State subsidies to political parties: Confronting rhetoric with reality

Transcription

1 West European Politics ISSN: (Print) (Online) Journal homepage: State subsidies to political parties: Confronting rhetoric with reality Jon Pierre, Lars Svåsand & Anders Widfeldt To cite this article: Jon Pierre, Lars Svåsand & Anders Widfeldt (2000) State subsidies to political parties: Confronting rhetoric with reality, West European Politics, 23:3, 1-24, DOI: / To link to this article: Published online: 03 Dec Submit your article to this journal Article views: 887 View related articles Citing articles: 35 View citing articles Full Terms & Conditions of access and use can be found at

2 State Subsidies to Political Parties: Confronting Rhetoric with Reality JON PIERRE, LARS SVÅSAND and ANDERS WIDFELDT This article investigates the political impact of the introduction of state subsidies to political parties. The arguments for and against subsidising political parties are outlined. Different models of party subsidies, and their regulatory frameworks, are discussed. We find little evidence of a cross-national impact of the introduction of party subsidies. The subsidies cannot explain the decline in party membership. Nor is there evidence to suggest that the subsidies were introduced as a response to membership decline. There is no support for the allegation that party subsidies lead to the petrification of party systems. The subsidies have not meant that other income sources have lost their significance for political parties. The introduction of public subsidies to political parties and their organisations is often described as one of the most significant developments in the relationship between political parties, the state and the citizenry in the advanced Western democracies. Some see such subsidies as a major threat to the historical role of parties as voluntaristic organisations and, ultimately, to the whole notion of political parties as participatory and representative structures. Others argue that subsidies have profound systemic effects, both in terms of consolidating the existing party system and also in terms of changing the nature of competition within the system. Certainly, a number of positive aspects of the subsidies have been highlighted in the debate as well, but even so the overall tenor of the debate has been quite critical of the subsidies. We are bothered by what we believe to be a distinctive gap between a normative debate concerning public subsidies on the one hand and a wealth of empirical data on party system change and party membership development on the other. Thus, this article confronts some of the normative assessments of public subsidies to political parties with empirical evidence of the developments which the subsidies are said to manipulate. The West European Politics, Vol.23, No.3 (July 2000), pp.1-24 PUBLISHED BY FRANK CASS, LONDON

3 2 WEST EUROPEAN POLITICS purpose of the analysis is not to provide a conclusive and final assessment of public subsidies to political parties but rather to investigate to what extent the different views on the subsidies are supported by empirical data. The argument proceeds in four steps. First, we rehearse the debate on public subsidies to the political parties and discuss the methodological complexities associated with the study of party finance. Following that, we describe in some detail the trajectory of public subsidies to political parties in a comparative perspective and the current structure of the subsidy systems in different jurisdictions. Once we have established the conceptual and empirical base we proceed to the third section, which assesses the impact of the subsidies on the three aspects of party development we are focusing on in the present analysis. A concluding section draws together the main findings. THE PUBLIC SUBSIDY DEBATE The introduction of various forms of public subsidies to political parties has brought forward a number of critical arguments against public financial involvement in the party organisations. We focus our analysis on three different aspects, or sources of controversy, of the subsidies. One topic of debate relates to the process through which the subsidies emerged. It has been suggested that the parties used the privileged position they have as actors in the legislative and executive branches of government and simply chose to allocate public money to themselves. Subsidies, in this perspective, are proof of the parties' successful co-optation of the state. Perhaps most importantly, the introduction of public funding is said to have blurred or confused the distinction between the state and political parties as voluntaristic, representative structures. 1 The growing dependence on the public treasury leads parties to become more interested in managing the state for their own good rather than for the polity tout court. This tendency is a defining characteristic of the development towards what Katz and Mair refer to as 'the cartel party' ; the state cannot be thought of as an autonomous actor when deliberating the introduction of subsidies to the parties, since the basic principles of state subsidies for parties are determined by parliamentary assemblies controlled by political parties. 2 Political parties, the argument goes, are the only voluntaristic organisations in society that themselves define the rules for their exchange with the state. Also, the argument has been made that state subsidies might entail some kind of control from the state over the parties. 3 Thus, public subsidies would create new problems at the same time as they purportedly solve other problems; public financial support to the parties cannot guarantee the supremacy of the parties in the democracy, but may well jeopardise the representative and

4 STATE SUBSIDIES TO POLITICAL PARTIES 3 voluntaristic nature of the parties by effectively placing parties under state control. 4 A second critique of subsidies holds that they interfere with party system change. They are said to do so because they give established parties a much stronger position than emerging party formations and hence petrify or cement the party system. 5 This argument is to some extent related to the previous comment that parties penetrated the state apparatus in pursuit of their own interests; that is to say, the established parties shared an interest in raising the thresholds for new parties. 6 Since the parties in government and parliament design subsidies themselves, the subsidy system is likely to be structured in ways which favour the established parties. This, in turn, interferes with the 'Darwinistic' life-cycle process of party formation, maturity and decline; 7 public subsidies are said to be a life-support system for ageing parties which obstruct the natural, organic development of the party system. Subsidies, in this view, sustain the status quo ante in the party system by strengthening the competitive advantage of the established parties vis-à-vis new actual or potential party formations. A third common critique of subsidies suggests that they virtually abolish the parties' interests in sustaining or increasing their membership. Once public subsidies are introduced, the argument goes, the parties' dependence on membership dues and other voluntary financial contributions to the party declines. By letting the state pay for the parties, an important incentive for parties to mobilise and activate members is abolished. Add to this that changing campaign techniques mean that party organisations will rely less on members in their campaign work, and the party leaderships' preoccupation with electoral performance, there will be less interest in catering for the interests and opinions of the rank-and-file in the intra-party decision-making process. As a result, traditional representative structures will lose much of their capacity and incentives to function as participatory, representative and communicative channels in the political system. For the individual member, involvement becomes less interesting because members no longer control a critical resource in their exchange with the party leadership; as Hirschman put it, 'exit' as a bargaining threat loses its significance when the consequences of 'exit' are reduced. 8 Public subsidies therefore contribute to the alienation and indifference of the rank-and-file membership and increase the centralisation of the party organisation, besides other factors that also push the parties in the same direction. In addition, the electorate will feel increasingly alienated from the parties because participation does not seem to have any real effect on the actual operation of the parties. Taken one step further, this critique sees political parties as more concerned with maintaining a good relationship with the state than with its constituencies.

5 4 WEST EUROPEAN POLITICS Thus, public subsidies may in fact exacerbate some of the problems they were intended to resolve. 9 However, there would not have been a debate if subsidies did not have strong advocates, too. Three overarching motives for the state to offer financial support to political parties have been advanced in the debate. First, political parties are seen as critical to democratic government; democracy as we know it is inconceivable without strong parties and party organisations. Parties perform roles such as electoral mobilisation, interest aggregation, socialisation of citizens, recruitment of future leaders and formulation of policy alternatives, which no other types of organisations can perform. 10 Some of these roles are said to be defined above and beyond partisan competition - for instance, by their mere existence, political parties offer support to the democratic process and democratic government - hence, it is only fair that the state offers some degree of compensation to the parties. The political and democratic nature of the modern state depends directly on the ability of the parties to serve as integrative links between state and society. Public subsidies to political parties contribute to this end. A second argument supporting state subsidies is also derived from the overarching democratic perspective on the parties, but looks at the role of the parties from the vantage point of the citizen. Political parties, in this view, are the key structures through which citizens can express their political opinions, and also the main instrument by which they can become actively involved in the governing of society. However, all societies display an enormous variation in resources - financial or otherwise - among citizens as well as among different groups and organisations." These differences, advocates of subsidies argue, could easily be translated into differences in political clout among different constituencies. Therefore, in order to safeguard political equality, the state must regulate the amount of financial resources that donors may provide to parties or candidates. Furthermore, by supplementing this regulation with public subsidies to the parties, the state can mitigate the problem of private (or corporate) support to some parties and ultimately contribute to ensuring responsiveness among the parties vis-à-vis the citizenry. A third argument in favour of subsidies, finally, relates to the costs of running a political party in a modern democracy. Political parties have a near monopoly in selecting candidates for public office, but interest groups and think-tanks exert a great deal of political influence. The problem here is that most political parties' financial resources do not come near to those controlled by many interest organisations, think-tanks and pressure groups. The growing significance of media campaigning has entailed tremendous escalation of the cost of electoral campaigns, not least in the United States. Also, parties are assumed to present coherent and appropriate policy

6 STATE SUBSIDIES TO POLITICAL PARTIES 5 proposals on a wide range of issues and policy sectors, something that requires a substantial staff and institutionalised expertise. Thus, if parties are to sustain their central role in the political system and be able to communicate their message to the voters, some financial support from outside sources becomes necessary, as membership dues alone cannot carry the financial burden of the modern party organisation. Thus, much of the debate has revolved around the issue of whose selfinterest is being pursued - that of the state or that of the parties - and, in a similar vein, who has co-opted whom in the process of designing state subsidies to the parties. Critics identify several potential consequences of the subsidies with regard to the relationship between citizens and parties. Meanwhile, supporters of the subsidies make a strong case that parties are critical to the democratic government of the polity. This suggests that we must use criteria at different institutional levels as we assess the impact of the subsidies on the party system and the political system. Public subsidies to political parties have now been implemented in several countries since the mid-1960s. This means that, to the extent that the critique is valid, we should now be able to see fairly clear signs of changes in party system development, or status of party members, in the advanced Western democracies. This article is an historical, evaluative analysis of the subsidies and the debate which has surrounded them. Empirically, we are drawing on a comprehensive data collection about the organisational development of political parties. The data collection was organised by the international comparative project on party organisation led by Richard Katz and Peter Mair.' 2 The dataset includes parties in Austria, Belgium, Denmark, Finland, Germany, Ireland, Italy, the Netherlands, Norway, Sweden, United Kingdom and United States. Thus it covers a variety of types of party system, both in terms of number of parties, ranging from typically competitive two-party systems like Britain and the United States, to more fragmented systems like Belgium and the Netherlands. It also includes countries in which the party system has remained relatively stable, Austria and Sweden, for example, to systems that have changed considerably, such as Denmark. The dataset also covers a wide variety of institutional arrangements: presidential and parliamentary systems, unitary and federal systems and a variety of electoral arrangements. Most important, the countries discussed here cover a range of different types of public subsidies to parties - including Britain, which is somewhat unique in not having introduced any direct funding for parties as such. Although the dataset covers several types of democratic systems it is far from complete. Long-standing Western democracies such as France, Switzerland, Luxembourg and Iceland are not included. Nor will our findings necessarily be relevant for the newer Western democracies of Spain, Portugal and

7 6 WEST EUROPEAN POLITICS Greece, not to speak of the 'third wave' democracies of central and eastern Europe. 13 Certainly, one could argue that this dataset is ageing, but given our historical approach that is not a significant problem. When the data was collected - around any changes in the development of the parties and their organisations triggered by the subsidies should have been observable. In addition, there are a number of more recent reports on comparative party finance and which provide additional empirical data on state subsidies. 14 MODELS OF PUBLIC SUBSIDIES In all contemporary Western democracies, the state offers some kind of financial support to political parties and their organisations. However, they do so according to different criteria, for different types of expenditures, and they identify different types of recipients. The subsidy programmes differ significantly among different jurisdictions, as we shall see below. Strange as it might seem, the main conceptual problem in comparing public subsidies for parties is defining 'party' and 'public subsidy'. First, who are the recipients of the subsidies, or, to put it more accurately, what does it mean that the parties are subsidised? We can distinguish between three different types of recipient: the party groups in the parliamentary assemblies; the national party organisation; and a particular branch or specialised section of the party. Table 1 provides information on when different types of public subsidies were introduced in the Western democracies. Subsidies to Parliamentary Parties Financial support to parliamentary groups is the oldest type of subsidy to political parties. This type of subsidy has generally been subject to less debate compared with subsidies to national party organisations. Supporting party groups in parliament could be legitimised on the grounds that such support served to strengthen the elected assembly vis-à-vis the executive TABLE 1 INTRODUCTION OF PUBLIC SUBSIDIES FOR PARLIAMENTARY GROUPS AND FOR ELEMENTS OF THE PARTY ORGANISATION Pari. Nat. org. Pari. Nat org. Pari. Nat. org. Austria Belgium Ireland Norway none 1970 Finland Italy* Netherlands UK none Sweden Denmark Germany USA Note: * Party subsidies were abolished in Italy Source: Katz and Mair (eds.), Party Organizations.

8 STATE SUBSIDIES TO POLITICAL PARTIES 7 branch of government. Furthermore, subsidies to the parliamentary parties were easily controlled by the state; hence there was little risk that money would be spent for purposes which could be called into question. That said, it is clear that as parliamentary support expanded it also had an impact on party organisation, if for no other reason than by relieving the party of some of the expenditures generated by the parliamentary party. However, financial support for parliamentarians is not necessarily the same as support for parliamentary parties. Farrell argues that Irish parliamentary parties have received 'substantial financial support from public funds'. 15 Among these categories are postal and travel allowances for members of parliament. Such subsidies, however, are also found in other countries and in periods without any political parties. 16 Thus, even for parliamentary support there are variations in terms of identifying the recipient, that is, the individual MP or the parliamentary party. Even between parliamentary groups there may be differences. In Ireland, for instance, the incumbent party did not receive any public funding until 1973 although the main opposition party had done so since the 1960s. 17 Even where there exists a clear distinction between support to the parliamentary groups and the regular party organisations, as indicated in Table 1, it is in practice very difficult to treat these as completely separate schemes because the regulatory frameworks frequently offer considerable discretion on how - and by whom - the money is spent. In Belgium, for instance, there were no direct subsidies to party organisations until When subsidies to the parliamentary parties were introduced in 1971, the subsidies were given to the central party organisation for distribution to the parliamentary parties with no formal control over the use of the financial resources. 18 In Italy, by contrast, the parliamentary party could decide for itself how the subsidies should be allocated. 19 In Denmark, the national party organisations moved part of their staff to the parliamentary secretariat when subsidies to parliamentary staff were introduced. Thus, although subsidies to national party organisations proper were introduced relatively late in Belgium and Denmark and 1987, respectively - both countries had witnessed indirect state support to the national organisations for some time. The late introduction of general party support in Denmark was the last element in a 'chain of events'. 20 A similar conceptual problem is related to the distinction between the party organisation strictu sensu and the specialised branches of the party. The dates indicated in Table 1 refer to the introduction of public subsidies to some unit in the extra-parliamentary party. In the Scandinavian countries, party organisations receive the subsidies. In Austria, the first extra-parliamentary subsidy was given to newly created party education institutes. Two years later this support was extended to the parties themselves. In the Netherlands,

9 8 WEST EUROPEAN POLITICS only specific elements of the parties such as educational organisations and women's and youth organisations are eligible for public subsidies. In Germany, the 1967 German Party Law allowed support to 'party foundations' and also reimbursements to parties for election expenses, but no direct and generalised payments to political parties as such. In 1983, however, direct party subsidies were introduced within certain limits. The American system of public subsidies, finally, differs from its European counterparts in some respects. The main concern, which propelled the introduction of public subsidies, was that large donors might exert influence on candidates. Thus, the 1974 programme regulated fundraising in some detail and, at the same time, introduced public financial support. However, most of the funding is not targeted at parties but at candidates. An even more problematic concept is that of 'subsidy'. In its simplest form, 'subsidy' refers to cash money paid to a recipient. However, it is generally recognised that this is not a very satisfactory definition. Thus, free access to radio and TV during election campaigns in countries with public service-style aired media could well be regarded as an indirect subsidy to the parties. By the same token, state subsidies to the press, as has been provided in the Scandinavian countries for a considerable time, is to the benefit of the parties because of the close connections between parties and newspapers. Defining 'public subsidies' is also complicated by the fact that in many countries' subsidies are provided at several tiers of government. Particularly in federal states, sub-national party subsidies appear to be a significant source of income for the parties. Thus, Müller notes that regional subsidies for Austrian parties in the period were three times the total of federal subsidies. In Sweden, regional and local subsidies are more than twice as large as the support provided by central government. 21 In addition, parties in EU member states receive subsidies for the European parliamentary election campaigns. Subsidies to National Party Organisations The structure of public subsidies for party organisations is far more complex than for the parliamentary fraction. Four variables seem to separate countries into different categories (see Table 2). First, different countries have different rules for identifying what party unit(s) should be the recipient(s) of the subsidy. Second, there is substantial cross-national variation with regard to what qualifying criteria parties must meet in order to receive public subsidies. Third, there are distinctions between countries with regard to what could be called the targeting of the subsidies, that is, for what purposes the public subsidies may be used. Finally, countries differ with regard to the control exercised by the state over the use of the subsidies.

10 STATE SUBSIDIES TO POLITICAL PARTIES 9 Recipients In most countries offering some type of public subsidy, the party organisation is the main recipient. However, there are often several recipients. We note a clear tendency in many countries to increase the number of recipients as a means of diversifying the subsidies. Party research institutes, educational institutions and local and regional branches of the party organisations have been added to the list of potential recipients in many jurisdictions. This diversification has greatly increased the total amount of subsidies. We also note the sharp distinction between the United States and the European countries with regard to the definition of recipients: the American (federal) public subsidies are much more candidate-oriented than party-oriented. Criteria With regard to the eligibility criteria in the public subsidy systems, there are two different principles, which sometimes are combined: winning of at least one parliamentary seat and a minimum level of electoral support. In Belgium, Finland and the Netherlands, only parties winning seats in parliament qualify for subsidies to the party organisation. In Norway, Denmark and Germany (since 1967), on the other hand, subsidies relate entirely to voting support. Austria, Italy and Sweden combine the two in an either/or combination. 22 There is a general development towards more liberal criteria for subsidies across western Europe. The only exception to the rule is Norway, where in 1977 a minimum level of 2.5 per cent of the votes was introduced. One way of preventing parties and candidates from becoming dependent on the state has been the 'matching funds' model; in the United States, Germany and the Netherlands subsidies are tied not only to electoral support but also to the ability of candidates and parties to raise funds themselves. 23 Targeting In the Scandinavian countries, the state imposes only very few restrictions on how the parties spend the financial resources they have been granted by the subsidy system. In Finland, however, the parties must transfer some of their subsidy to women's organisations and to lower levels of the party organisation. Since 1990, part of the Finnish party subsidy is also earmarked for international activities. The Dutch party support is specifically targeted at research activity (since 1971) and education (since 1975). Austria added an extra subsidy for campaign expenses in 1990, but the general party subsidy is not targeted. The most elaborate rules concerning what kind of funding can be used for which purposes, and by whom, can be found in the United States. 24

11 TABLE 2 REGULATORY FRAMEWORK OF STATE SUBSIDIES FOR EXTRA-PARLIAMENTARY PARTY ORGANISATIONS Country Year Recipient Basic Allocation Criteria Targeting Accountability Austria Party educational unit Central party organisation Parliamentary representation Parties with 1% of vote receive money in election years. Parties with seats in parliament receive money permanently Training of party staff (extra campaign contribution to parties with seats in parliament introduced 1990) Publish party budget; report campaign expenditures (the latter abolished 1990) Belgium 1989 Central party organisation (via state agency)* Parliamentary representation Accounts of state agency* to be public Denmark Central, regional and local party organisations Minimum 1000 votes (central), 500 votes (regional) and 100 votes (local) Declare that money has been spent on political activities Publish accounts Finland Central party organisation Parliamentary representation 8% to be transferred to women's organisation 8% to be transferred also to regional levels ; also targeting international activity Report on the use of public funds Report entire party accounts Germany Central party organisation Parliamentary representation Min. 2.5% of 2nd votes or 10% of 1st votes Min. 0.5% of 2nd votes Campaign reimbursement ; direct subsidy also introduced Publish income accounts + publish expenditure accounts Italy** 1974 Central party (paid via parliamentary group) Parliamentary representation. Campaign expenses: 2% of vote (differs for regional parties) Regular expenses; separate subsidy for campaign expenses Publish party budget

12 TABLE 2 (Com.) Country Year Recipient Basic Allocation Criteria Targeting Accountability Netherlands*** Party research institutes Party educational institutes Basic amount + per parliamentary seat. Not to exceed what party can raise on its own, but not to exceed 10% of salary expenditure and 30% of all expenditures Research Education Ministry of finance to approve accounts Norway Sweden Central party organisation Regional and local party organisations Central party organisation Central party organisation Central party organisation Regional and local party organisations Central party organisation Central party organisation All registered parties eligible Representation in regional and/or local assemblies Min. 2.5 per cent of vote Minimum 2 per cent of vote + representation in 2nd Chamber Representation in regional and local government, respectively 2.5% of votes or parliamentary Publish income items Voluntary agreement among parties to publish accounts United States 1974 Candidates Central Party organisation Both major parties eligible Campaign expenses Nominations; Party Conventions Candidate accounts audited by Federal Election Commission Notes: * A separate state agency has been established to receive the subsidy. ** Party subsidies were abolished in Italy in *** The Netherlands also has a subsidy to facilitate radio and TV broadcasts by the parties. It was first introduced in Sources: Katz and Mair (eds.), Party Organizations; idem (eds.), How Parties Organize.

13 12 WEST EUROPEAN POLITICS Accountability An increasing number of countries require parties and candidates to make information about their finances available to the public, and also to show that public subsidies have been spent according to the rules. In practice, the rules vary considerably among different countries, from leaving parties and candidates completely autonomous to very strict control with regard to how the subsidies are spent. The most extensive monitoring is exercised in the United States and Germany. In Norway, on the other hand, there were until 1998 no requirements concerning the availability of finance information; indeed, when subsidies were first introduced, discretion on how the subsidies were spent was seen as a key precondition for the introduction of the subsidies in order to ensure that parties would not become bound in any way by the state. 25 In Sweden, there is a voluntary agreement among the parties from 1981 to make their accounts publicly available, while the Austrian, Danish and Finnish parties are required by law to provide their financial records. However, in these countries there is no regulation of how parties spend their money, or of how they raise it. Part of the problem of ascertaining the parties' use of the subsidies is that this money is funnelled into the party funds. Hence, it becomes difficult to isolate what specific expenditures were covered by the subsidies. Thus when asked if he could give an account of how the subsidies had been spent, a Danish party official claimed that 'it is like mixing hot and cold water and then try[ing] to separate them again'. 26 Thus, the public subsidy systems vary across countries as well as over time within a country. In addition, we have seen that the parties benefit to a varying degree from other forms of state subsidies, for example to the press or different programmes for youth activities. Therefore, it is almost impossible to calculate how much money the state (let alone regional and local governments) spends on political parties. But even with these caveats borne in mind it is clear that the Scandinavian countries, particularly Sweden and Norway, have very generous subsidy programmes. Here, the state spends more than twice the amount of, for example, the Netherlands per parliamentary seat in direct party subsidies. Table 2 summarises the discussion on the regulatory framework of the subsidy systems. THE IMPACT OF PUBLIC SUBSIDIES Although there are now few countries without some form of public subsidy to political parties, assessing the precise effects of party subsidies is not an easy task. The methodological problems are, firstly, the comparative complexity; given the few cases without any type of subsidy we are prevented from comparing party development in countries with and without

14 STATE SUBSIDIES TO POLITICAL PARTIES 13 subsidies. Also, the regulatory framework pertaining to the subsidies within each country have been revised several times, which makes comparisons difficult both synchronically and diachronically. Secondly, there is the problem of isolating the effects of subsidies from the myriad of other factors which shape party development. Party subsidies have been said to affect the party system, the internal life of parties, the electorate's view of parties, and the saliency of parties vis-à-vis other political actors. Needless to say, merely observing changes in parties and party systems, which correlate chronologically with the introduction of public subsidies, does not suffice to show that public subsidies caused that particular change. Testing the 'null hypothesis' - that is, that subsidies did not have any significant impact on the parties or the party system - faces a similar problem; negative effects of the subsidies could have been counterweighted by forces pulling in the opposite direction. There are also significant empirical problems in assessing the effects of party subsidies. Party finance is not very accessible or transparent; hence it is not self-evident what is a good benchmark for assessing the impact of public subsidies. Party finance data are only partly publicly available, depending on official rules and the parties' willingness to make their records available. Even when party finance records are available, there may still be serious doubts concerning their validity. Public subsidies are usually allocated over several budgetary categories, only one of which may actually be called 'party subsidies'. State subsidies have been said to have had effects both on the microlevel, that is, on intra-party developments in different respects, as well as on the macro-level, that is, the level of the party system. At the macro-level, there is some evidence of decreasing party activity in some countries, 27 although it is problematic to substantiate that there has been an overall, significant decline in intra-party participation and involvement over the past couple of decades. And even if such a decline has occurred, it is very hard to claim that the decline is on account of the introduction of state subsidies for parties. State-Dependency As we discussed earlier, one critical view on the subsidies holds that they have made parties more dependent on the state. The extent to which this should be the case is related not just to the 'generosity' of the state, but also with the traditional sources of income for different individual parties. Table 3 presents the significance of subsidies in these respects. It shows very clearly the tremendous cross-national variation with regard to the state's financial contribution to parties. This should not be very surprising given the different regulatory frameworks for the subsidy systems in different

15 14 WEST EUROPEAN POLITICS jurisdictions, discussed earlier. Finland has the highest level of 'statedependency', followed by Germany, Sweden and Norway. More importantly, the data do not display the pattern of continuously increasing financial dependency on the state predicted by the critics of the subsidies. Thus, contrary to much of the critique of party subsidies, parties have not become more state-dependent over time. The only slight exception to this pattern is Italy, where state-dependency increased continuously between 1986 and Finland generally has the highest level of statedependency, but even here the total state subsidy share of the parties' budgets in the first five-year period has since declined to about threequarters of total party income. Denmark represents the other extreme in these respects; subsidies to party organisations were introduced late and do not make up more than a fifth of the parties' income. 29 Throughout the Western democracies, parties have developed a wide range of different organisational models, varying significantly in terms of membership structure and linkages to society. 30 Thus, party structures and TABLE 3 PERCENTAGE STATE SUBSIDY OF TOTAL PARTY INCOME PER YEAR BY COUNTRY Year Sweden Germany Norway Italy Austria Finland* n/a n/a n/a _ _ _ Note: * Finnish data are five-year averages. Source: Adapted from Katz and Mair (eds.), Party Organizations.

16 STATE SUBSIDIES TO POLITICAL PARTIES 15 TABLE 4 PERCENTAGE PUBLIC SUBSIDY OF TOTAL PARTY INCOME IN SELECTED PARTIES (CENTRAL ORGANISATIONS, 1989) Denmark Germany Austria Norway* Sweden Social Soc. SPD Greens Social Greens Labour Soc. Social Greens Dem. People's Dem. Left Dem Note: * Norway: Source: Entries are calculated on the basis of Katz and Mair (eds.), Party Organizations. party history in and of themselves are likely to yield variations in statedependency also within a particular country. Parties like the Social Democratic parties in Sweden and Austria have traditionally relied on large membership organisations and close links to interest organisations. The impact of state finance will be different on these parties compared with parties with a weaker membership structure, such as newly formed parties. In most countries, more recent party formations have often been quite successful electorally but have found it more difficult to recruit members. Cases in point are the Socialist Left party in Norway, the Greens in Germany and the many new parties formed in Denmark in connection with the 1973 'political earthquake'. For these parties, therefore, public funding is more important than for the established parties. However, as Table 4 shows, there exists no clear-cut, universal pattern with regard to differences between old and new parties with regard to their statedependency. In Table 4, we use Social Democratic parties as the archetype of the mass party model in order to have a basis for comparison with selected new parties in Scandinavia, Germany and Austria. In Austria, Denmark and Germany, the new parties are financially more dependent on the state compared with the Social Democratic parties. In Norway and Sweden we see the opposite pattern; new parties have a smaller percentage of public subsidies in their budgets. Other new parties, such as the Progress parties in Denmark and Norway and New Democracy in Sweden have tended to be even more dependent on state subsidies. The often 'over-night' electoral successes of newly formed parties demonstrate that access to the media, not an elaborate grass-roots organisation, is the key precondition for becoming a party in parliament. While traditional parties first built an organisation and then became successful in the electoral struggle, new parties frequently proceed the other way around; electoral success brings public subsidies, which in turn may be used to build a regular party organisation.

17 16 WEST EUROPEAN POLITICS In sum, we cannot detect any clear patterns concerning which type of party tends to become most dependent on public subsidies. What does seem clear is that subsidies frequently play a fundamental role in the consolidation of new parties. This is obviously not the same as saying that they entail pétrification of the party system - an issue we shall address below - but the absence of any clear correlation between party size and state-dependency does call into question some of the criticisms of state subsidies to political parties. Public Funding and Party Change The implications of public subsidies on the parties' overall dynamic and viability are empirically elusive. While there may be correlations between state subsidies and different aspects of party change such as party decline, it is a different matter to claim that there is a causal relationship. Critics of public subsidies to political parties submit that they have a detrimental effect on party membership because subsidies would eliminate all incentives for retaining a high membership level. Also, the financial contribution of the rank-and-file membership would become largely insignificant; hence members leaving the organisation would go unnoticed by the senior ranks of the party organisation. This, in turn, is said to entail increased autonomy for the party leadership, and party membership will decline. Some see a clear, causal relationship between the emergence of public subsidies and declining party membership; 31 the introduction of public subsidies followed a period of substantial decline in membership - and a subsequent decline in party income - for the major parties. Subsidies, in this perspective, were a response or a strategic adaptation of the party to identify and exploit new sources of income when traditional sources could no longer sustain the organisations; a strategy not very different from what Tsebelis refers to as an attempt to 'modify the rules of the game'. 32 Seen this way, the introduction of public subsidies would be a case of political parties exercising the prerogatives they derive from their privileged position as agents of civil society, but simultaneously deeply rooted in the state apparatus. What appears to be the prevailing perspective on the subsidies is that they represent an ingenious invention of the party élites to guarantee the party organisations a significant, safe and continuous source of income in an era of increasing campaign and other expenditures and increasingly volatile voters and members. However, this interpretation ignores the fact that parties continue to implement membership mobilisation campaigns, and express deep concerns over the lack of grass-root involvement in party activity In order to generate some understanding of the significance of the subsidies with regard to party development and change we focus on the

18 STATE SUBSIDIES TO POLITICAL PARTIES 17 development of party membership over time and use the year for the introduction of party subsidies as a cut-off point to see to what extent subsidies have had an effect on party membership. That having been said, however, a plausible effect of public subsidies could have been an exacerbation of the decline since parties came to rely less on membership dues once the subsidies were introduced. 33 Furthermore, while it is true that party membership has declined in most countries, simultaneously with the expansion of the public subsidy schemes, it is unclear to what extent these two developments are related to each other. It is certainly worth noticing here that in the British case, where extensive state subsidies for parties have not been introduced, party membership has nevertheless declined significantly since the 1960s. 34 In Figure 1 we have indexed party membership by party family. 35 The underlying assumption here is that party membership is accorded different degrees of significance in different types of parties and also that differences in their process of formation and historical trajectory is an important factor in assessing the impact of state subsidies. Figure 1 suggests that there is little to suggest a systematic membership development surrounding the introduction of party subsidies if we look at party families. In the case of Socialist/Social Democratic parties, party membership fluctuated slightly in the years preceding public subsidies, but increased consistently in the years following their introduction. Thus, at least in terms of timing, there is a correlation between the two phenomena. For the Conservative parties, there is a decline in membership starting in advance of, and continuing after, the FIGURE 1 INDEX OF PARTY MEMBERSHIP DEVELOPMENT BEFORE AND AFTER INTRODUCTION OF STATE SUBSIDIES, BY PARTY FAMILY Chr. Dem. o Liberal X-Socialist Conserv I 1 1 1I Year

19 18 WEST EUROPEAN POLITICS introduction of the subsidies. The same can be said of the Liberal parties. Christian Democratic parties display a curvilinear development in the years leading up to the start of the public subsidies, but a strong increase in the index after the subsidies had started to be paid. Thus, there is no real support for the argument that public subsidies interfered in the organic development of the parties by reversing a decline in membership. Indeed, the patterns displayed in Figure 1 indicate a very limited impact of the subsidies; party families which had experienced a declining membership prior to the introduction of public subsidies largely continued to do so once they were implemented, and parties with increasing membership also continued to increase after the subsidy programmes had been enacted. A part of the explanation of the absence of clear correlations between membership development and the introduction of public subsidies is that parties recruited members for other reasons than collecting dues before the subsidies were established and they have continued to do so after the introduction of the subsidies as well. Members continue to be important to the parties for a number of reasons other than financial motives. As Scarrow has pointed out, members are believed to be essential to the parties; for example, as providers of legitimacy, communicative channels between the party and society, financial benefits, free labour, ideas, and as a recruitment base for senior party posts. 36 How can we account for the timing of the introduction of public subsidies and for its specific arrangements? We noticed earlier that most countries introduced some form of public subsidy in the late 1960s and early 1970s. The lateness of Belgium and Denmark in taking the final step towards subsidising party organisations may be only an appearance. In reality the rules opened up for this earlier as the parliamentary groups were largely free to spend the money as they pleased. Given the differences in timing and format of public subsidies across countries, it appears unlikely that one single factor can explain why the state took on the burden of financing the parties. We earlier pointed to the arguments in favour of public subsidies. These arguments could easily be found in all states. However, this 'state-centric' explanation must be supplemented by a 'diffusion' perspective. 37 Throughout Europe there has been a diffusion of public subsidies for political parties, originating in (West) Germany and Sweden. Once the subsidy was introduced in one country, it could be diffused across national boundaries. Information on party finance schemes may have been diffused through international party organisations or international co-operation organisations like the Interparliamentary Union and the Council of Europe, which provide arenas for exchange of information. 38 In recent years, reforms of state party subsidies have shown a similar 'diffusion pattern'. Germany

20 STATE SUBSIDIES TO POLITICAL PARTIES 19 and France (from 1991) have adopted the American and Canadian models of finance regulations by creating agencies similar to the Federal Elections Commission and by linking contributions from individuals to political parties and candidates to tax deduction. The claims made in 1967 by the Norwegian party leaders that the parties play an 'indispensable role', and consequently should be publicly subsidised, is perhaps better understood in the light of developments in other democracies. Sweden had introduced public subsidies for parties in 1966, and Norwegian parties - most of them have close contacts with their Swedish counterparts - had not failed to observe the effect. In addition to the 'diffusion theory' of the emergence of public subsidies in different national contexts, there is also another hypothesis which calls into question the idea that subsidies were proof of parties penetrating the state in pursuit of their own interests. The emergence of public subsidies to parties, we argue, was a path-dependent response to the parties' financial problems. That is to say, subsidies were introduced in countries with distinctly state-centric regimes and a strong societal belief in the state as a regulator, provider and mediator. These regimes have a long history in distributive and redistributive policies. Offering financial support to the political parties was perfectly consistent with offering support to the press, voluntaristic organisations, study groups and a large number of other functions and activities in society. Thus, state subsidies to political parties were seen as an integrated part of the state's responsibilities and societal roles. Put slightly differenly, the solution - subsidies - is a path-dependent choice of policy instrument, applied to a variety of problems, which the state wishes to address. Theories of path-dependency in public policy seem to account for the emergence of public subsidies to the political parties much more than the theory that the parties penetrated the state to promote their own interests and secure massive funding support from sources there. 39 If this model of the emergence of public subsidies to the party organisations is correct, then we might also want to address its consequences to the discussion of the changing interaction between civil society and the state with the political parties as intermediary agents. According to the 'cartel theory' of the parties' role in state-society interaction and exchange, 40 parties are assumed to have successfully penetrated the state apparatus while at the same time becoming gradually detached from civil society and the linkage roles, which the traditional model of party government accords them. Indeed, the introduction of public subsidies to the parties has been taken as proof - or at least an important indicator - of this process. However, if public subsidies were introduced to a significant extent not because the parties experienced a serious financial crisis but rather because of a general expansion of the

21 20 WEST EUROPEAN POLITICS public domain in the Scandinavian societies of the 1960s, the process could be described more accurately as a co-optation of the parties by the state than a penetration of the state by the parties. Thus, in the language of the cartel theory, the development did not primarily see the parties moving closer to the state but - at least to the same extent - the state moving closer to the established parties. State Subsidies and Party System Pétrification Have public subsidies preserved the existing party system? Have they helped build walls around the party system in order to keep new parties off the political scene, as some critics suggest? A closer look at the party system change in western Europe shows that party system dynamics appear to have no relationship to the emergence of public subsidies, at least not in the expected direction. If there were any relationship between public subsidies and constrained dynamic and change in the party systems we should expect that countries in which subsidies were introduced early and generously - would display only moderate change, whereas the opposite would be true for countries without subsidies. As Table 1 shows, the countries we have included in this study fall into three categories. The first category consists of Germany, Sweden, Norway, Finland, Austria and Italy, which introduced fairly generous subsidies to extra-parliamentary party organisations in the late 1960s and early 1970s. The Netherlands also introduced subsidies early, but in a limited form. The second category comprises Belgium and Denmark, which both introduced subsidies fairly late. The third category consists of Ireland and Britain as countries without party subsidies. Among the early and generously subsidised party systems, Germany, Austria and Sweden experienced little change in the existing party systems until the late 1980s when Green parties secured parliamentary representation. This might indicate some support for the pétrification hypotheses, but Norway experienced an upheaval in its party system following the first EU membership fight in 1972, with the successful formation of two new parties and sustained changes in the balance between the old parties. Looking at the two other categories of countries, it appears there is no relationship between stability and subsidies. Denmark experienced a political earthquake in 1973 with lasting consequences for the structure of the party system, 41 whereas in Belgium the systemic changes have been much more moderate. Moreover, in Ireland and in the UK, where the conditions for new parties - according to the pétrification hypotheses - should be the most favourable, there has been very little change indeed in terms of successful new parties. It appears that the pétrification hypothesis places an enormous weight on a technicality (state subsidies), ignoring other

22 STATE SUBSIDIES TO POLITICAL PARTIES 21 institutional factors, such as the electoral system, the new media structure; not to speak of changes in the social structure, issue orientation in the electorate and new social and political values. These patterns, to reiterate an earlier remark, do not provide complete and exhaustive insights into the dynamics of party change or party system change; the purpose here is strictly to review the critique against public subsidies to the parties and the degree to which actual developments corroborate the critique. In doing so, the modest empirical test we have undertaken here suggests that the introduction of public subsidies to the parties in and of itself cannot delay the emergence of new parties. CONCLUSIONS Have subsidies set in train the negative developments the critics argued they would lead to? There is very little in our data which suggest that that should be the case. Pétrification of the party system has not taken place in systems with extensive support for parties as organisations. Ironically, two countries in which the 'old' party system has retained its strongest grip on political representation, Britain and the United States, are also systems in which parties either are not supported at all, or, as in the case of the United States, support is directed towards candidates, not to parties. Membership levels have fallen in parties in political systems across Europe with apparently little relationship to public subsidy schemes. And centralisation in political parties is a process which appears to be about as old as the parties themselves. 42 In fact, direct subsidies to particular organisational sub-units may have contributed to organisational fragmentation, as the central party organisation no longer has complete control of the inter-organisational flow of the party's financial resources. To the extent that it is possible to substantiate claims of organisational fragmentation, public subsidies have had the opposite effect of centralisation, which otherwise may have resulted from this type of financing. Taken together, these observations suggest that public subsidies, in and of themselves, have not had a significant impact on the structure of the party system, nor have they - to the extent that this was ever the intention - reversed the process of party membership decline. This finding corroborates Sartori's succinct statement that 'party finances do not explain why party systems are as they are, nor their variance across the world'. 43 To be sure, the data presented here indicate that public subsidies can make a positive more than a negative difference. Put differently, in countries like (West) Germany where the subsidies were introduced in an era of party organisational consolidation or even expansion, these subsidies may have provided the organisations with the funds necessary to continue this

23 22 WEST EUROPEAN POLITICS development further. However, in countries where the parties were experiencing decreasing membership figures, public subsidies were not able to convert this development into a positive one, even if that had been their purpose. It is interesting to note that public subsidies - coupled with long-term, significant changes in electoral campaigning - can even be said to encourage the creation of new political parties rather than obstructing it. Parties campaigning via the media instead of, as was previously the case, with the aid of large numbers of rank-and-file campaign activists can gain entrance to parliament and become eligible for public subsidies. A number of successful newer parties in the Scandinavian countries, like the Danish and Norwegian Progress parties and the Ny Demokrati in Sweden in the 1991 election, display, this strategy. Finally, the introduction of public subsidies did not mean that other sources of income lost all significance. If we look at the percentage of party income deriving from other sources than public subsidies we see that, after the initial major decrease, these sources soon regained a significant percentage of total party income. Thus, the evidence presented here seems to raise questions about the validity of at least some of the critique voiced against public subsidies. Party development and party system change are complex and multi-dimensional processes, which are related to overarching changes in society. Public subsidies may well play a role in that process but they are too weak to make a decisive difference in any crucial aspect of party development. NOTES 1. See Charles S. Maier (ed.), Changing Boundaries of the Political (Cambridge, MA: Cambridge University Press 1987). 2. Richard S. Katz and Peter Mair, 'Changing Models of Party Organization: The Emergence of the Cartel Party', Party Politics 1/1 (1995), pp.5-28; Eric Nordlinger, On the Autonomy of the Democratic State (Cambridge, MA: Harvard University Press 1981). 3. Gullan Gidlund, Partistod (Lund: CWK Gleerup 1983). 4. Arthus Lipow, Political Parties and Democracy: Explorations in History and Theory (Buckingham: Polity Press 1996), Chapter Karl Heinz Nassmacher, 'Party Finance in Western Democracies: A Framework for Analysis' (paper presented at the conference 'Party Politics in the year 2000', UMIST, Manchester, Jan. 1995). 6. See, e.g., Katz and Mair, 'Changing Models of Party Organization'. 7. Mogens N. Pedersen, 'The Birth, Life and Death of Small Parties in Denmark', in Ferdinand Müller-Rommel and Geoffrey Pridham (eds.), Small Parties in Western Europe: Comparative and National Perspectives (London: Sage 1991), pp Albert O. Hirschman, Exit, Voice and Loyalty. Responses to Decline in Firms, Organizations and States (Cambridge, MA: Harvard UP 1970). 9. It appears that the initial opposition of the Norwegian Conservative Party to state subsidies for party organisations was not motivated by its own financial strength. On the contrary, Sejersted has shown that the central office of the Conservative Party was constantly in financial difficulties. However, state funding for parties as such was seen as blurring the

24 STATE SUBSIDIES TO POLITICAL PARTIES 23 distinction between the state and the voluntary associations of society. Albinsson has shown that the Swedish Conservative Party was initially of the same opinion. Francis Sejersted, Opposisjon og posisjon ( ). H0yres historie, 3 (Oslo: Cappelen 1984); Per Albinsson, Skiftningar I Blätt. Förändringar I Moderata Samlingspartiets Riksorganisation (Lund: Kommunfakta Förlag 1986). 10. Anthony King, 'Political Parties in Western Democracies', Polity 2/2 (1969), pp Sidney Verba and Steven Kelman, Elites and the Idea of Equality: A Comparison of Japan, Sweden, and the United States (Cambridge, MA: Harvard University Press 1987); Sidney Verba, Norman H. Nie and Jae-on Kim, Participation and Political Equality: A Seven Nation Comparison (Chicago: The University of Chicago Press 1987); G.K. Wilson, Business and Politics: A Comparative Introduction (Chatham, NJ: Chatham House 1990). 12. Richard S. Katz and Peter Mair (eds.), Party Organizations: A Data Handbook (London: Sage 1992); Richard S. Katz and Peter Mair (eds.), How Parties Organize: Adaptation and Change in Party Organization in Western Democracies (London: Sage 1994). 13. Peter Burnell and Alan Ware (eds.), Funding Democratization (Manchester: Manchester University Press 1998). 14. See, e.g., Herbert E. Alexander, Comparative Political Finance in the 1980s (Cambridge, MA: Cambridge University Press 1989); A.T. Barbrook, 'Atlantic Crossing: Campaign Finance in Britain and the USA', Parliamentary Affairs MB (1993), pp ; Justin Fisher, 'Donations to Political Parties', Parliamentary Affairs 50/2 (1997), pp ; M. Hofnung, 'Public Financing, Party Membership and Internal Party Competition', European Journal of Political Research 29/1 (1996), pp.73-86; R.I. Johnston and C.J. Pattie, 'The Impact of Spending on Party Constituency Campaigning in Recent British General Elections', Party Politics 1/2 (1995), pp ; Jonathan Mendilow, 'Public Party Funding and the Schemes of Mice and Men: The 1992 Elections in Israel', Party Politics 2/3 (1996), pp David M. Farrell, 'Ireland: Centralization, Professionalization and Competitive Pressures', in Katz and Mair (eds.), How Parties Organize, p The Norwegian Constitution of 1814 for example specified that MPs could travel free of charge and would receive allowances. 17. David Farrell, 'Ireland', in Katz and Mair (eds.), Party Organizations, p Kris Deschouwer, 'Belgium', in ibid., p Luciano Bardi and Leonardo Morlino, 'Italy', in ibid., pp.608ff. 20. Mogens N. Pedersen and Lars Bille, 'Public Financing and Public Control of Political Parties in Denmark', in Matti Wiberg (ed.), The Public Purse and Political Parties (Helsinki: The Finnish Political Science Association, 1991), p Wolfgang C. Müller, 'The Development of Austrian Party Organizations in the Post-War Period', in Katz and Mair (eds.), How Parties Organize, p.55; Gullan Gidlund, 'Public Investments in Swedish Democracy', in Wiberg (ed.), The Public Purse, pp.13-54; Ds. 1994: 31, Det offentliga stodet till partierna - Inriktning och omfattning (Stockholm: Fritzes 1994), p A number of political scandals in France, Belgium, Spain and Italy in during the early 1990s reveal extensive covert and illegal financial activities of parties discrediting official party data. Bardi and Morlini report that illegal contributions to Italian parties were probably ten times the official party figures. Luciano Bardi and Leonardo Morlini, 'Italy: Tracing the Roots of the Great Transformation', in Katz and Mair (eds.), How Parties Organize, p.260, Italy abolished state subsidies to parties in 1994, following a referendum. At the time of writing, attempts to re-introduce party subventions have been unsuccessful. 23. This also now applies to France. 24. Herbert E. Alexander, 'American Presidential Elections since Public Funding, ', in Herbert E. Alexander (ed.), Comparative Political Finance in the 1990s (Cambridge, MA: Cambridge University Press 1989), pp ; Frank J. Sorauf, Inside Campaign Finance (New Haven, CT: Yale University Press 1992). 25. Since 1998 parties must make available information about the sources of their income. 26. Personal communication, Lars Bille, Department of Political Science, University of Copenhagen. 27. E.g., Denmark: Jörgen Goul Andersen, 'Politisk deltagelse i 1990 sammenlignet med 1979', in J.G. Andersen et al., Medborgerskab: Demokrati of politisk deltagelse (Herning: Forlaget Systime 1993), pp.45-74; and Sweden: Anders Widfeldt, Linking Parties with People? Party Membership in Sweden (Aldershot: Ashgate 1999).

25 24 WEST EUROPEAN POLITICS 28. See, however, note Koole reports that despite the introduction of some forms of state subsidies to the Dutch parties in the 1960s/1970s (see Table 1), more than half of the annual income for several parties comes from the members. Ruud Koole, 'The "Modesty" of Dutch Party Finance', in Alexander (ed.), Comparative Political Finance in the 1980s, pp ; Ruud Koole, 'The Vulnerability of the Modern Cadre Party in the Netherlands', in Katz and Mair (eds.), How Parties Organize, pp See, e.g., Katz and Mair, 'Cartel Party'; Richard S. Katz and Peter Mair, 'Cadre, Catch-All or Cartel? A Rejoinder', Party Politics 2/4 (1996), pp ; Thomas Poguntke, 'Parties and Society in Western Europe: Declining Linkage?' (paper presented at the conference 'Party Politics in the Year 2000, UMIST, Manchester, Jan. 1995); Ruud Koole, 'Cadre, Catch-All or Cartel? A Comment', Party Politics 2/4 (1996), pp Lars Bille, 'Denmark: The Decline of the Membership Party?', in Katz and Mair (eds.), How Parties Organize, p.145; Ruud Koole, 'The Vulnerability of the Modern Cadre Party in the Netherlands', in ibid., p George Tsebelis, Nested Games (Berkeley, CA: University of California Press 1990), p Richard S. Katz, 'Party as Linkage: A Vestigial Function?', European Journal of Political Research 18/1 (1990), pp Paul D. Webb, 'Are British Political Parties in Decline', Party Politics 1/3 (1995), p Comments to Figure 1: The vertical axis represent indexes on total party membership. Index=100 at the year of introduction of public subsidies. The indexes are averages of the respective index for different party families. The horizontal axis represents the years, where year 0 is the year of the introduction of state subsidies, negative years are years before the introduction and positive years after the introduction. For Denmark we have used the year of introduction of parliamentary subsidies, for other countries the year of introduction of subsidies to central party organisations. Included parties are as follows: Christian Democratic parties: Christian People's Party (Norway), DC (Italy), CDU/CSU (West Germany), ÖVP (Austria). Liberal parties: FPÖ (Austria), Venstre (Denmark), Folkpartiet (Sweden), PLI (Italy), and FDP (West Germany). Socialist parties: SPD (West Germany), Arbeiderpartiet (Norway), Socialdemokraterna (Sweden), Socialdemokratiet (Denmark), Suomen Sosialidemokrattiinen Puolue (Finland), PSI (Italy), and SPÖ (Austria). Conservative parties: H0yre (Norway), Konservative Folkeparti (Denmark), Moderata Samlingspartiet (Sweden), Kansallinen Kokoomus (Finland). Source: Data from Katz and Mair (eds.), Party Organizations, reanalysed by the present authors. We have also conducted a similar analysis comparing membership development in different national contexts. There had been a negative membership trend in Sweden prior to the introduction of party subsidies, but in Germany the trend was positive, and in other countries it is difficult to speak of any trend at all. The development after the subsidies had been introduced has either been stable or negative. The only exception is Germany, where the positive trend continued. Thus, on the country level, there is no general support for the hypothesis of party subsidies as a response to a membership decline; nor is there anything to suggest that the introduction of subsidies had a systematic and general effect on membership development. 36. Susan Scarrow, Parties and Their Members: Organizing for Victory in Britain and Germany (New York and Oxford: Oxford University Press 1996), pp Lauri Karvonen, Med vari vastra grannland somforebild (Turku: Åbo Akademi 1982). 38. See, e.g., Financing of Political Parties: A Cornerstone of Pluralist Democracies Report on the committee on parliamentary and public relations (Council of Europe 1989). 39. Kathleen Thelen and Sven Steinmo, 'Historical Institutionalism in Comparative Politics', in Sven Steinmo, Kathleen Thelen and Frank Longstreth (eds.), Structuring Politics: Historical Institutionalism in Comparative Analysis (Cambridge, MA: Cambridge University Press 1992), pp Katz and Mair, 'Cartel Party'. 41. Mogens N. Pedersen, 'The Danish Folketing Election of 1973: The Defeat of all Parties', in Kay Lawson and Peter H. Merkl (eds.), When Parties Fail (Princeton, NJ: Princeton University Press 1988), pp Robert Michels, Political Parties. A Sociological Study of the Oligarchical Tendencies of Modern Democracy (New York: The Free Press 1968 [1911]). 43. Giovanni Sartori, Parties and Party Systems. A Framework for Analysis, vol.1 (Cambridge, MA: Cambridge University Press 1976), p.95.