Irita Kise,

politics, 3rd yr

Theories of Welfare State Expansionism:

Do Parties Matter?

The major trend after the second world war in Europe has been the economic growth and the growth of the state. Percentage of revenues raised by means of taxes in GNP has grown from level of 25% to roughly 45% in 1985. (Lane and Ersson, 1994:327) Such changes in public sector have aroused interest of scientists and a number of theories has been developed. Until now, no one has been able to work out a theory which would explain all because developments in different countries are too different. Therefore, scientists have made efforts to find variables that determine why in one country like in Sweden tax state has expanded from 26% (1950) to 63% (1990) but in others like Switzerland from 25% to only 34%.

Besides demand and supply theories another dimension was introduced: does politics matter? Thomas Dye in 1966 raised the issue of the role of politics as cause of public policy. It used to be believed that political parties were the main intermediators between citizens and the outputs of governmental activity meaning that it made a difference which political party was in power. However, Dye argued that broad environmental factors such as affluence, urbanisation and industrialisation meant much more. (Lane and Ersson, 1994:340)

The aim of this essay is to examine the hypothesis that parties do not make any difference and that environmental factors like demographic, economic, cultural and institutional play greater role in determining public sector size and structure. Actually, all these factors including political ones must be regarded as variables and not as separate explanatory theories. They all are present in modern state and they all in different degree influence government outcomes. I will look at influence of parties without denying the significance of other variables because “explanation of public policy outcomes in terms of any single paradigm is unlikely to be a fruitful way forward in comparative public policy”.(Castles, 1982:3) The question about party influence on government outcomes aroused my interest because the statement “If voting changed anything they’d abolish it” (Ken Livingstone) undermines the image of the very basis of representative democracy. Although elections have no necessary connection with democracy, they are thought to be the first and for some citizens the only means for influencing government policy. If we assume that there is no difference which party makes the government and the policy outcome is determined by other factors, then people’s choice in elections is only an illusion of democracy. To believe in such thesis is very hard because it means that the choice we make in elections does not change anything and all party programmes between which we choose are only means for parties to get more votes. Francis Castles asks: “ What price democracy, if it does not matter who you vote for or what party is in office?” (Castles, 1982:5) Then another question seems to be important: do parties themselves decide to let all as it is or is it more dependent on environment and parties must follow the development without any choice to influence it?

Several theories have stated that parties do not influence policy outcomes. Thomas Dye has argued that “once one has controlled for various measures of economic development ( such as personal income, urbanisation, industrialisation and education level), the apparent connection between party competition and state expenditure disappears”. (Castles, 1982:25) He must admit that in some cases the link between parties and policies exist and only in society which is rather homogeneous it is absent. In homogeneous societies the gap between different social groups is not critical, parties mostly are catch-all parties whose ideologies do not differ in such extent to radically change policies. Dye comments: “Party influence is only effective where the parties represent separate and distinct socio-economic coalitions. Where the constituencies are divided along social and economic lines and where the party division coincides with these constituency divisions, only then will party program and discipline be effective in shaping policy in legislative chambers.” (Dye, 1978:303) It is evident that changes in Europe have led to more homogeneous societies. There are no more separate workers’ parties or rural parties with voters only from those social classes. Parties do not engage themselves in bitter ideological battles with each other. Value systems of people change. Many authors have used the changes in Western European countries to explain the welfare state development. Alber has found that only before the second world war political forces were of great significance in shaping public policy. He has compared the growth of welfare state from its beginnings in Europe. The thesis of divided societies has proved to be true. For example, at the end of the 19th century authoritarian regimes were active to introduce social insurance programmes because their legitimacy was threatened by the political mobilisation of the working class. (Castles, 1982:32) After the second world war, Alber and others argue, “social policies appeared to be a by-product of prosperity throughout Western Europe, independent of the inner-political situation.” (Castles, 1982:32) This leads to another two theories which try to describe and explain present situation - convergence theory and the “end of ideology”. Already in 1950s the changes in society were described as the end of ideology and tied to economic development. S.Lipset writes: “This change in political life reflects the fact that the fundamental political problems of industrial revolution have been solved; the workers have achieved industrial and political citizenship; the conservatives have accepted the welfare state; and the democratic left has recognised that an increase in overall state power carries with it more dangers to freedom than solutions for economic problems.” (Christenson, 1972:298) I can conclude that two significant changes which have been unanimously recognised as shaping the modern society are: the elimination of class conflict and the economic affluence. Technological change has produced consensus and similarity. That serves for ground in convergence theory. Galbraith, Kerr and others suggest that “there is a logic of industrialism in which the exigencies of modern technology and an advanced economy override political factors making for diversity, and progressively shape social structures and public policies in a similar mould.” (Castles, 1982:6) Definitely, the process of technological change has shaped Western Europe. Now it has not only similar economic processes but the countries are interdependent. Information flows have made more possible “snowballing” effect. The integration process in EU has provided with new chances for convergence although not all European countries are involved in the process. Those who want to join EU are eager to change their legislation in order to meet the requirements which exist in European Union. In such a way public policy, especially social policy, becomes more and more similar in the whole Europe independent on which party is in government in each country.

The decline of ideological factor is proved by several researches. In 1975 Wilensky examined the impact of GNP, age of social security systems and age structure of the population on the percentage of GNP devoted to social security. His conclusions support the convergence theory. “Economic growth makes countries with contrasting cultural and political traditions more alike in their strategy for constructing the floor below which no one sinks. The richer countries become, the more likely they are to broaden the coverage of both population and risks.” (Castles, 1982:28) Wilensky’s findings also prove the hypothesis of the “end of ideology”.

Very important changes which have influenced all European states are demographic changes. Increasing proportion of older people equally affects all countries and governments must provide services, health care, pensions for that part of population whose needs are very demanding. That determines that increase in social spending by no means is dependent on political forces in government but is inevitable response to dramatic demographic changes. Similarly, other economic factors as unemployment and inflation are characteristic for the whole Europe due to its interdependence, affect government outcomes in rather similar ways.

There are other theories that do not put political factors on the top when explaining the growth of welfare state. Cameron argues that the openness of the economy contributes to the expansions of the public economy. (Castles, 1982:31) It seems as if the extent of openness in an economy depends on how large the nation is in terms of population. The larger the nation the less open is the economy. It is quite understandable that state must provide security for its citizens which is considerably less in small European countries.

Another theory seeks the link between political centralization in federal and unitary states and public policy. The findings show that the less centralized is country the lower public spending it has. Institutionalists think that patterns of governance matter ( the rules of electoral competition, the relationships between legislature and executive, the role of the courts, etc.). Institutions establish the rules of the game for political struggles and they affect government capacities. Institutionalists argue that strong states are likely to produce strong welfare states. Where political authority is fragmented, minorities can block social legislation. Not only federalism, but also separation of powers , strong bicameralism or reliance on referenda may restrict welfare state development. This approach is very useful in explaining the slow growth of welfare expenditure in Switzerland, which is very decentralized state. But it fails to meet the description of developments in Germany, also very decentralized state.

Harold Wilensky has also developed a theory which explains public expenditure by cultural variables. Already Max Weber confronted two society patterns of Protestantism and Catholicism. The differences between both are clearly visible.
The most famous comparison - collectivism versus individualism - is also of importance in explaining social spending. Although the socio-economic processes in Europe have made countries more similar the differences based on traditional political culture have remained but these differences cannot be fully expressed because of economic problems which require in most cases universal responses. Some other scientists also have used religion as a variable in social policy analysis. Most recently, Frances Castles in his article suggests that “differences in religious adherence and/or in degrees of secularisation between advanced nations may be as relevant to understanding cross-national variance in a wide range of public policy outcomes as the impact of socio-economic and political factors”. (Castles, 1994) He does not deny the impact of economic and political factors and admits that he himself has taken part in the discussion about which factor was more important. That is characteristic to writings on this theme in last couple of years: more and more authors acknowledge that economical, political and institutional factors of welfare state development cannot be separated. Castles offers a division of states in families of nations based on religion. Each “family” has its own distinct pattern of social policy “as a result of shared historical and cultural attributes”. (Castles, 1994) He concentrates on Catholicism and identifies a Catholic family of nations in which are included Southern European countries. The difference between Catholic and Protestant countries is reflected in their social policy structure. For example, the countries with strong Protestant state churches developed early a notion of state responsibility for public welfare, whereas in the religiously mixed and Catholic countries the tradition of private charity and the principle of subsidiarity, giving priority to the responsibility of smaller collectivities, remained strong. (Flora and Alber, 1981:43)

Having looked at some theories which emphasise other determinants of public policy than political ones , now it is time to examine some ideas which advocate the “politics matters” position.

The socio-economic factors could not explain all differences between social policy patterns in different countries. While this argument explained broad social policy differences between rich and poor nations, it proved to be far less helpful in explaining differences within the scope of rich democracies. That left room for other factors of which one was political factor and especially class struggle, worker mobilization and strength of left and right parties.

The strength of political explanation is that it points out the importance of necessary precondition of the welfare state: political decision-making. Social expenditures do not grow automatically, but their growth is initialized and supported by political decisions. In a sense even those automatic or built-in growth tendencies caused by demographic changes are within the powers of political decision making. As important as political factors may appear to be, they might be merely intermediate factors mediating the effect of structural and economic variables and having only minor causal significance. (Uusitalo, 1984) Those who believe in importance of political determinant have been quite united in making conclusions about different parties’ influence on public policy patterns, although there are problems in addressing particular policies to parties. Very often, especially in last decade, parties can no more be associated with particular kind of ideas and distinct group of supporters. That is due to the changes all over the Europe. Voter volatility, class voting replaced by issue voting, erosion of ideologies. Politicians generally worry first and foremost about getting elected. Helping improve the economy may make that easier, but not if it requires hugely unpopular policies, and not if the economic benefits are likely to appear at some point in the distant future and will not influence the results of the next election. The fact that parties no more are strong advocates of one social group interests weakens the argument about party significance in welfare state development. Ideology and class interests are no more equal with the term “party”. In the past the ideological factor played greater role and parties representing each wing of left-right scale could be identified more easily. Political scientist have made researches on these divisions: left / non-left, right / non-right and Christian Democratic Parties. First there was asumption that left parties in government contribute to expansion of welfare state. That is not unexpected due to the well-known ideology of the left-wing parties. Later the division was changed to right / non-right when some researchers found that also centrist governments favoured growth of welfare state and only right-wing parties were the ones which wanted to restrict social spending. A number of consequences of leftist government for social policy have been identified in lieature. Douglas Hibbs in 1977 examined the impact of parties on macroeconomic policy: “The “revealed preference” of leftist governments has been for relatively low unemployment at the expense of high rates of inflation, whereas, comparatively low inflation and high unemployment characterize political systems dominated by center and right-wing parties.” (Hibbs, 1977) Other leftist “success”: reducing gap between rich and poor (redistributive income policy), welfare expansion. These characteristic ideological objectives could be achieved if the government worked in political vacuum. In every democratic state there exists opposition whose influence cannot be underestimated. Hicks and Swank verify that welfare policies of Left-led governments may be moderated by strong center and right oppositions and conservative governments under the pressure of left parties must adopt welfare-expanding reforms. (Hicks and Swank, 1992) It is also argued that in highly competitive electoral environments parties make pro-welfare appeals to “median” voters in hopes of gaining electoral majority, but not always the pre-electoral promises are kept. Theories of electoral cycle emphasize that the pro-welfare reforms are introduced by incumbent party before elections in order to remain in office. Sometimes it results in huge economic problems after elections, but then there is time to coup with these problems until next elections.

Strength of Christian democratic parties was the measure used by Wilensky in 1981 demonstrating a linkage between Catholicism and social security expenditure effort.

The existing variance of welfare state patterns shows that some ideological factors have been working at least in the formation years of welfare state. Esping-Andersen, having examined data from 18 developed democratic countries devides them in three types of welfare-state regimes: liberal (USA, UK, Canada, Australia), corporatist ( Austria, France, Germany) and social democratic (Scandinavia). “The historical forces behind the regime differences are interactive. They involve, first, the pattern of working-class political formation and, second, political coalition-building in the transition from a rural economy to a middle-class society.” (Esping-Andersen, 1990:32) So different welfare-state regimes can be associated with different ideologies and parties in government. Joan Higgins illustrates that with facts from the 1950s when successive political parties in government in the UK and USA were responsible for shifts in emphasis (from punitive and deterrent programmes to liberal and generous policies and back once more to selectivist, regressive programmes). (Higgins, 1981:52) The most popular example in “parties matter” discussion has been Sweden and other Scandinavian countries where Social Democratic parties have dominated governments since 1945. In these countries the welfare state growth has been very significant.

“Comparative public policy analysis is a child of both sociology and political science. It has been an area of profound contestation woth initial hegemonic claims that only industrialization or politics mattered gradually giving way to an uneasy truce built on the basis of multivariate findings which showed that both could matter simultaneously.” (Castles, 1994) Some areas are more likely to be affected by overtly political factors, some - by economic. There is difference between deciding ewpenditure priorities, modifying income distribution by taxes and controlling inflation and unemployment. Nevertheless, the process of policy making is complex and it has various determinants which differ from time to time and from state to state. That alliws to hope that our voting means something, at least parties must accomplish their promises in order to get voters’ trust and enough votes in next elections. That in turs means that our choice can be regarded as decisive.

Literature:

Castles, Francis (1982) “The Impact of Parties”

Castles, Francis (1994) “On Religion and Public Policy: Does Catholicism Make a Difference?” from European Journal of Political Research, vol.25, pp.19-40

Christenson, Reo (1972) “Ideologies and Modern Politics”

Dye, Thomas (1978) “Understanding Public Policy”

Esping-Andersen, Gosta (1990) “The Three Worlds of Welfare Capitalism”

Flora, Peter and Alber, Jens (1981) “Modernization, Democratization and the Development of Welfare States in Western Europe” from Flora and Heidenheimer eds. “The Development of welfare States in Europe and America” pp.37-80

Hibbs, Douglas (1977) “Political Parties and Macroeconomic Policy” from APSR, vol. 71, No.4, pp.1467-1487

Hicks, Alexander and Swank, Duane (1992) “Politics, Institutions and Welfare Spending in Industrialized Democracies, 1960-82” from APSR, vol.86, No.3, pp.658-674

Higgins, Joan (1981) “States of Welfare. A Comparative Analysis of Social Policy”

Uusitalo, Hannu (1984) “Comparative Research on the Determinants of the Welfare State: the State of the Art” from European Journal og Political Research, vol.12, pp.403-422