Care.” A Pastoral letter of the American Catholic Bishops, Nov. 19, 1981, United States Catholic Conference, p. 5, 11. 12 “Hearings on Women’s Health.” Testimony of Helen Alvare, Secretariat for Pro-Life Activities, National Conference of Catholic Bishops, Hearings before the Subcommittee on Health and the Environment, Committee on Energy and Commerce, House of Representatives, 103rd Congress, 2nd Session, January 26, 1994, pp. 171. 13 “Hearings on Women’s Health.” Testimony of C. Ben Mitchell, Director for Biomedical and Life Issues, Southern Baptist Christian Life Commission, Hearings before the Subcommittee on Health and the Environment, Committee on Energy and Commerce, House of Representatives, 103rd Congress, 2nd Session, January 26, 1994, pp. 186.

253

so-hidden agenda to expand government bureaucracy, pay for abortions with tax dollars, and promote a radical social agenda.” Concerned Women for America echoed these viewers, calling on Clinton to “protect religious freedom, defend the sanctity of all life,” born and unborn, and “protect marriage and the home.”14 Religious conservatives launched a massive campaign against Health Security, lobbying legislators, distributing postcards and letters through churches and developing a national media campaign (Diamond 1994; Hardisty 1999). The Christian Coalition alone spent $1.4 million to defeat Health Security. By the spring of 1994 Clinton’s plan was dead. Its demise is usually attributed to the opposition of the health insurance industry and small business organizations, who hired thousands of lobbyists and public relations firms and spent more than $140 million to kill it (Quadagno 2004, 2005; West and Loomis 1999). However, the significance of religious conservatives in stirring up grassroots opposition should not be underestimated. In the 1994 congressional elections, the Christian Coalition organized a Road to Victory plan (Diamond 1995). Thirty-eight percent of evangelicals who attended church on a regular basis reported being urged to vote for Republican candidates by church leaders (Rozell and Wilcox 1996). In Oklahoma, a state where 50 percent of voters were Biblical literalists, the Christian Coalition distributed over 100,000 voter guides, reporting candidates’ views on tax increases, gun control, school prayer and tax funding for abortions. Republicans swept the offices of governor and lieutenant governor, picked up seats in the state legislature and won a 7–1 majority in the House. All five of Oklahoma’s Republican House members and both Senators received strong backing from evangelicals (Bednar and Hertzke 1995: 11–13). The NRLC also worked to defeat pro-choice politicians in both parties. In all religious conservatives gained control of 18 state Republican parties and were influential in 13 more. Nationally, the Republican party recaptured the Senate and broke the Democratic party’s 40-year hold on the House of Representatives (Skocpol 1996). One-quarter of House Republicans claimed to be evangelicals. Of the 137 congressional candidates supported by the NRLC, 93 won: “Not a single pro-life incumbent governor or member of Congress was defeated by a pro-abortionist. Not one!” (Franz 1994b: 3). 14 An open letter to President Clinton, Concerned Women of America, USA Today, January 25, 1994, p. 5A.

254

9.4.4 Reform of AFDC Following the election, the new Speaker of the House, Newt Gingrich (R-GA), announced he would bring to fruition the Republican Party’s “Contract with America” (Diamond 1995). Gingrich would abolish the Department of Education (to punish teachers’ unions), cut spending on Medicare, Medicaid and education, slash the Earned Income Tax Credit for the working poor, and enact a tough welfare reform bill. Reform of AFDC was the first priority. Out-of-wedlock births had increased from 5.3 percent of all births in 1960 to 32.2 percent by 1996;15 almost one-third of these births were to teenage mothers, a trend Republicans attributed to AFDC: The plain truth is that if we are ever going to cope with the deficit, and the social programs that inflate it, we are going to have to begin with a very different view of human nature and human responsibility in relation to such issues as criminality, sexuality, welfare dependency, even medical insurance. (Kristol 1994: ix)

Phyllis Schafley, the head of the Eagle Forum, called AFDC a “grossly unjust and immoral system… millions of people were taught that they were entitled to pick the pockets of law-abiding, taxpaying families if they met two conditions: they didn’t work and they were not married to someone who does work” (Schlafly 1996: 1–2). The Christian Coalition supported many provisions of the Republican Contract with America but felt it paid insufficient attention to moral issues. They responded with an alternative “Contract with the American Family” that included a religious equality amendment to the Constitution, a family tax credit, spending cuts for the arts, humanities and education (all institutions that subjected Christian children to secular humanist values), restrictions on abortion and expansion of the Earned Income Tax Credit for two-parent households. The one priority on which there was clear agreement was AFDC reform. AFDC offended evangelicals’ basic moral values – that sex should take place within marriage and be tied to procreation, that the father should hold patriarchal authority in the family and that children should be strictly disciplined (Oldfield 1996: 60). Evangelicals believed that AFDC subsidized family breakup, encouraged promiscuity and allowed men to shirk their familial responsi-

15 U.S. Department of Health and Human Services. 1996. Indicators of Welfare Dependence and Well-Being: Interim Report to Congress. Washington, D.C.: U.S. Department of Health and Human Services.

255

bilities. AFDC reform would empower “private citizens to free themselves from dependency on government programs” (Reed 1994: 10). For the next two years the Christian Coalition gave welfare reform priority even over anti-abortion activities. Members visited senators and their staffs and camped out in the Senate hallways during floor debates (Reese 2005). Seeking to bridge the conflicts with traditional Republicans that had fractured the party, Reed and other Christian leaders also met with heads of conservative think tanks to carve out positions and strategies. President Clinton had promised to “put an end to welfare as we know it” during his 1992 presidential campaign. In his initial version of welfare reform, AFDC recipients would receive assistance for two years, then would have to find a job or “start earning their way through community service” (DeParle 2004: 103). However, they would also receive medical insurance, child care and job training. When the Republicancontrolled House of Representatives passed a harsh welfare reform bill (HR 4) in March 1995, it contained no guarantee of health insurance or child care. Single mothers under age 18 would be ineligible for any benefits and states could prohibit payments to single mothers under 21. A “family cap” would deny additional payments to welfare recipients who bore a child while receiving benefits. Ironically, these stringent provisions generated divisions among Christian groups. The Christian Coalition favored even harsher measures and sought to end all cash, food and housing assistance to any unmarried mother younger than 26 and to require mothers with children above pre-school age to work (DeParle 2004: 115–16). The Christian Coalition also endorsed bonuses to poor mothers who married, “responsible fatherhood” programs, and mandatory child-support (Reed 1994: 10). Concerned Women of America agreed that these provisions would “break the cycle of dependency” and demand “moral accountability” from the poor (Reese 2005).16 The Catholic Church opposed many of these provisions, however, on the grounds that they would eliminate an important safety net for the poor. When Wisconsin en-

16 Child support enforcement also won the support of liberal feminist groups like the National Women’s Law Center and the Women’s Legal Defense Fund who believed that fathers should be forced to support their children.

256

acted a state welfare program that became the blueprint for national welfare reform, the Catholic bishops declared that “calls for affirming personal responsibility must not become a signal to abandon the poor … Welfare reform that is purely pragmatic, based only on reducing costs and that is conceived in isolation from the needed values of our whole society will not be fruitful” (Yamane 2005: 2). Catholic Charities USA, the social service arm of the Catholic Church and the nation’s largest private social service network, had 1,400 agencies, 272,000 staff and volunteers and served 10.6 million people a year. Testifying on behalf of Catholic Charities, the Reverend Fred Kammer declared: We are opposed to a family cap provision as untested, dangerous and cruel. We oppose rigid lifetime limits for AFDC in the face of chronic and systemic unemployment … The churches and charities, beginning with our own, say this would produce a tidal wave of hungry and homeless kids and moms, a torrent of newly abused children and profoundly wrong social and moral outcomes.17

When the Personal Responsibility Act passed in the House, the U.S. Conference of Catholic Bishops noted, “We cannot support reform that … eliminates resources that have provided an essential safety net for vulnerable children.”18 The NRLC, too, charged that the family cap and ban on payments to teenagers would encourage abortion (Franz 1995a). Although AFDC did subsidize immoral and irresponsible behavior, the child could not be held accountable: “How can it be morally acceptable to punish the child, even execute him by abortion, for the sins of his parents?” (Franz 1995b: 3). To pacify pro-life groups, the Republicans agreed to include a bonus to states that reduced illegitimate births without increasing abortion. The final measure that President Clinton signed into law on July 31, 1996, the Personal Responsibility and Work Opportunity Reconciliation Act (PRWOA), replaced AFDC with Temporary Assistance to Needy Families (TANF). PRWOA converted AFDC into a program of block grants to the states at a fixed level of federal funding,

17 Statement of Rev. Fred Kammer, President Catholic Charities USA. Hearing before the Subcommittee on Human Resources, Committee on Ways and Means, House of Representatives. 104th Cong, 1st Sess, January 13, 1995, Part 1. Washington, D.C. U.S. Government Printing Office, pp. 714–17. 18 U.S. Conference of Catholic Bishops. 1995. “Welfare Reform.” Washington, D.C.: Office of Social Development and Peace.

257

eliminating the federally guaranteed entitlement to public assistance that had been in place since 1935 (Walker 1996; Gilbert 2002). It required welfare recipients to engage in work or training after two years. It also gave the states considerable latitude in determining how to allocate funds between cash benefits, child care, job training and education and in designing incentives and sanctions to motivate welfare recipients to work. Yet PRWOA did not require states to provide education or job training to mothers displaced from welfare and allowed states to waive the work requirements. As a disgusted Rep. James McDermott (D-WA) proclaimed: “every bit of work requirement in this bill is a fraud.”19 While Congress was most concerned with the work provisions, evangelical Christians focused on “family values” and the moral behavior of poor women. The “findings” reported in the opening paragraph of the bill declared that marriage was “the foundation of a successful society,” responsible parenthood was integral to successful child rearing, and the prevention of out-of-wedlock pregnancy was an “important government interest.”20 To this end PRWOA allocated $100 million to reward states that reduced illegitimate births and provided $50 million annually for abstinence education and marriage promotion programs. As a concession to pro-life groups states were allowed to retain benefits for teen mothers at their discretion but could refuse assistance to teenagers who were living on their own or to mothers who failed to cooperate in establishing paternity (Durham 2000: 95–96; Reese 2005). A proviso was included that the abortion rate must not rise. The Christian Coalition also won a “charitable choice” provision that allowed religious organizations to contract for service delivery “on the same basis as other governmental providers. PRWOA further specified that neither the state nor the federal government could require a faith-based organization to alter its internal governance or remove “religious art, icons, scripture or other symbols.”21

19 Remarks of Senator McDermott, Conference Report on H.R. 3734, Personal Responsibility and Work Opportunity Reconciliation Act of 1996, House of Representatives, Washington, D.C., July 31, 1996, p 4. 20 H.R. 3734, Sec. 101. Findings. 21 H.R. 3734, Sec. 104. Services Provided by Charitable, Religious or Private Organizations.

258

The conflict between evangelical Christians and the Catholic Church over the PRWOA provisions illustrates the fragility of the alliance that makes up the so-called Christian Right. While religious conservatives may be unified on some moral issues, they have fundamental disagreements on how to use economic incentives and for what ends. These disagreements make it difficult for them to frame policy issues in ways that garner support among constituents with different beliefs and values. Such conflicts were highlighted in the debate over Social Security privatization.

9.4.5 Christian Conservatives and Social Security Privatization During the 2000 presidential campaign, Republican candidate George W. Bush proposed privatizing Social Security by diverting a portion of payroll tax revenues into private accounts. Although Bush, consumed by the aftermath of the Sept. 11, 2001 terrorist attack, made no mention of privatization in his first term, when he was reelected in 2004, he interpreted his victory as a mandate to move forward. Given that evangelicals had helped defeat Clinton’s Health Security and supported Republicans on AFDC reform, Bush presumed that they would also back Social Security privatization. However, Bush was unenthusiastic about a constitutional amendment banning gay marriage, a key evangelical issue, and the Christian Coalition threatened to withhold support for his Social Security plan unless he rejected gay marriage. As a commentator for Family Voice, Robert Regier (February 27, 2002), explained: (Homosexual marriage) would also put a strain on a Social Security System that is already struggling to stay afloat. As it stands now, we barely have enough workers to support those collecting benefits. Adding more beneficiaries would help drive the last nail in the coffin on any chance that people under 40 will ever see a dime of Social Security when they retire.

Once Bush endorsed the amendment, the Christian Coalition made Social Security privatization its second highest priority (Phillips 2006: 242). “President Bush wants Social Security reform to include turning President Roosevelt’s New Deal-era retirement program into a self-financing private investment accounts system which workers could own and control” (Christian Coalition 2005). A feature article by Jessica Wadkins in CWA’s monthly magazine, Family Voice (April 1998), noted: The Bible says we must be wise in our dealings- as shrewd as serpents and innocent as doves (Matt 10: 16, NIV). God calls us to be good stewards of our resources. Therefore,

259

we should encourage our legislators to privatize Social Security. We must ensure that our nation’s retirees will receive what has been promised to them. We must defuse this fiscal time bomb NOW!

Yet grassroots Christian groups did not find these messages convincing. Polls showed that most Americans feared that privatization would destroy the safety net for the elderly poor and reduce income security for the middle class. A public opinion survey conducted after the 2000 election revealed that one of the main reasons Democratic presidential nominee, Al Gore, won forty-two percent of the evangelical Christian vote was his support for Social Security. The Social Security debate highlighted a fissure in the Christian Right/Republican party coalition and demonstrated the limitations of religious doctrine. Religion can ignite passions and move people to action but it cannot provide an all-encompassing political framework. When universalistic welfare programs, like Social Security, are attacked in the name of religion, religious support for an anti-welfare state agenda dwindles.

9.5

Conclusion

The diversity of American society coupled with a two-party system make it necessary for groups to form coalitions to achieve electoral success. Through coalitions the minority can participate in majority rule. Throughout American history religion has provided a durable base for political coalitions. Unlike the class struggle, which arises from a sense of economic injustice and worker oppression, the religious struggle arises from perceptions of moral pathology and a reaction against the suppression of religious principles. While economic conditions vary considerably and often change too rapidly to generate the intellectual frameworks needed to interpret them, religious values are deeply and permanently embedded in the way people perceive the world. When religious groups mobilize to defend cherished values and beliefs, they have certain inherent advantages: they can draw upon a coherent theology to convince adherents that their cause is just, they can use the threat of informal sanctions available to members of a congregation and they have a natural base of operations in the church (Leege 1992; Brooks 2002; Williams 2002). The church provides organizational resources and a forum for political messages, which can be transmit-

260

ted through sermons, classes and publications and reinforced through informal social interaction among members who are bound by affective ties (Regnerus, Sikkink and Smith 1999). Such characteristics make churches unique political communities (Wald, Owen and Hill 1988). For most of American history, the primary source of religious cohesion has been denominational membership. In the last quarter of the twentieth century, however, degree of commitment to orthodox religious beliefs has transcended denomination (Layman 1997). In this period, conservative Christians of varying denominations moved into politics because of perceived threats to fundamental values, especially around issues concerning sex, childrearing, abortion and the position of women. Wooed by the Republican Party, they formed a sometimes uneasy coalition with traditional Republicans concerned about fiscal responsibility and national defense. The two party system that allows voters to either support the opposing party or drop out of politics altogether has given Christian conservatives leverage within the Republican party. If Republicans fail to take positions that mobilize conservative Christians, they risk losing important organizational resources. Church and ministry networks won’t be mobilized, voter registration drives won’t take place and a valuable pool of volunteers for party work won’t be available (Oldfield 1996: 79). State variations in delegate selection processes have also given Christian conservatives the ability to insinuate themselves deeply into local and state party organizations and propel their candidates to victory and leadership positions within the Republican party, often at the expense of traditional Republicans. Of all the issues that mobilized religious conservatives, the abortion battle was most crucial. Indeed it could be argued that Roe v. Wade was the source of a fundamental realignment in American politics. It unified Catholics and evangelicals, two groups that were otherwise at odds on most political issues, and it transformed the agenda of the Republican party. The efforts of this political coalition had crucial consequences for the issues that came to the forefront in the era of welfare state restructuring. The first direct consequence was the Hyde Amendment, which rescinded a basic social and constitutional right for poor women. The battle over abortion then spilled over into Clinton’s Health Security plan, which might otherwise have won public support if abortion services had been excluded. The abortion issue and more general moral concerns permeated the debate about AFDC reform.

261

The failure of Social Security privatization demonstrates limits on what a religious political movement can achieve. Certainly, Christian conservative leaders failed to make a convincing link between privatization and gay marriage. More importantly, however, unlike Health Security and AFDC, Social Security’s political base derives from its universal constituency and from the recognition by people of working age that they will eventually be beneficiaries themselves. Over the past quarter century, Christian conservatives have helped frame welfare policy debates and reshape the political landscape more generally. Religious conservatives have influenced welfare state politics in the U.S. by framing policy debates in moral terms, penetrating deeply into the Republican Party machinery at the local, state and national level and forcing the Democratic Party to move in a conservative direction on social welfare issues. What are the broader implications of these events for welfare state theory? The current alliance between the Christian Right movement and the Republican party parallels, in many respects, the close alliance between the labor movement the Democratic party in the 1960s. Yet its impact has largely been invisible to political theorists, most probably because the focus on class-based movements has obscured the activities of religious based movements. The result has been a shift from an emphasis on politics to a focus on institutions. Thus, Myles and Quadagno (2002: 52) argue that while political parties, labor unions and social movements were central in the period of postwar World War II expansion, “during the past quarter century the variety of responses to globalization and industrialism have instead emphasized the decisive role of political institutions.” Similarly, Pierson (1994; 2000) suggests that the legacy of previous policy choices creates embedded structure and constituencies with an inherent interest in defending existing benefits from cuts. Perhaps the failure of Social Security privatization might be understood in these terms, but such a perspective is not well-suited to explaining the elimination of federal funds for abortion, the defeat of Health Security and the demise of AFDC. The peculiar path of welfare state restructuring in the United States from the 1970s to the 2000s was neither a product of a weak, left-center political coalition nor of self-serving beneficiaries defending their interests but rather of concerted action by a right wing coalition of fiscal and religious conservatives. If labor-based movements create possibilities for coalitions that push

262

parties toward the left, then these are the forces that can push parties toward the right. This is the unexamined dimension of welfare state theory.

References Bednar, Nancy and Allen Hertzke. 1995. “The Christian Right and Republican Realignment in Oklahoma.” PS: Political Science and Politics 28 (1): 11–15. Blanchard, Dallas. 1995. The Anti-Abortion Movement and the Rise of the Religious Right. New York: Twayne Publishers. Bruce, Steve. 1990. “Modernity and Fundamentalism: The New Christian Right in America.” British Journal of Sociology 41 (4): 477–496. Burke, W.M. 1908. “The Anti-Saloon League as a Political Force.” Annals of the American Academy of Political and Social Science (November): 27–37. Burns, Gene. 1992. The Frontiers of Catholicism: The Politics of Ideology in a Liberal World. Berkeley: University of California Press. Calhoun-Brown, Allison. 2000. “Upon This Rock: The Black Church, Nonviolence, and the Civil Rights Movement. PS: Political Science and Politics 33 (2): 168–174. Cato Institute. 2005. “The real issues of Social Security reform.” New York Times, June 13: 3A. Chalfant, Harry. 1923. “The Anti-Saloon League – Why and What?” Annals of the American Academy of Political and Social Science (September): 279–283. Christian Coalition. 2005. “Christian Coalition of America’s Agenda for the 109th Congress.” June 19, 2005. http://www.cc.org/issues.cfm. Clifton, Brett. 2004. “Romancing the GOP: Assessing the Strategies Used by the Christian Coalition to Influence the Republician Party.“ Party Politics 10: 475–498. Coughlin, Bernard. 1965. Church and State in Social Welfare. New York: Columbia University Press. Crawford, Alan. 1980. Thunder on the Right: The “New Right” and the Politics of Resentment. New York: Pantheon. DeParle, Jason. 2004. American Dream: Three Women, Ten Kids and A Nation’s Drive to End Welfare. New York: Viking. Derickson, Alan. 2005. Health Security for All. Baltimore: Johns Hopkins University Press. Diamond, Sara. 1995. Roads to Dominion: Right Wing Movements and Political Power in the United States. New York: The Guilford Press. Durham, Martin. 2000. The Christian Right, the Far Right and the Boundaries of American Conservativism. Manchester, UK: Manchester University Press. Eagle Forum News. 2005. “ What Happened in the 108th Congress, 2nd Session?” Eagle Forum News January, 2005: 1–10. Esping-Andersen, Gosta. 1990. The Three Worlds of Welfare Capitalism. Princeton University Press.

263

Esping-Andersen, Gosta and Kees van Kersbergen. 1992. Contemporary Research on Social Democracy.” Annual Review of Sociology 18: 187–208. Fones-Wolf, Elizabeth. 1994. Selling Free Enterprise. Urbana, IL: University of Illinois Press. Franz, Wanda.. 1994a. “The REAL Health Care Crisis Created by the Clinton Health Proposal.” National Right to Life News, Feb. 3, 1994: 3. Franz, Wanda. 1994b. “Reflections on a Political Earthquake.” National Right to Life News, Dec. 13, 1994: 3. Franz, Wanda. 1995a. “The State of the Union: The Pro-Life View.” National Right to Life News, Feb. 3, 1995: 3. Franz, Wanda. 1995b. “Welfare Reforms Must Not Increase Abortion Rates.” National Right to Life News, March 16, 1995: : 3. Gilbert, Neil. 2002. Transformation of the Welfare State: The Silent Surrender of Public Responsibility. New York: Oxford University Press. Guth, James L. 1995. “South Carolina: The Christian Right Wins One.” PS: Political Science and Politics 28 (1): 8–11. Hacker, Jacob and Paul Pierson. 2005. Off Center: The Republican Revolution and the Erosion of American Democracy. New Haven: Yale University Press. Hardisty, Jean. 1999. Mobilizing Resentment: Conservative Resurgence from the John Birch Society to the Promise Keepers. Boston: Beacon Press. Hicks, Alexander. 1999. Social Democracy and Welfare Capitalism. Ithaca: Cornell University Press. Huber, Evelyne and John D. Stephens. 2000. “Partisan governance, women’s employment, and the social democratic service state.” American Sociological Review 65: 323–42. Huber, Evelyne, Charles Ragin and John D. Stephens. 1993. “Social Democracy, Christian Democracy, Constitutional Stucture, and the Welfare State.” American Journal of Sociology 99 (3): 711–749. Inglehart, Ronald and Pippa Norris. 2004. Sacred and Secular: Religion and Politics Worldwide. Cambridge University Press. Jackson, J.C. 1908. “The Work of the Anti-Saloon League.” Annals of the American Academy of Political and Social Science (November): 12–26. Kahl, Sigrun. 2004. “The Religious Roots of Modern Poverty Policy: Catholic, Lutheran and Calvinist Traditions Compared.” Paper presented at the workshop on Religion and the Western Welfare State, April, 2004. Max Planck Institute for the Study of Societies, Cologne, Germany. Kalyvas, Stathis N. 1998. “From Pulpit to Party: Party Formation and the Christian Democratic Phenomenon.” Comparative Politics 30, 3 (April): 293–312. Keddie, Nikki R. 1998. “The New Religious Politics: Where, When and Why Do Fundamentalisms Appear.” Comparative Studies in Society and History 40 (4): 696–723. Klatch, Rebecca. 1999. A Generation Divided: The New Left, The New Right and the 1960s. Los Angles: University of California Press.

264

Korpi, Walter. 1989. “Power, Politics and State Autonomy in the Development of Social Citizenship: Social Rights during Sickness in Eighteen OECD Countries Since 1930.” American Sociological Review (54): 309–328. Kristol, Irving. 1994. “Taking Religious Conservatives Seriously.” Pp. Vii-xi in Disciples and Democracy: Religious Conservatives and the Future of American Politics, Michael Cromartie, Ed. Washington, D.C.: Ethics and Public Policy Center. Layman, G.C.. 1997. “Religion and Political Behavior in the United States: The Impact of Beliefs, Commitment and Affiliation from 1980 to 1994.” Public Opinion Quarterly 61 (2): 288– 316. Leege, David C. 1992. “Coalitions, Cues, Strategic Politics and the Staying Power of the Religious Right, or Why Political Scientists Ought to Pay Attention to Cultural Politics.” PS: Political Science and Politics (June): 198–204. Lienesch, Michael. 1982. “Right-Wing Religion: Christian Conservativism as a Political Movement.” Political Science Quarterly 97 (3): 403–425. Lipset, Seymour Martin. 1996. American Exceptionalism: A Double-Edged Sword. New York: Norton. Lynch, Julia. 2004. “Italy: A Christian Democratic Welfare State?” Paper presented at the workshop on Religion and the Western Welfare State, April, 2004. Max Planck Institute for the Study of Societies, Cologne, Germany. Manow, Philip. 2004. “The Good, the Bad and the Ugly: Esping-Anderson’s Regime Typology and the Religious Roots of the Western Welfare State.” Paper presented at the workshop on Religion and the Western Welfare State, April, 2004. Max Planck Institute for the Study of Societies, Cologne, Germany. Manza, Jeff and Clem Brooks. 1997. The Religious Factor in U.S. Presidential Elections, 1960– 1992.” American Journal of Sociology 103 (1): 38–81. Maxwell, Carol. 2002. Pro-Life Activists in America: Meaning, Motivation, and Direct Action. New York: Cambridge University Press. McAdam, Doug. 1996. “The Framing Function of Movement Tactics: Strategic Dramaturgy in the American Civil Rights Movement.“ Pp. 338–355 in Comparative Perspectives of Social Movements: Political Opportunities, Mobilizing Structures, and Cultural Framings, edited by D. McAdam, J. McCarthy, and M. Zald. Cambridge: Cambridge University Press. McAdam, Doug. 1999. Political Process and Development of Black Insurgency, 1930–1970. Chicago: University of Chicago Press. Moen, Matthew. 1992. The Transformation of the Christian Right. Tusacaloosa: The University of Alabama Press. Morgan, Kimberly. 2006. Working Mothers and the Welfare State. Stanford University Press. Morone, James. 2003. Hellfire Nation: Morality in American Politics. New Haven: Yale University Press. Morris, Aldon D. 1999. “A Retrospective on the Civil Rights Movement: Political and Intellectual Landmarks.” Annual Review of Sociology 25: 517–539. Morris, Aldon. 1984. The Origins of the Civil Rights Movement. New York: Free Press. Myles, John and Jill Quadagno. 2000. “Envisioning a Third Way: The Welfare State in the Twenty-First Century.” Contemporary Sociology 29 (1): 156–167. CCC. 2002. “Political Theories of the Welfare State.” Social Service Review March: 1–22.

265

Nord, Philip G. 1984. “Three Views of Christian Democracy in Fin de Siecle France.” Journal of Contemporary History 19: 713–727. Oldfield, Duane. 1996. The Right and the Righteous: The Christian Right Confronts the Republican Party. Lanham, Maryland: Rowman and Littlefield. O’Connor, Julia, Ann Orloff and Sheila Shaver. 1999. States, Markets and Families: Gender, Liberalism and Social Policy in Australia, Canada, Great Britain and the United States. Cambridge, UK: Cambridge University Press. Palme, Joakim. 1990. Pension Rights in Welfare Capitalism: The Develoment of Old-Age Pensions in 18 OECD Countries, 1930 to 1985. Stockholm: Swedish Institute for Social Research. Pierson, Paul. 1994. Dismantling the Welfare State: Reagan, Thatcher, and the Politics of Retrenchment. Cambridge: Cambridge University Press, 1994. CCC. 2000. “Increasing Returns: Path Dependence and the Study of Politics.“ American Political Science Review 94, 2: 251–267. Phillips, Kevin. 2006. American Theocracy: The Peril and Politics of Radical Religion, Oil and Borrowed Money in the 21st Century. London, England: Viking. Quadagno, Jill. 1994. The Color of Welfare. New York: Oxford University Press. CCC. 2004. “Why the US Has No National Health Insurance: Stakeholder Mobilization Against the Welfare State, 1945–1996.” Journal of Health and Social Behavior 45: 25–44. CCC. 2005. One Nation, Uninsured: Why the US Has No National Health Insurance. New York: Oxford University Press. Reed, Ralph. 1994. “What Do Religious Conservatives Really Want?” Pp. 1–31 in Disciples and Democracy: Religious Conservatives and the Future of American Politics, Michael Cromartie, Ed. Washington, D.C.: Ethics and Public Policy Center. Reese, Ellen. 2005. Backlash Against Welfare Mothers: Past and Present. University of California Press. Regier. P. 38 Regnerus, Mark, David Sikkink and Christian Smith. 1999. “Voting with the Christian Right: Contextual and Individual Patterns of Electoral Influence.” Social Forces 77 (4): 1375–1401. Reichley, A. James. 1981–82. “The Conservative Roots of the Nixon, Ford, and Reagan Administrations.” Political Science Quarterly 96 (4): 537–550. CCC. 1986. “Religion and the Future of American Politics.” Political Science Quarterly 101 (1): 23–47. Rohlinger, Deana. 2002. “Framing the Abortion Debate: Organizational Resources, Media Strategies, and Movement-Countermovement Dynamics.“ The Sociological Quarterly 43: 479–507. CCC. 2004. “Getting into Mass Media: A Comparative Analysis of Social Movement Organizations and Mass Media Outlets in the Abortion Debate.“ Dissertation Thesis, Sociology, University of California, Irvine. Rohlinger, Deana and David Meyer. 2005. “Framing Abortion Globally: Transnational Framing of Access to Abortion in the United States, England, and Ireland.“ Pp. 197–214 in U.S. Women’s Movement in a Dynamic and Global Perspective, edited by L. A. Banaszak. New York: Rowan and Littlefield Publishers.

266

Rohlinger, Deana and Jill Quadagno. 2006. “Social Security Rights.” In Public Sociologies Judith Blau and Keri Iyall Smith, Editors. Rowman and Littlefield. Rozell, Mark J. and Clyde Wilcox. 1996. “Second Coming: The Strategies of the New Christian Right.” Political Science Quarterly 111 (2): 271–294. Schafly, Phyllis. 1996. “Welfare: Fraud and Failure.” The Phyllis Schafly Report, February, 1996: 1–2. Shibley, Mark A. 1998. “Contemporary Evangelicals: Born-Again and World Affirming.” Annals of the American Academy of Political and Social Science 558 (July): 67–87. Skocpol, Theda. 1996. Boomerang: Health Care Reform and the Turn Against Government. New York: W.W. Norton. Smith, Christian. 1996. Resiting Reagan: The U.S. Central America Peace Movement. Chicago: University of Chicago Press. Sombart, Werner. 1906 (1976). Why Is There No Socialism in America? White Plains, New York. Stevens, Beth. 1988. “Blurring the Boundaries: How the Federal Government Has Influenced Welfare Benefits in the Private Sector.” In The Politics of Social Policy in the United States, edited by Margaret Weir, Ann Shola Orloff, and Theda Skocpol. Princeton, N.J.: Princeton University Press. Szymankski, Ann-Marie. 2003. Pathways to Prohibition: Radicals, Moderates, and Social Movement Outcomes. Durham: Duke University Press. Tiryakian, Edward. 1993. “American Religious Exceptionalism: A Reconsideration.” Annals of the American Academy of Political and Social Science 527: 40–54. Tocqueville, Alexis. 1835. Democracy in America. Tribe, Laurence H. 1990. Abortion: The Clash of Absolutes. New York: W.W. Norton. Usher, Douglas. 2000. “Strategy, Rules and Participation: Issue Activists in Republican National Convention Delegations, 1976–1996.” Political Research Quarterly 53 (December): 887–903. Van Kersbergen, Kees. 1995. Social Capitalism. London: Routledge. CCC. 2004. “Social Capitalism Revisited? Reconsidering the Religious Roots of the Welfare State.”Paper presented at the workshop on Religion and the Western Welfare State, April, 2004. Max Planck Institute for the Study of Societies, Cologne, Germany. Wald, Kenneth D., Dennis Owen and Samuel S. Hill Jr. 1988. “Churches as Political Communities.” American Political Science Review 82 (2): 531–548. Walker, Martin. 1996. “The US Presidential Election, 1996.” International Affairs 72 (4): 657–674. West, Darrell M. and Burdette Loomis. 1999. The Sound of Money: How Political Interests Get What They Want. New York: W.W. Norton. Wilcox, Clyde. 2000. Onward Christian Soldiers: The Religious Right in American Politics. Boulder, CO: Westview Press. Williams, Rhys. 2002. “From “Beloved Community” to “Family Values”: Religious Language,Symbolic Repertoires, and Democratic Culture.” in Social Movements: Identity, Culture, and the State, edited by D. Meyer, N. Whittier, and B. Robnett. New York: Oxford University Press. Yamane, David. 2005. The Catholic Church in State Politics. New York: Rowman and Littlefield. Young, Michael P. 2002. “Confessional Protest: The Religious Birth of U.S. National Social Movements.” American Sociological Review 67 (October): 660–688.

267

10

Religion as a Cultural Force: Social Doctrines and Poor Relief Traditions

Sigrun Kahl

The contributions to this volume set out to develop a comprehensive argument of how religion shaped the Western welfare state, aiming to supersede an older, fragmentary line of reasoning that focused on the effects of religious (particularly Catholic) social doctrine on social policy via the influence of Christian Democratic parties. However, as I show in this chapter, in some sectors of the welfare state there is yet a deeper consensus that underlies, and conditions, the formulas of political compromise between different electoral and societal groups that Manow and Van Kersbergen (this volume) posit to constitute welfare state regimes. Comparing how social doctrines shaped the history of poor relief in Catholic, Lutheran, and Reformed Protestant countries, this chapter looks at religion as a cultural force. This perspective complements the actor and cleavage centered perspective pursued in the other contributions to this volume. Religion can also work its way into policy by creating conditions that render many decisions consensual, so that its influence remains virtually unnoticed. As a result, the impact of religion can be hard to detect. Moreover, because religion is shaped by the institutions it once helped to create, causal direction can be unclear. Still, in this chapter, I show that in the domain of poor relief there exist important institutional continuities that are driven by principles of charity which are firmly embedded in Catholic, Lutheran and Reformed Protestantism’s social doctrines. The influence of religion becomes clear once these principles are taken into account. My argument is twofold. (1) Social doctrines are embedded into the “deep structure” of secular institutions of poor relief and today social assistance, reflecting and reenforcing particular values and attitudes regarding poverty, work, and charity. (2) Social doctrines also condition how religious actors form and articulate their political interests, which can for instance be seen at the churches’ stance towards state involvement. (1) In the first part of the chapter, I show that there has historically been a correlation between social doctrines and institutional traditions of poor relief. I argue that the

268

Reformation led to a differentiation of the general Christian postulate that it is a duty of society to support those members who are in need. This is the most basic moral imperative of the welfare state. It is anything but a clear imperative, though, because it does not say how to help the poor, leaving societies to grapple with a goal conflict between the principle of ‘feeding one’s poor brother’ and that of ‘letting only those eat who work.’ Catholic, Lutheran, and Calvinist/Reformed Protestant societies have picked different moral and institutional solutions to this basic conflict. In essence, the Catholic tradition is closest to the “feed the poor” postulate; the Calvinist tradition lets only those eat who work; and the Lutheran tradition gives the poor bread first and tries to get them to work afterwards. These institutionalized social doctrines may also have affected the timing and organizing principles of social assistance. The principles have been remarkably stable, even beyond the advent of modern social policies. My counterfactual argument is that had the Christian confessional tradition not differentiated into three distinct social doctrines, the institutional trajectories of poor relief and patterns of welfare state evolution in the mature welfare states of the west would have been very different. (2) My second point is more a possible agenda for future research. Looking at the churches’ stance towards state involvement, I hypothesize that social doctrines affect the interest formation of religious actors in the political arena. However, while social doctrines underlie political power struggles and social reforms, patterns of statechurch cleavage also condition how the church forms and pursues its interests. As a political actor, the church adjusts its preferences to the realities of an emerging and expanding welfare state. Why particular state-church balances emerge and which effects they engender can well be explained with the Rokkanian theoretical framework adopted in this volume. What a focus on social doctrines can add to such accounts is to open the black box of the church as a political actor. I argue that the interests of religious actors are not identical between the different denominations, and they change over time as actors adapt to the political realities brought about by the state regulation of social welfare.

269

10.1

Three Traditions of Work, Poverty, and Charity

Ensuring that no citizen lacks the basic means of subsistence is the most basic concern of the welfare state. Historically this postulate is the starting point of public social policy to care for the poor. But what exactly should be done with the poor? Should they be fed, clothed, and healed; i.e., should society ensure their welfare? Or should they be forced to work for their own welfare? This is a very old question that remains painfully unsolved, as evidenced today in the periodically resurfacing debates about poverty, laziness, work and the duty of society to provide versus the duty of the individual to support him- or herself. The bible does not provide Christians with a clear answer. Rather, it reflects that societies have always been in conflict about these two goals. For instance, the bible postulates that “he who is kind shall be blessed, for he gives of his bread to the poor” (Proverbs 22: 9). Yet it also states that “if any will not work, neither let him eat” (II Thessalonians 3: 12;10). The goal conflict between the principle of granting economic support and ensuring that everybody who can work in fact does so is an old one and has been pervasive in the history of poverty relief in Christian societies for two millennia.

The Medieval Sacralization of Poverty and the Reformation The medieval unity of poverty and work somewhat alleviated this goal conflict. “Work” was a fatiguing and painful effort required of the poor and powerless in exchange for their subsistence. To be poor meant to be working. Those who had to work to survive were poor by definition. Thus, while the sloth was not socially desired and should work in exchange for his food, the dominant societal program was not to bring the poor to work. Poverty was associated with powerlessness, manual labor, and social problems. But all this was outweighed by the glorification of the poor as images of Christ. The most telling effect of this view of poverty was a large share of “voluntary poverty” – orders and individuals who gave up all their possessions to be closer to Christ. Prayers from the poor were the most effective way to ensure entrance to heaven in a world oriented towards the afterlife. Pauper and potens thus engaged in a reciprocal and for both sides essential commitment: The potens passed out the alms, and in return the pauper prayed for the donor’s soul. The medieval church’s task,

270

then, was to mediate between the rich and the poor. The rich were to donate to the church and give alms in person. As a result, a sinner could be “saved from hell by throwing a loaf of bread to a beggar” as “the alms bestowed today would be repaid a thousand-fould, when the soul took its dreadful journey amid rending briars and scourching flames” (Tawney 1990 [1922]: 259). Between a third and a quarter of church income went to the poor, who were to accept their destitute situation (Geremek 1991: 52). It was common to donate large sums to be passed out to the beggars at each anniversary of the donor’s death, to ensure that the poor pray for salvation on this day. Municipalities started to regulate begging well before the Reformation, not only as a reaction to the social problem and the public nuisance, but also out of the Christian duty to care for the “truly” poor. For instance, societal attitudes towards giving to the wandering beggars were ambiguous. Beyond fears that they might spread disease and difficulties in verifying who was truly poor, people wanted assurance that the poor would do their part of the gift exchange by praying. For this reason, municipal edicts on begging often required a “beggars’ exam” to ensure that beggars could tell the Lord’s Prayer, the Ave Maria, the Apostle’s Creed and the Ten Commandments (see the edicts in e.g. Waldau 1789; Rüger 1932).

10.1.1 Catholic Poor Relief: Feed the Poor, Get Saved The Catholic tradition continued the medieval principle of caritas. In the course of the Counter-Reformation, Christian benevolence became an important part of Catholic renewal. The Council of Trent confirmed the traditional principles of poor relief and rejected the repression of begging, while it acknowledged that begging needed to be regulated. Because people were justified by faith in Christ and by a life of good works, almsgiving remained an individual act and begging was not forbidden (Battenberg 1991: 68f; Fairchilds 1976: 27; Pullan 1988: 200). Not only Catholic policymakers and church officials, but also the population opposed the secularization of poor relief. The fear was that a secular system would erase the divine benevolence of the giver.

271

When in the 16th, 17th, and 18th centuries secular authorities did try to regulate begging, their declarations and edicts were similar to the Protestant legislation in that they sought to force the able-bodied poor to work and to deport vagrants. Yet, whereas secular authorities in Protestant countries actively enforced such legislation, in Catholic countries institutions of poor relief were not secularized and there was thus little authority to enforce the legislation. When secular authorities attempted to force the poor to work, they often met plain resistance (e.g. Gutton 1971; 1974; Stolberg 2002). In Rome, Genoa, and Turin, for instance, the opposition to the workhouse was so great that the newly established workhouses had to be shut down. In Lucca and Siena, similar plans were not even realized (Gorski 2003: 135). Moreover, legislation in Catholic countries usually went less far than in Protestant societies. For instance, begging was often only forbidden for the wandering beggars but not for the resident poor. Finally, much of the legislation regulating poverty was not systematic but situational, reacting for instance to a strong increase in the number of beggars or to an epidemic. Ensuring Salvation through Almsgiving: The Hospital System Catholic poor relief continued the relatively indiscriminate passing out of alms, stressing that “giving to the poor from one’s affluence is a moral duty, which, however, cannot be called for by the poor as a right” (Fösser (1889: 567). The hospital was the major institutionalization of that principle, and it remained in the hands of the church, monastic orders, lay confraternities, and pious foundations. Through personal alms giving, it guaranteed that the alms reached the recipient, ensuring the donors’ salvation. The hospital at the same time allowed the church and local authorities to carry out social monitoring and to confine and control the poor, to separate men from women, and to teach them religion. In Spain and Italy, the traditional institutions of charity were largely unaffected by the Reformation (Pullan 1988; Geremek 1991). Protestant ideas informed some proposals how to reform poor relief, but these never made it into legislation. France differed from its Catholic neighbors in making some attempt to forbid begging and to make the poor work or to even establish a public poor relief system, like the Aumône Générale in Lyon (Davis 1968). However, these efforts were never consistent and Catholic charity persisted (Metz 1985: 2). There was no coordinated supervisory system but a hodgepodge of local institutions. By the end of the ancien régime, Paris

272

alone had about 50 hospitals for various groups of the poor, complemented by a large number of private institutions (Ramsey 2002: 282). The two most important of these institutions were the Hôpitaux généraux and the Dépôts de mendicité. The latter were royal workhouses that were introduced in the 18th century to force able-bodied beggars to work. While the Dépôts were a clearly punitive policy, there were simply no functioning institutions to enforce it. Catholic hospitals were not to pay off economically, nor were they just places of punishment, but they were places of support, shelter and medical care. The majority of the clientele were elderly people, children, the sick, disabled and mentally ill (Metz 1985: 3, 7; Gorski 2003: 135). Caritas, Subsidiarity and the Lack of Secular Poor Relief In the tradition of Catholic caritas, helping the poor was a responsibility of the local congregation and should arise from compassion rather than legal force by the state. Relatives, friends, employers and the church all felt individually responsible for the poor. Typical examples of Catholic poor relief are Spain and Italy, where relief stayed localized and church and family remained the most important providers of assistance. Until the 20th century, Italy and Spain had no national regulation on minimum benefits. The French case is more complex because many cities introduced centralized and partially secular systems of poor relief. France also saw a violent church–state conflict during and after the Revolution. Despite strong secularization trends, the church remained the predominant institution to execute state poor relief until the 19th century. The national convention wanted a national poor relief system and introduced the principle of obligatory public assistance. The right to subsistence was written into the declaration of rights of the Jacobin Constitution (Ramsey 2002: 288). However, the traditional principles and providers quickly returned. The state seized church property, but did not replace the hospitals. It continued to pay the clergy to do the work. Thus, the institutions of poor relief changed owner but were not themselves changed. Following the fall of Robespierre, the national convention embraced a less etatist model. Already in 1796, the unsold property of the hospitals was returned to the church and the church was to be compensated for the property already sold. Poor relief became a municipal and voluntary matter that was entrusted to renowned citizens (Metz 1985: 10f). The Committee on Poverty increasingly focused on particular

273

groups, like the sick and disabled, and stressed the role of voluntary charity. Poor relief was to rely on private contributions for its finances (Ramsey 2002: 287). French liberals and Catholic conservatives mostly agreed on the key principles of poor relief. Both considered assistance a social necessity and thought that the state could play a useful role in the coordination of various governmental and voluntary efforts, but they refused to introduce a state guarantee. In short, until the last quarter of the 19th century, France retained the core principles of Catholic social thought. In revolutionary France, charity turned into a duty of the good Citizen and the good Christian. By the late 19th century, the Enlightenment terms bienfaisance and philantrophie enjoyed less currency than Christian charité (Ramsey 2002: 302; Kesselman 2002). In the course of the religious revival at the beginning of the 19th century, the old system was strengthened. No less than 300 orders were created in France between 1810 and the Second Empire. During the 1850s no fewer than 100,000 women worked in such orders (Faure 2002: 312). France was thus characterized by the public regulation and religious implementation of assistance (Gouda 1995), a “surprisingly limited direct involvement by the state in poor relief,” and an “explosive growth of charity and mutualism” (Ramsey 2002: 303). The French case is thus somewhat less state-driven than one might have expected. Public poor relief was less well-developed and far less centralized than in other European countries. The system relied primarily on local institutions, private charity and mutual aid. Even when state involvement expanded at the end of the 19th century and established various regulations, transfers and services of assistance sociale for children, the elderly, and the disabled (see Manow/Palier, this volume), a national poor relief system did not emerge.

10.1.2

Lutheranism: Bread First, Work Second

With Luther’s translation of the Bible and the Reformation, the goal conflict between feeding the poor and forcing the poor to work for their own bread became more visible and pressing, because the medieval/Catholic unity of work and poverty no longer applied. The concept of work changed in two respects. First, Luther raised the status of work immensely by construing it as an intrinsically positive activity that was pleasing to God. Second, work no longer equaled poverty but instead was seen

274

as a way to overcome poverty, which, in turn, became associated with non-work and laziness. Salvation did not depend on the kind of work a person was doing. The poor peasant’s work was worth as much as that of the wealthy artisan. The pursuit of material gain beyond individual needs, however, was reprehensible. As a consequence, Lutheran doctrine altered the status of both the beggar and the almsgiver considerably. Excoriating the sale of indulgences by the Catholic church, Luther called begging “blackmail” and postulated that Christian truth could be found only in Scripture (sola scriptura), and that only by faith could man be justified (sola fide). God’s gratia amissibilis could always be regained by true faith. All human works were sins, as long as the person performing them was a sinner. He thus strongly rejected the idea that generous donations could prevent sinners from eternal damnation and agony in fire and brimstone, or that the poor would be justified by living in poverty. Luther rejected individual almsgiving and denounced able-bodied beggars. In his foreword to the 1523 German edition of the Liber Vagatorum, a famous collection of fraudulent begging techniques, he demanded that the “undeserving” poor – the cheaters, idlers and vagrants – be excluded from the alms. The differences between Catholic and Lutheran approaches were reflected in the heated controversies theologians fought from the 16th century onwards about Catholic almsgiving, how to deal with beggars, the role of the Reformation, and whose system was more in line with the Gospel (Geremek 1991: 91,147; Davis 1968: 217ff). These debates were revived again and again in the centuries following the Reformation. In 19th century Germany, Lutheran historians criticized the Catholic church for not countering “undeserving” poverty, most of all the “strong beggars.” They argued that the Catholics failed to develop criteria for the proper distribution of the alms among the poor, leaving troops of wandering beggars to grow into a “beggars plague” (e.g. Uhlhorn/Aschrott 1859: 828). That is why Protestant poor relief was argued to have been mere “self-defense” by the police against the dangerous nuisance of begging (Uhlhorn/Aschrott 1859: 919). Building on that critique, Lutheran social reformers developed their own version of poverty policy. The state was to protect the community from the “undeserving” poor, but the state also was to ensure all the help possible for the “truly needy” (e.g. Aschrott/Uhlhorn/Münsterberg 1909). The Catholic side in turn chastised the Protestant approach, criticizing that “the feelings of voluntarily giving medieval Catholics were different from the feelings of to-

275

day’s Protestants who must be forced to pay their taxes” (my translation of Fösser 1889: 490). From Individual Almsgiving to Centralized Outdoor Relief One of the most obvious signs of diverging developments in Catholic and Protestant countries and territories was the change in municipal edicts on begging. Until the early 17th century, almost all Lutheran cities transformed reactive medieval begging edicts that had negatively regulated begging into active poor laws that formulated a positive responsibility of the emerging secular authorities to care for the poor. As first outlined in the Order for a Common Purse (Beutelordnung) for Wittenberg in 1520/21, the poor were to be registered and to be supported out of a common chest that was financed through weekly collections. This system reduced the role of the hospitals to the care for the sick and the weak. Poor relief rigorously enforced the distinction between the “deserving” and the “undeserving” and relief tended to be restricted to the residential, authentic and morally upright poor, whereas the able-bodied were to work. In Luther’s famous words, a basic principle of a “healthy” system of poor relief is that “no one should live idle on the work of others.” As an example, the Nuremberg alms edict of 1522 mandated that able bodied beggars not be supported so that the deserving poor could get all they needed. Applying for relief became a bureaucratic process requiring a formal examination of need and eligibility. Likewise, in 1522 Christian II of Denmark required that the worthy poor be separated from the unworthy (Brinker 1994: 33). The Danish Poor Law of 1708 (Fattigforordning) stipulated that the poor be classified into three classes – those who were unable to sustain themselves at all, orphans, and those who were able to work. Indoor relief was to be given to the sick and the weak (Bonderup 2002: 173–176). As a complement to outdoor relief, Lutheran cities adopted the workhouse from Calvinist cities. The workhouse was to deter the able-bodied poor from claiming relief, and to remind them demonstratively of their duty to support themselves. At the same time, it was an institution of punishment and correction for those who were socially deviant, often also serving as an ordinary jail. Copenhagen erected a workhouse in 1605 to instill upon beggars and vagrants both the ability and the will to

276

work (Brinker 1994: 54f). However, nowhere was the workhouse so widespread as in England, the Netherlands, and the United States. In Germany, territorial differences in poor relief practices were still significant as late as the end of the 19th century. The share of the population in Lutheran German cities that received public poor relief was much higher than in the Catholic and mixeddenominational cities. In 1892, in Lutheran Berlin, 6.7% of the population received outdoor relief. In the Lutheran cities of Frankfurt am Main and Stuttgart, the percentage was similar, 6% and 6.5% respectively. In contrast, in Catholic Cologne 3.8% received outdoor relief, and in Catholic Munich, the share was only 1.2% (Levine 1988: 45). The State as Uncontested Provider Lutheran poor relief was centralized and laicized in cooperation with the church. The new secularized system was built with the existing religious institutions. This cooperation, for instance, helped introduce the 1522 Nuremberg Poor Law. This law replaced the alms and introduced a municipal common chest for the poor. It was, however, read from the pulpit, and the chest was set up inside the church. In the Lutheran countries public responsibility for poor relief was unquestioned and the secularization of social welfare was a smooth process. The common chest later developed into poor taxes, to be collected and delivered by lay administrators. In the course of the Danish Reformation, poor relief became a direct responsibility of the Danish state. A 1539 Danish church law ordered that those resources that had thus far been devoted to poor relief should now be paid to the common chest, which was to be administered by the municipality (Brinker 1994: 39ff). Likewise, in Sweden, cities introduced poor rates (for instance, Stockholm 1663) to be collected by officially appointed collectors (Grell/Cunningham 1997: 23–26). This process was smooth but slow. In Sweden and Denmark, the kings sought to regulate nationally local practice without establishing a national system. They sought to deal with poverty by ordering municipalities to look after their own poor, but made contributions to the hospitals and outdoor relief. This changed in the 17th century. For instance, in 1630 poor relief was re-organized in Copenhagen, begging was forbidden, and the deserving poor were to receive quarterly alms from the overseers (Riis 1997). In Sweden, Gustavus Adolphus in 1624 tried to outlaw begging to reduce

277

the number of hospitals and to establish workhouses to be financed by an annual poor tax. He did not succeed and in 1642 the government re-introduced parish relief (Grell/Cunningham 1997: 23–26). Beginning in the late 1520s, the process of reform moved from the local to the regional level in Germany. By the mid-sixteenth century, the three leading states of the Reformation (Saxony, Hessia, and Württemberg) had already implemented statewide reforms. By the late 16th century, Lutheran princes all over Germany were asserting control over poor relief as part of their general police powers (Gorski 2003: 128–129). By 1794, poor relief had become a general responsibility of the Prussian state (Allgemeines Landrecht). With this law, the state committed itself “to provide for the nutrition and feeding of those citizens who are unable to provide for themselves, and who are unable to receive provisions from others who are bound to provide care in accordance with other special laws” (quoted from Dross 2002: 72). In the 1808 Städteordnung, Prussian municipalities were compelled to establish communal authorities on a uniform basis. As a result, relief efforts increased dramatically. For instance, while in 1750, one in 82 Berlin residents got poor relief, in 1801 the ratio was 1 to 14 (Dross 2002: 74).

10.1.3

Reformed Protestantism: Work for Your Own Bread

Calvinism developed two different, and partially contradicting, ideas of poverty, which both stigmatized the poor. One is the doctrine of predestination in Calvin’s writings; the other is the ethos of work and individual responsibility. Predestination states that God’s “unconditional election” creates every human being as either damned or saved prior to birth. The condemnation of the poor did not necessarily follow from this, and it was not explicit in Calvin’s writings. Calvinists, however, were searching for signs of damnation or salvation. The principle of understanding a morally rigorous worldly life and economic success as signs of election marks the last step in the emergence of a Calvinist moral in 17th century England, Holland and United States (Borkenau 1980 [1934]: 154–161). In practice this means that God helps those who help themselves. Thus the Calvinist, as it is sometimes put, himself creates his own salvation, or, as would be more correct, the conviction of it. But this creation cannot, as in Catholicism, consist in a gradual accumu-

278

lation of individual good works to one’s credit, but rather in a systematic self-control which at every moment stands before the inexorable alternative, chosen or damned. (Weber [1904–05] 1958 [1904/05]: 115)

The most certain mark of election was proving one’s faith in a worldly activity, and success in a worldly occupation and wealth became a possible indication that one was saved by God from the start, while poverty became a possible indicator of damnation. The Calvinist creation of the Protestant work ethos and the strict and systematic requirements about what constitutes a life that increases the glory of God (e.g. personal responsibility, individualism, discipline, and asceticism) made poverty appear to be the punishment for laziness and sinful behavior and could be taken as a sign of damnation. Along these lines, good works were a necessary but not a sufficient sign of being chosen. “There was no place for the very Catholic cycle of sin, repentance, atonement, release, followed by renewed sin” (Weber [1904–05] 1958 [1904/05]: 117). The Calvinist solution to the goal conflict between feeding one’s poor brother and forcing him to work for his food was to stress the latter objective. Calvin took Luther’s interpretation of work much further. The exalted position of work, if ambiguous for Luther, was clear for Calvin: “In the things of this life, the labourer is most like to God”, he declared (Tawney 1990 [1922]: 23). He made work an absolute duty, a spiritual end in itself and the best way to please the Lord. Calvin also fundamentally changed the requirements on how people should work. Whereas in Lutheranism, the sinner could always regain God’s mercy if he was humble and believing, in Calvinism sins were black marks never to be erased. Only systematic and constant selfcontrol provided a feeling of security about one’s state of grace, in particular working without rest in a disciplined and rational manner (Weber [1904–05] 1958 [1904/05]: 70–88). Ultimately, of course, predestination rendered that sense of security illusive. As Weber pointed out, there is no word like Beruf/calling/vocation among the predominantly Catholic peoples for whom the medieval concept of work more or less persisted (Weber [1904–05] 1958 [1904/05]: 79). In the Catholic societies, work remained a means to ensure subsistence and the traditional inefficiency and casualness remained characteristic of Catholicism (Weber 1996 [1904/05]: 22f). This had a discernable impact on European societies: “by the middle of the seventeenth century the contrast between the social conservatism of Catholic Europe and the strenuous enterprise of Calvinist communities had become a commonplace” (Tawney 1958: 6).

279

Both predestination and its marks – the ethics of worldly life – have in common that they could be taken as an indication that the poor are damned and the rich are saved. Also here, the rich could (only God knew) still be damned, but their richness could be taken as a sign of grace. Predestination implied that the community had no positive responsibility for the poor; Calvinist moralism implied that the poor needed to be punished and corrected. The “main problem was to transform the mental outlook of the lower orders so that they no longer waited at the rich man’s gate for charity, but went out to offer their services on the labour market” (Hill 1952: 36). Work Ethos and Workhouse System The intent to bring the poor to work existed everywhere, but only in Reformed Protestant poor relief was the workhouse programmatic, because of the “elective affinity” between the workhouse and the Calvinist view of poverty. The dominance of the workhouse is reflected both in its early introduction and in the high number of workhouses (that were often redefined hospitals). The workhouse was the invention of Calvinist social reformers, with the first workhouse, the London Bridewell (1555), inspired by the Dutch Reformed Protestant communities in London. The first Continental workhouse was the Amsterdam Tuchthuis (1595), which was a model to Northern German cities that followed suit in the early 17th century. The first German workhouse was in Bremen (1609) – where, from the 1540s on, the Reformation had assumed a distinctively Calvinist character. Catholic workhouses were founded very late and remained very few and met considerable opposition, so that many of them had soon to be shut down again. In England there were 200 workhouses in the 18th century, whereas there were just 63 workhouses in the German Lutheran territories (which includes cities and territories where Reformed Protestantism was influential) even though they had a considerably larger population. The Catholic regions of Germany had only 5 workhouses by the 18th century (Köln, Münster, Paderborn, Würzburg, Passau) (Geremek 1991: 260; also Ayaß 1992: 16–21, 25–31; Gorski 2003: 135). Settlers brought to the New World these Calvinist conceptions of poverty and work (Feagin 1975). The colonial population was overwhelmingly British and Reformed Protestant (Puritan, Dutch Reformed, Presbyterian, Anglican). The history of poor relief in the United States thus largely followed the English and Dutch precedents (Lacey 1953). The Dutch established the first workhouses in the 1650s in present New

280

York City and in Albany, to be followed by the English colonies (Huey 2001: 140). London’s Bridewell inspired the founding of a Bridewell in New York City in 1776 (Baugher 2001). All in all, there were about 2,300 workhouses/poorhouses in the United States, some of which operated well into the second half of the 20th century (Wagner 2005: 14). The able-bodied poor in these facilities were generally expected to work, and “inmates” (this is what they were actually called) were under constant supervision and could not leave without permission (Wagner 2005). In the Catholic countries, Lutheran outdoor relief was rejected because it made the giver-receiver relationship anonymous and did not guarantee that the needy actually received the alms. Reformed Protestant reasoning also opposed it, but for a very different reason, namely that it pauperized individuals. Outdoor relief provided no incentive to the poor to develop work habits and improve themselves, and it deprived authorities of any possibility to control their behavior and circumstances. Since poverty was considered to be associated with sin and poor relief was considered to cause laziness, true generosity required that the poor be set to work. The poorhouse was so popular because, as Superintendent Trottier of the Rockingham County Farm in the U.S. said, “indoor relief” ensured that only the deserving poor were supported and those “too lazy to care for themselves” were not admitted so as to “in no way encourage indolence and pauperism” (quoted from Wagner 2005: 39). In Catholic systems, the community had to support the poor, but it was a charity and not an entitlement. In Lutheran systems, the poor had a right to relief (but were denied citizenship). In England and the United States, the poor had only the option to enter the workhouse. In England, a 1572 act, the 1723 Work House Test Act, and the 1834 Poor Law Report allowed support for “paupers” only in the workhouse. Municipal outdoor relief was never a stable institution, was never endorsed as a societal responsibility, and was heavily attacked, and sometimes completely abolished and then occasionally reintroduced. For instance, following the Yates and Quincey reports, New York state and Massachusetts completely replaced outdoor relief with poorhouses in the 1820s. The reasoning was that the present poor laws “tend to encourage the sturdy beggar and profligate vagrant to become pensioners upon public funds.” The poor law provisions “operate as so many invitations to become beggars” and were said to weaken the “desire of honest independence” (Yates report, quoted in: Wagner 2005: 9). Or, as of 1820, Scottish poor relief did not support anybody who

281

was not disabled (Mitchinson 2002: 248f). This reflected the impossibility of cleanly separating the deserving and undeserving, a belief that the workhouse was the proper institution to care for the able-bodied poor, and an acknowledgement that outdoor relief was appropriate only for the “deserving” poor (Katz 1996 [1986]: 3–59; Driver 1993; Wagner 2005). Reality, however, differed somewhat from statute, both in England and the United States. Outdoor relief was, in fact, always important for those who could not work, because it was much cheaper to administer for the parishes than the poorhouse and workhouse. In addition, there were never enough workhouses to accommodate all the able-bodied. Indeed, it was often the deserving poor who entered the poorhouse/workhouse. The original intention of the poor law, that the different classes of the poor (sick, elderly, able bodied, children) be housed in separate buildings, was abandoned and the “general mixed workhouse,” as it was then called, was more common. As a result of these developments, the “workhouse” increasingly turned into a place for the sick, elderly and infirm. The able-bodied poor only entered in times of economic depression (Crowther 2002: 213; Mitchinson 2002). Perhaps it was an unintended consequence, but the principle of indoor relief fitted very well with Calvinist reformers’ ideas by de facto excluding able-bodied men from relief. Calvinist doctrine postulates the glorification of God not by prayer only, but by striving and laboring – “labore est orare.” Only in the workhouse could the duty of industry be enforced and the danger of relaxing the incentive to work be avoided. Relief was to be so low and conditions in the workhouse so hard that any work was more desirable and only the most destitute would ask for relief (less eligibility principle). Public assistance was to be restricted to the absolute minimum to keep wages low, a principle which Young’s famous 1771 quote summarized as follows: “Every one but an idiot knows that the lower classes must be kept poor, or they will never be industrious” (quoted from Englander 1998: 1). There are numerous 17th and 18th century pamphlets that advance schemes for further developing the workhouse, often justifying this with the famous Thessalonians quote from the bible. Hartlib, for instance wrote in 1650 that “the law of God saith, ‘he that will not work, let him not eat’. This would be a sore scourge and smart whip for idle persons if … none should be suffered to eat till they had wrought for it” (quoted from Tawney 1990 [1922]: 262).

282

In a way, the Calvinist approach created two classes of work – work as a calling for the elected; and work as punishment and toil for the poor. Whereas Luther had said that any work is of equal value for God, Calvinism and Reformed Protestantism in general held that only rational work (e.g. striving for profits) was pleasing to God and could help to find a sign of grace. Whereas in Lutheranism the state of grace was only determined by faith, Calvinist predestination implied that it did not really matter what one did during life. Therefore the signs of grace assumed such an important place in solving the resulting salvation panic of the Calvinist believer. Thus, the state of grace was predetermined but it was reflected somehow in signs like wealth (elected) and poverty (damned). Calvinism required rational and profitable work from the electi and considered work in the workhouse to be the proper punishment for the poor.

Rejection of State Involvement According to Calvin, poor relief should be part of the church’s ministry. In addition, Calvinism’s emphasis on independence and localism implied that poor relief should be funded by voluntary donations and not government funds. The anti-statist position of Reformed Protestantism resembled Catholicism. Both viewed poor relief as a private and a church responsibility, but the defining principles behind this postulate differed. In late 19th century terminology, Catholics stressed subsidiarity, and Calvinists stressed sovereignty in one’s own circles (Cox 1993: 66). The Reformed poor relief systems were “less centralized and definitely less laicized than their Lutheran analogues” (Gorski 2003: 130). Church and private charities retained a key role in the administration of poor relief and private charity remained part of proving and displaying election. In this sense, Calvinism kept the traditional ostentation of public giving. Poor relief officials were usually lay officials of the church, whereas in the Lutheran countries they were state representatives. This is due to Calvin’s conviction that poor relief was a part of the church’s ministry and to the diaconal system which flowed from that conviction (Gorski 2003: 130). The Dutch poor law of 1854, for instance, ruled that poor relief was the responsibility of the churches. Only if the churches did not have sufficient funds was the municipal poor administration to step in (Gijswijt-Hofstra 2002: 269).

283

Whereas in the Netherlands only few city councils controlled relief administration and poor relief was largely in the hands of various private and church institutions, England developed a more secularized and centralized system with national poor relief legislation and local poor rates. As early as 1572 municipalities were required to introduce a compulsory poor rate, and by the 17th century, poor rates were successfully introduced. However, the English state church was very different from the Lutheran state church in that its doctrine can (roughly speaking) be characterized as a mix of Catholicism and Puritan Calvinism. In addition, the Anglican church developed later than the Lutheran churches and was soon challenged and partially dislodged by non-conformist sects and churches also rooted in Calvinism (Grell/Cunningham 1997). Thus, as Heidenheimer (1983: 8) noted, Anglo-Saxon Protestants had “Puritan Calvinist or sectarian consciences.” This helps explain the half-way equilibrium between state control and local (state) church influence in England. The English system also remained localized, with church and private charities retaining considerable influence over the collection and distribution of funds. Moreover, poor relief legislation by the state tended to regulate poor relief negatively (e.g. by prohibiting outdoor relief and restricting poor relief to the poorhouses and the workhouses) and to provide a framework for local action (e.g. blueprints for the classification and institutionalization of the poor). As a result, the British Poor Law was “a tool of social policy of infinite variety and unlimited versatility” (Fraser 1976: 32) that gave space to experiments like the notorious Speenhamland system of 1795 in Berkshire (Polanyi 1957 [1944]). Moreover, the content of poor relief was not affected by the early introduction (at least on paper) of poor taxes. Contrary to the Lutheran countries, the poor laws did not positively formulate an ultimate state responsibility for the poor. Private Christian charity and voluntary organizations remained central actors (Lewis 1999: 13). In the United States, the (national) state played only a small role, stepping in only rarely and with strictly limited powers (Levine 1988: 265).

10.2

Timing and Principles of Modern Social Assistance Programs

It is commonly held that there was a “pattern of development” from outdated, insufficient poor laws toward “comprehensive, universal programs” (Quadagno 1988: 2),

284

in which the welfare state constitutes a break with poor law principles, which are prior, but not antecedent, to the welfare state (see also Flora 1986: XV; Levine 1988; Ritter 1991; Geremek 1991; Rimlinger 1971). This view is simplistic at best. Rather, poor relief was historically foundational for the welfare state, because it was the first realm of public redistribution towards the needy and constituted a central part of early modern state capacity. The design of poor relief also conditioned to which extent state involvement in social assistance was legitimate and institutionally feasible. The extent to which the state was already involved in poor relief co-determined the scope of wider welfare state activity. Social insurance was historically built on top of an already existing poor relief system. The latter system continued and developed into a social support of last resort for those who were not covered by social insurance. Those countries that by the end of the 19th century had centralized, tax-financed poor relief systems that were administered by state appointed officials were also the first to introduce social insurance. They did so basically in order to supersede the existing system of poor relief, which was deemed insufficient and unable to deal with the extent and variation of needs and risks brought about by industrialization. These countries were Lutheran. There is a strong correlation between dominant denomination and the timing of mainstream welfare state benefits. Lutheran countries started early with the introduction of social insurance, while those with a Catholic and Reformed Protestant heritage began late (Manow 2004). The timing of a national last resort safety net (social assistance) followed a different, complementary pattern. The Reformed Protestant states introduced social assistance before Lutheran states; Catholic states launched it very late or not at all. As to benefit structure and generosity, Catholic and Calvinist social assistance is fragmented and ungenerous, with different benefits for different groups of the poor. Lutheran social assistance is unitary and generous, with one uniform social assistance program (Kahl 2005). Because of the supreme role of the church as a provider of relief in the Catholic countries, the welfare state started late and did not cover the neediest. A last resort safety net was introduced late or not at all. Issues of social assistance remain a core domain of local and church charity in the Catholic world and this world is characterized by “rudimentary” welfare states that lack a provider of last resort. If public assistance

285

programs exist, they are ungenerous. Indeed, Italy, Spain and Greece are still without national social assistance even today. French social assistance was similar to Italy’s until 1990, because it did not have a national benefit and provided only local relief, albeit neither everywhere nor for everybody in need (Commission Nationale d’Evaluation du Revenu Minimum d’Insertion 1992: 101–105; Neyret/Madinier/al. 1988). In 1990, however, the Revenu Minimum d’Insertion (RMI) was successfully introduced. Because the Lutheran state churches supported secular social welfare, these states could introduce social protection earlier and more easily. Germany and the Scandinavian countries launched such programs already in the late 19th century and they adopted the social insurance principle to cover major social risks. In the course of social insurance expansion, poor relief became residual, though it continued to exist as a last resort supplement, covering only the “left over” exigencies that fell through the cracks of social insurance. Thus, national assistance was introduced at a very late stage in the process of welfare state building and, like poor relief, social assistance was a residual program. Important differences between Germany and Scandinavia relate to the influence of a strong Catholic minority in Germany, which can for instance be seen at the means test, which requires the immediate family to provide support before the state steps in, and in the importance of the third sector in the delivery of social services. Because Reformed Protestantism viewed state involvement in social welfare as undesirable, the welfare state neither started as early nor did it become as comprehensive as in the countries with a Lutheran tradition. These countries did not introduce a national social insurance system. Reformed Protestant countries were first to adopt national social assistance schemes, to provide a modest level of relief to most urgent cases of need where self-help failed, without restricting private insurance and nonstatist welfare provision. The United States still today lacks a comprehensive social assistance network, a national health care system and comprehensive insurance against unemployment. Indeed, there is a trend in the U.S. towards even less state responsibility. Many policymakers consider the church and charities to be better able than the state to design and deliver such policies. With the “charitable choice” provision in the 1996 welfare reform, the separation between church and state has been effectively abandoned.

286

The three historical traditions of poor relief correlate with the recent introduction of “welfare-to-work” policies into social assistance programs. The principles and programs that comprise national welfare-to-work strategies differ substantially across countries because policymakers combine those elements that fit in with the institutional legacy. For instance, the United Kingdom and the United States opted for upfront job search, strong obligations of the unemployed, and “making work pay” measures. Denmark, Sweden, and Germany take a longer route of enhancing employability. France and Italy never adopted the idea that the poor have work obligations and should not get benefits if they do not comply (Kahl 2006).

10.3

Social Doctrines and Church Interests within the Welfare State

Differences in the doctrinal conceptions of state involvement affected patterns in the evolution of state involvement in poor relief and social assistance, but the degree of state involvement in these three traditions was certainly not only a result of doctrinal stances over state responsibility. In line with the general thrust of the argument in this volume, the state–church cleavage considerably influenced patterns of state responsibility for the poor (e.g. Fix/Fix 2002; Schmid 1995). A focus on social doctrines provides a better understanding of the church as a political actor and of other religious actors. Their political interests are neither an objective given, nor are they carved in stone. For example, social doctrines influence how the churches conceptualize their political interests and how they adjust their interests as the state expands its reign over efforts in aid of the poor. There are two scenarios in which the church re-articulates its view of state involvement. (1) The church reacts to a welfare state that expands despite church opposition. In this case, the church seeks to maximize its interests within, and to utilize, the state. (2) The church calls for more state involvement in a situation where traditional church institutions fail, for instance because they are unable to cope with the extent and character of need. How the church reacts to the emerging state and its attempts to take over poor relief initially depends on how church doctrine views worldly involvement. That is why the evolution of poor relief was a relatively smooth process in the Lutheran countries.

287

Luther’s doctrine of two estates precluded conflict because the religious estate was responsible for the soul and the governing estate for order. Since the 16th century, the church has been an integrated part of the state in Scandinavia. Though there were important variations in the degree of state-church integration, and in Calvinist, Roman Catholic, and Pietistic and revivalist influences (see Thorkildsen 1997), the overarching feature in all of these countries was a homogenous state Lutheranism (Østergård 1997). Countries under Lutheran dominance secularized church property in the course of the Reformation and assigned a positive role to the state very early on, and the church withdrew from poor relief even as it implemented the poor law for the state. In accordance with Lutheran poor law, these countries established tax-based and centralized systems of poor relief. Poor relief officials were laymen and employed by secular authorities. This also reinforced the endorsement of state responsibility in Lutheran social teachings. The Catholic and Calvinist reaction to a nascent and expanding welfare state, on the other hand, was opposition, even in economic crises, because Catholic subsidiarity and Reformed Protestant individualism and voluntarism both attribute a negative role to the state. For instance, in the middle of the New Deal debates in the United States, Bishop Aloisius Muench argued that “the poor belong to us” and “we will not let them be taken from us” (quoted in Brown/McKeown 1998: 1). (This resistance was also due to the minority status of U.S. Catholics who sought to utilize their charity activities to establish a public voice Brown/McKeown 1998)). Likewise, the Reformed Protestant ideology of mutual and self-help opposed state involvement in an opposition that delayed the implementation of the welfare state (Manow 2004). In countries under Catholic or Calvinist dominance, poor relief was not secularized as early and as comprehensively as in the Lutheran countries and private charity, families and mutual help remained important sources of support. Poor relief officials were mainly representatives of the clergy. Only if there was a strong state-church conflict in which the state prevailed (France) could poor relief be secularized – but never as early and to the same extent as in the Lutheran countries. But the fact that Catholics and Reformed Protestants were wary of state involvement is just one part of the story. The other part is how their opposition played out in different periods and countries and how they shaped welfare state provision. The existing literature has shown that key actors in the welfare state, such as employers and

288

unions, seek to remain in a key position so that they can shape policy even after welfare programs are cemented among the litany of unquestioned state responsibilities. And so it is with religious actors, particularly when it comes to questions of state activity. The churches’ position towards state involvement is not static. Rather, as Therborn (1994: 106f) has noted, it adapts to new realities. “Once they were modernized to cope with the challenges of the urban and industrial world, the churches have tended to support social concern.” Opposition could transform into support when private charity and self-help failed miserably, and/or when Catholics or Reformed Protestants reasoned that they could better achieve their ends by utilizing the state and thus sought to initiate and influence state policy. The state can thus be an instrument to realize the desired social order. Reformed Protestant actors in particular had a “purely utilitarian idea of the state” (quoted in Heidenheimer 1983: 10). For instance, most of the leading figures in English poor law reforms were Puritans (Grell/Cunningham 1997: 31); or, at least 8 out of a group of 15 “founders of the welfare state” in Britain had a Reformed Protestant background (Barker 1984. A coalition of non-conformists and Liberals made possible the victory of the Liberals in 1906, which paved the way for Lloyd George’s – who was himself non-conformist – reforms (Koss, 1975: 848) In situations of crisis characterized by myriad failures of the institutions of self-help, even Reformed Protestants called for state involvement, but they also stipulated the principles of that involvement, stressing, for instance, individual fault and individual responsibility. Examples are the depression in the late 19th century in the U.S., the failure of the friendly societies in the UK, and the Great Depression. Roosevelt would not have called for national insurance had there not first been a total breakdown of the institutions of first resort as originally postulated by Reformed Protestant doctrine – the market, self-help, and private charity. Another example is the family policies implemented by an incipient French welfare state. The majority of these policies where initiated by representatives of French social Catholicism as “part of a broader attempt to make capitalism conform to Christian teachings” (which ultimately caused cleavages between Catholic social doctrine and the interests of employers) (Pedersen 1993: 227, 231; see also Schultheis 1988). Once social Catholics realized that they lacked the capacity to control family allow-

289

ances in the 1920s, their opposition to state involvement waned and they successfully shaped state provision according to the Catholic concept of the dignity of labor and the just wage (Pedersen 1993: 233, 372f). This is also why the Catholic church has developed a more activist conception of state involvement than the churches rooted in Calvinism still today are more hesitant to support state involvement. When unable to cope with the scope of poverty, the church and other (Christian) associations became strong supporters for social assistance in Italy and France in the 1980s and 1990s. The Italian Catholic church and Christian Democrats opposed the Reddito Minimo di Inserimento when it was first introduced on an experimental basis and this is the major reason why Italy still has no social assistance despite efforts to the contrary. Today, the church wants a national measure because it is unable to provide for the poor, but it insists on a very strong voice in designing that measure. Even in France, secular charity organizations and the church not only initiated the debate that lead to the introduction of the Revenu Minimum d’Insertion (RMI) in 1989, they also kept pushing for reforms, like the 1998 Loi de lutte contre les exclusions. Moreover, the RMI did not replace the existing third sector structures but incorporated them. The church and religious organizations remain present, both at the national level (in the evaluation committees, and as a powerful lobby for changes in the legislation) and at the local level (in the insertion committees and as providers). How the historical state-church balance affects the objectives of the church today can also be seen at the different ways the Catholic church’s welfare organization, Caritas, is organized. In France, Caritas (Secours Catholique) is an advocate for the poor in the political arena, but it does not want to provide social services. In Germany, Caritas is politically influential and provides extensive services, which the state pays for. In Italy, Caritas operates according to the traditional Catholic logic, filling all the gaps in state provision, and Caritas does not want the state to provide, but to finance their services provision.

10.4

Conclusion

In this chapter, I investigated the historical evolution of different denominational solutions to the goal conflict between granting economic support to the poor and

290

ensuring that everybody who can work in fact does. The Catholic tradition systematized traditional caritas, which embraced a variety of needs, thus showing a relatively high level of social tolerance, even though it precluded a right to relief. Countries with a Catholic legacy have stressed “welfare” in the sense of delivering money and services to the poorest of the poor and generally lack a work objective. Lutheran outdoor relief institutionalized a societal responsibility for supporting the poor that was guaranteed through formalized eligibility determination. Countries with a Lutheran legacy have sought to both provide generous welfare benefits and to require the ablebodied poor to work. The Reformed Protestant system did not institutionalize responsibility for but of the poor, enforcing work discipline and providing only meager relief. Countries with a Reformed Protestant legacy unequivocally support and enforce “work first” as the best guarantor of economic and social inclusion. The relationship between confessional divisions and differences in poor relief policies is systematic, despite the considerable variety and overlap between the different traditions, and despite the fact that there is no one-to-one correspondence between doctrines and policies. But how significant is this relationship in the face of a commonly perceived secularization trend? Why should religion continue to influence social policy? The argument presented here implies that, as Christian societies secularized, part of these religious doctrines became associated with secular institutions and value sets rather than confessional beliefs. In the Protestant Ethic, Weber described this transformation in terms of several stages over which “asceticism was carried out of monastic cells into everyday life, and began to dominate worldly morality.” Therefore, “the capitalistic system (…) no longer needs the support of any religious forces” (Weber [1904–05] 1958 [1904/05]: 181, 72). Likewise, religion has worked its way into the welfare state via a process of institutionalization and secularization, thus shaping the historical antecedents, defining the playing grounds for actors, and producing societal values even as it ceases explicitly to be identified as religion. As a result, there is today a “hidden” religious curriculum that is deeply engrained into the secular institution of social assistance. Catholic, Lutheran, and Reformed Protestant poor relief reflected particular values regarding work, poverty, and charity. As I demonstrated in the first part of this chapter, originally, these values were powerful because they were connected to ideas of

291

salvation. For example, the medieval Catholic gave to the poor because almsgiving was considered a ticket to heaven. Once institutionalized in poor relief systems, the mechanism of influence became less instrumental. The (secular) institutions of poor relief unfolded a socialization potential on its own that re-enforced those values in society. Thus, the religious influence may last even as the churches no longer hold the power to define how to address the poor. For example, the workhouse reproduced the individualistic work ethic, Catholic almsgiving the idea that poverty is God’s ordeal rather than a result of individual failure, and Lutheran outdoor relief and poor taxes stressed society’s duty to provide coupled with a more social version of the work postulate (it is both an individual responsibility and a duty towards society to work). These institutionalized values still today structure societies’ responses to need and integration strategies for the unemployed poor. The longue durée, institutional perspective suggested here thus reveals important continuities and complements the cleavage/coalition-centered perspective presented in the preceding chapters of this volume. Confining the analysis to how social doctrine influences policy would generate wrong conclusions, though. As I argued in the second part of the paper, power constellations between the state and the church need to be taken into account in order to explain variations in the degrees of state involvement in poor relief as well as differences in the church’s political preferences. For example, the French case cannot be understood without giving the fierce state-church conflicts in which the state prevailed their historical due. This is why France developed a more comprehensive public poor relief system than Italy or Spain. Confining the analysis to organized religious actors and state-church conflict would not capture the whole story either because religion may also become embedded into the welfare state via processes of institutionalization and cultural transposition. Ultimately, an approach that incorporates both religious actors and the effects of social doctrines on institutional and value traditions captures best how religion shaped the welfare state.

292

References Aschrott/Uhlhorn/Münsterberg, 1909: Artikel Armenwesen Handwörterbuch der Staatswissenschaften, 3. Aufl., Bd. 2. 1–196. Ayaß, Wolfgang, 1992: Das Arbeitshaus Breitenau. Bettler, Landstreicher, Prostitutierte, Zuhälter und Fürsorgeempfänger in der Korrektions- und Landarmenanstalt Breitenau (1874–1949). Kassel: Gesamthochschule Kassel/Verein für hessische Geschichte und Landeskunde e.V. Barker, Paul, 1984: Founders of the welfare state London: Heinemann. Battenberg, J. Friedrich, 1991: Obrigkeitliche Sozialpolitik und Gesetzgebung. Einige Gedanken zu mittelrheinischen Bettel- und Almosenordnungen des 16. Jahrhunderts. In: Zeitschrift für historische Forschung 18, 33–70. Baugher, Sherene, 2001: Visible Charity: The Archeology, Material Culture, and Landscape Design of New York City’s Municipal Almshouse Complex, 1736–1797. In: International Journal of Historical Archeology 5, 175–202. Bonderup, Gerda, 2002: Health Care Provision and Poor Relief in Enlightment and 19th Century Denmark. In: Ole Peter Grell/Andrew Cunningham/Robert Jütte (eds.), Health Care and Poor Relief in 18th and 19th Century Northern Europe. Aldershot: Ashgate, 172–188. Borkenau, Franz, 1980 [1934]: Der Übergang vom feudalen zum bürgerlichen Weltbild. Studien zur Geschichte der Philosophie der Manufakturperiode. unveränd. reprogr. Nachdr. d. Ausg. Paris 1934, Darmstadt: Wissenschaftliche Buchgesellschaft. Brinker, Anne Kathrin, 1994: Armenfürsorge als Sozialpolitik im frühmodernen dänischen Staat. Hamburg: Kraemer. Brown, Dorothy M./Elizabeth McKeown, 1998: The Poor Belong to Us: Catholic Charities and American Welfare Cambridge: Harvard University Press Commission Nationale d’Evaluation du Revenu Minimum d’Insertion, 1992: RMI. Le Pari de l’Insertion. Paris: La Documentation Francaise. Cox, Robert Henry, 1993: The Development of the Dutch Welfare State: From Worker’s Insurance to Universal Entitlement. Pittsburgh&London: University of Pittsburgh Press. Crowther, Anne, 2002: Health Care and Poor Relief in Provincial England. In: Ole Peter Grell/Andrew Cunningham (eds.), Health Care and Poor Relief in 18th and 19th Century Northern Europe. Aldershot: Ashgate, 203–219. Davis, Natalie Z., 1968: Poor Relief, Humanism, and Heresy: The Case of Lyon. In: Studies in Medieval and Renaissance History 5, 217–275. Driver, Felix, 1993: Power and Pauperism. The workhouse system, 1834–1884. Cambridge: Cambridge University Press. Dross, Fritz, 2002: Health Care Provision and Poor Relief in Enlightment and 19th Century Prussia. In: Ole Peter Grell/Andrew Cunningham/Robert Jütte (eds.), Health Care and Poor Relief in 18th and 19th Century Northern Europe. Aldershot: Ashgate, 69–111. Englander, David, 1998: Poverty and Poor Law Reform in 19th Century Britain, 1834–1914. London: Longman. Fairchilds, Cissie, 1976: Poverty and Charity in Aix-en-Provence, 1640–1789. Baltimore: John Hopkins University Press.

293

Faure, Olivier, 2002: Health Care Provision and Poor Relief in 19th Century Provincial France. In: Ole Peter Grell/Andrew Cunningham/Robert Jütte (eds.), Health Care and Poor Relief in 18th and 19th Century Northern Europe. Aldershot: Ashgate, 309–324. Feagin, Joe R., 1975: Subordinating the poor: Welfare and American beliefs. Englewood Cliffs: Prentice Hall. Fix, Brigit/Elisabeth Fix, 2002: From charity to client-oriented social service production: a socialprofile of religios welfare associations in Western European comparison. In: European Journal of Social Work 5, 55–62. Flora, Peter (ed.) 1986: Growth to Limits. The Western European Welfare States Since World War II. Vol: 1: Sweden, Norway, Finland, Denmark. Berlin/New York: Walter de Gruyter. Fösser, 1889: Artikel “Armenpflege“ und “Armenrecht“ Staatslexikon. 465–507. Fraser, Derek, 1976: The Evoluation of the British Welfare State. A History of Social Policy since the Industrial Revolution. London: Macmillan. Geremek, Bronislaw, 1991: Geschichte der Armut. Elend und Barmherzigkeit in Europa. München: Deutscher Taschenbuch Verlag. Gijswijt-Hofstra, Mareike, 2002: Dutch Approaches to Problems of Illness and Poverty between the Golden Age and the Fin de Siècle. In: Ole Peter Grell/Andrew Cunningham (eds.), Health Care and Poor Relief in 18th and 19th Century Northern Europe. Aldershot: Ashgate, 259–276. Gorski, Philip S., 2003: The Disciplinary Revolution. Calvinism and the Rise of the State in Early Modern Europe. Chicago: University of Chicago Press. Gouda, Frances, 1995: Poverty and Political Culture. The Rhetoric of Social Welfare in the Netherlands and France, 1815–1854. Lanham: Rowman&Littlefield. Grell, Ole Peter/Andrew Cunningham, 1997: The Reformation and changes in welfare provision in early modern Northern Europe. In: Ole Peter Grell/Andrew Cunningham (eds.), Health Care and Poor Relief in Protestant Europe 1500–1700. London/New York: Routledge, 1–42. Gutton, Jean-Pierre, 1971: La société et les pauvres. L’exemple de la généralité de Lyon 1534–1789. Paris: Les Belles Lettres. Gutton, Jean-Pierre, 1974: La société et les pauvres en Europe. XVIe-XVIIIe siècle. Paris: Presses Universitaires de France. Heidenheimer, Arnold J., 1983: Secularization Patterns and the Westward Spread of the Welfare State, 1883–1983. Two Dialogues about how and why Britain, the Netherlands, and the United States have Differed. In: Comparative Social Research 6, 3–65. Hill, Christopher, 1952: Puritans and the Poor. In: Past&Present 2, 32–50. Huey, Paul R., 2001: The Almshouse in Dutch and English Colonial North America and its Precedent in the Old World: Historical and Archeological Evidence. In: International Journal of Historical Archeology 5, 123–154. Kahl, Sigrun, 2005: The Religious Roots of Modern Poverty Policy: Catholic, Lutheran, and Reformed Protestant Traditions Compared. In: European Journal of Sociology XLVI, 91–126. Kahl, Sigrun, 2006: The Religious Foundations of the Welfare State: Poverty Regimes, Unemployment, and Welfare-to-Work in Europe and the United States. PhD Thesis, Berlin: Humboldt University.

294

Katz, Michael B., 1996 [1986]: In the Shadow of the Poorhouse. A Social History of Welfare in America. Revised and Updated Edition, New York: Basic Books. Kesselman, Mark, 2002: The Triple Exceptionalism of the French Welfare State. In: Gertrude Schaffner Goldberg/Marguerite G. Rosenthal (eds.), Diminishing Welfare. A Cross-National Study of Social Provision. Westport/London: Auburn House, 181–210. Lacey, Forrest W., 1953: Vagrancy and other crimes of personal condition. In: Harvard Law Review 66, 1203–1226. Levine, Daniel, 1988: Poverty and Society. The Growth of the American Welfare State in International Comparison. New Brunswick/London: Rutgers University Press. Lewis, Jane, 1999: The Voluntary Sector in the Mixed Economy of Welfare. In: David Gladstone (ed.) Before Beverdige. Welfare Before the Welfare State. London: IEA Health and Welfare Unit, 10–17. Manow, Philip, 2004: ‘The Good, the Bad, and the Ugly’. Esping-Andersen’s regime typology and the religious roots of the western welfare state. Max-Planck Institute for the Study of Societies Working Paper 04/3, Cologne: MPIfG. Metz, Karl H., 1985: Staatsraison und Menschenfreundlichkeit. Formen und Wandlungen der Armenpflege im Ancien Régime Frankreichs, Deutschlands und Großbritanniens. In: Vierteljahrschrift für Sozial- und Wirtschaftsgeschichte 72, 1–26. Mitchinson, Rosalind, 2002: Poor Relief and Health Care in 19th Century Scotland. In: Ole Peter Grell/Andrew Cunningham (eds.), Health Care and Poor Relief in 18th and 19th Century Northern Europe. Aldershot: Ashgate, 246–255. Neyret, Guy/Philippe Madinier/Michel Estéve et. al., 1988: Protection Sociale et Pauvreté. Protection légale et expériences locales de Revenu Minium Garanti. Documents du Centre d’Étude des Revenus et des Coûts, No. 88, Paris: La Documentation Francaise. Østergård, Uffe, 1997: The Geopolitics of Nordic Identity – From Composite States to Nation States. In: Øystein Sørensen/Bo Stråth (eds.), The Cultural Construction of Norden. Oslo et. al.: Scandinavian University Press, 25–71. Pedersen, Susan, 1993: Family, Dependence, and the Origins of the Welfare State. Britain and France, 1914–1945. Cambridge: Cambridge University Press. Polanyi, Karl, 1957 [1944]: The Great Transformation. The Political and Economic Origins of Our Time. Boston: Beacon Press. Pullan, Brian, 1988: Support and Redeem: charity and poor relief in Italian cities from the fourteenth to the seventeenth century. In: Continuity and Change 3, 177–208. Quadagno, Jill, 1988: The Transformation of Old Age Security: Class and Politics in the American Welfare State. Chicago: Univeristy of Chicago Press. Ramsey, Matthew, 2002: Poor Relief and Medical Assistance in 18th and 19th Century Paris. In: Ole Peter Grell/Andrew Cunningham/Robert Jütte (eds.), Health Care and Poor Relief in 18th and 19th Century Northern Europe. Aldershot: Ashgate, 280–308. Riis, Thomas, 1997: Poor relief and health care provision in sixteenth-century Denmark. In: Ole Peter Grell/Andrew Cunningham (eds.), Health Care and Poor Relief in Protestant Europe 1500–1700. London: Routledge, 129–146. Rimlinger, Gaston, 1971: Welfare Policy and Industrialization in Europe, America, and Russia. New York: John Wiley & Son.

295

Ritter, Gerhard A., 1991: Der Sozialstaat. Entstehung und Entwicklung im internationalen Vergleich. München. Rüger, Willi (ed.) 1932: Mittelalterliches Almosenwesen. Die Almosenordnungen der Reichsstadt Nürnberg. Nürnberg. Schmid, Josef, 1995: Verbändewohlfahrt im modernen Wohlfahrtsstaat: Strukturbildende Effekte des Staat-Kirche-Konflikts. In: Historical Social Research 20, 88–118. Schultheis, Franz, 1988: Sozialgeschichte der französischen Familienpolitik. Frankfurt/Main: Campus. Stolberg, Michael, 2002: Health Care Provision and Poor Relief in the Electorate and Kingdom of Bavaria. In: Ole Peter Grell/Andrew Cunningham/Robert Jütte (eds.), Health Care and Poor Relief in 18th and 19th Century Northern Europe. Alsdershot: Ashgate, 112–135. Tawney, Richard Henry, 1958: Foreword to “The Protestant Ethic and the Spirit of Capitalism“ by Max Weber. New York: Charles Scribner’s Books. Tawney, Richard Henry, 1990 [1922]: Religion and the Rise of Capitalism. A Historical Study. Holland Memorial Lectures 1922, Middlesex et.al.: Penguin Books. Therborn, Göran, 1994: Another way of taking religion seriously. Comment on Francis G. Castles. In: European Journal of Political Research 26, 103–110. Thorkildsen, Dag, 1997: Religious Identity and Nordic Identity. In: Øystein Sørensen/Bo Stråth (eds.), The Cultural Construction of Norden. Oslo et al: Scandinavian University Press, 138–160. Uhlhorn, Gerhard/Aschrott, 1859: Artikel “Armenwesen“. In: Conrad, et al. (eds.), Handwörterbuch der Staatswissenschaften. Jena, 819–924. Wagner, David, 2005: The Poorhouse: America’s Forgotten Institution. Lanham: Rowman & Littlefield. Waldau, Georg Ernst (ed.) 1789: Vermischte Beiträge zur Geschichte der Stadt Nürnberg, Bd. IV. Nürnberg. Weber, Max, [1904–05] 1958 [1904/05]: The Protestant Ethic and the Spirit of Capitalism. New York: Charles Scribner’s Sons. Weber, Max, 1996 [1904/05]: Die protestantische Ethik und der “Geist“ des Kapitalismus. Textausgabe auf Grundlage der ersten Fassung von 1904/05 mit einem Verzeichnis der wichtigsten Zusätze und Veränderungen aus der zweiten Fassung von 1920. Auflage. Weinheim: Beltz Athenäum Verlag.