Populist right-wing parties (PRWPs) and the welfare state

One of the most important developments in West European party systems over the last thirty years has probably been the emergence and ‘mainstreaming’ of PRWPs (Mudde, Reference Mudde2007, Reference Mudde2013). Kriesi et al. (Reference Kriesi, Grande, Lachat, Dolezal, Bornschier and Frey2008: 19) consider these parties that champion immigration control, tough law and order policies and a critique of the political establishment as the driving force behind the current transformation of West European party systems. In countries such as Austria, Switzerland, Denmark, Italy, the Netherlands, or Norway, PRWPs are no longer a protest force only, but have become pivotal players in the building of parliamentary majorities (De Lange, Reference De Lange2012). While a now substantial body of literature has delved into the ideology of these parties (Mudde, Reference Mudde2007), the factors influencing their success (Bornschier, Reference Bornschier2010), their electorate (Ivarsflaten, Reference Ivarsflaten2005; Oesch, Reference Oesch2008) and their impact on other parties (Bale, Reference Bale2003; Van Spanje, Reference Van Spanje2010), few studies have sought to assess their influence on policymaking, and more specifically on welfare state reforms. The few works on their policy impact have focused on the domains that they are considered to ‘own’, such as immigration and law and order (Schain, Reference Schain2006; Akkerman, Reference Akkerman2012). In contrast, very little attention has been paid to the way they may enable change in other domains of government intervention, notably by changing coalition opportunities for other parties.

This article investigates the impact of PRWPs on welfare reforms in Western Europe with a comparative case study of pension policymaking in Switzerland, Austria, and the Netherlands. The main argument of the article is that PRWPs face a trade-off between office and votes when it comes to welfare state reforms, and that their policy impact depends on how they prioritize these two objectives. On the one hand, PRWPs appeal to a constituency of traditionally left-leaning blue-collar ‘insiders’ (Häusermann and Walter, Reference Häusermann and Walter2010: 144) generally supportive of social insurance schemes, such as public pensions. On the other hand, PRWPs have only been able – as far as Western Europe is concerned – to take part in government as junior coalition partners with liberal or conservative parties who are more likely to retrench these very same welfare programmes (Giger and Nelson, Reference Giger and Nelson2011). In this context, the article argues that parties have to choose between betraying their electorate (and losing votes), and betraying their coalition partners (and losing office). When they choose office, it enables welfare state retrenchment by allowing their coalition partners to curtail left-wing opposition, but entails high electoral costs for PRWPs. When they choose votes, it may jeopardize their participation in government by blocking the retrenchment plans of their coalition partners.

The article contributes both to the literature on PRWPs and on welfare state reforms. First, while research on PRWPs has shown that socio-economic factors do not figure at the centre of the electoral platforms of PRWPs or in the preferences of their voters, notably because they are keen to ‘blur’ their position to appeal to a wider range of voters (Rovny, Reference Rovny2013), this article argues that socio-economic policies can become issues of utmost importance once PRWPs are in office because their position on these domains becomes much more difficult to conceal. Second, recent scholarship on the welfare state is increasingly moving away from the idea that welfare states are ‘immovable objects’ and emphasizes how partisanship and party coalitions can enable change (Häusermann et al., Reference Häusermann, Picot and Geering2013). If one considers that party politics and coalitions are major factors in explaining welfare state reforms, the new coalition alternatives created by the rise of PRWPs can potentially influence policies in this domain as well.

The article is structured as follows. The first section outlines the dilemma between office and votes that PRWPs have to face, and proposes a number of hypotheses about their impact derived from party strategies. After outlining the methodology, the article tests the argument in a comparative case study analysis of Switzerland, Austria, and the Netherlands drawing on a variety of primary and secondary sources, including interviews. The article concludes with a discussion of findings and insights for further research.

Office, votes and policy: the dilemma of PRWPs and welfare state reforms

Since the mid-1980s, PRWPs have been the fastest growing party family in Western Europe. These parties combining nativism, populism, and authoritarianismFootnote ¹ have emerged as a significant electoral and parliamentary force in a number of countries (Mudde, Reference Mudde2007). From 9.3% in 1990, the average share of seats of these parties in the parliaments of Switzerland, Austria, Belgium, the Netherlands, Denmark, and Norway had jumped to 17% in 2013. As of 2013, their share of parliamentary seats had peaked at 27.8% in AustriaFootnote ² (2013), 30.1% in Switzerland (2007) and 16% in the Netherlands (2010). While they were first ostracized by other parties, PRWPs have progressively acquired a more respectable status, and have either taken part directly in government as coalition partners, or have struck agreements to provide support to minority governments (De Lange, Reference De Lange2012).

Although a substantial body of research has analysed the ideology and positions of PRWPs, few works have analysed their policy impact, and none so far has focused on welfare reforms. Schain (Reference Schain2006) looks at their direct and indirect impact on immigration policy, and Zaslove (Reference Zaslove2004) assesses how the ideologies of the FPÖ and the Lega Nord have influenced immigration policy reforms towards more restriction. Whereas these studies mostly assess how the ideas of these parties are taken up by ruling parties mostly for electoral purposes, they do not deal with how PRWPs directly negotiate policy within government, and how they trade off their preferences for policy, office, and votes (Strom, Reference Strom1990). Akkerman and De Lange (Reference Akkerman and De Lange2012) analyse the participation of PRWPs in government, but focus on their impact on immigration policy, assuming that this is the foremost issue that these parties care about. In many ways, this assumption is in line with the statement of Mudde (Reference Mudde2007) arguing that the core concern of these parties is nativism, socio-economic issues being at best subordinate.

While it is clear that immigration control and law and order have been the main ‘selling points’ of PRWPs, there are good reasons to believe that their policy influence is not restricted to these issues alone, especially once they come to office. First, the logic of policymaking is different from the logic of electoral politics. While it may be easier to conceal their position in electoral campaigns and focus on specific issues that they ‘own’, it becomes more difficult to do so once in office, when explicit parliamentary support has to be given or withdrawn to government proposals. If a PRWP takes part in a government coalition, it must explicitly give or withdraw its support to the policies of other parties in the coalition. Second, PRWPs may find it difficult to gain credibility if they are perceived as ‘one-issue’ parties. As one of the core issues of government intervention, it is reasonable to assume that welfare is one issue on which PRWPs will have to position themselves once in government.

So far, the role of PRWPs in welfare politics has only been marginally tackled in the literature. Kitschelt and McGann (Reference Kitschelt and McGann1995: 257–273) analysed how the welfare state may structure anti-immigration sentiments and the success of PRWPs, and Swank and Betz (Reference Swank and Betz2003) seek to assess if the welfare state tempers support for right-wing populist parties. As the participation of PRWPs in government has spread, we now have comparative evidence to assess the impact of PRWP on the welfare state, instead of the other way around. In this article, I argue that this role can be best understood in the light of the particularly high ‘costs of governing’ that these parties face (Van Spanje, Reference Van Spanje2011), due to the trade-off between the preferences of their electorate and those of their potential coalition partners. For this, I draw on Strom’s (Reference Strom1990) now classic theory of party goals highlighting the trade-offs between office, policy and votes. The main idea is that PRWPs cannot have them all, and their impact on welfare reforms depends on how they prioritize these goals.

Cases and methods

The empirical section of this paper draws on a comparative case study analysis of pension reforms in Austria, Switzerland, and the Netherlands during three periods of government participation of PRWPs: the Schüssel I and II cabinets in Austria (2000–06), the Rutte I cabinet in the Netherlands (2010–12) and three pension reforms in Switzerland: the 10th (1995), 11th (first attempt, 2003) and 11th (second attempt, 2010) revisions of state pensions (AHV/AVS). I use a most-similar systems design where a number of scope conditions (consensus democracy; Bismarckian welfare system; government participation of PRWPs) are kept constant, but where both the independent (party strategy) and the dependent variable (impact on welfare reform) are allowed to vary. There are a number of reasons why negotiation democracies and Bismarckian systems are appropriate cases. First, besides the fact that proportional representation makes it easier for new parties to make their way into parliament, negotiation democracies are also particularly vulnerable to populist critiques. Since decision making in these countries entails relatively opaque negotiations and compromise-building between parties and organized interests, populist appeals against a ‘cartel of elites’ have been a particularly fruitful strategy for radical right parties (Papadopoulos, Reference Papadopoulos2005; Hakhverdian and Koop, Reference Hakhverdian and Koop2007).

Second, Bismarckian welfare systems are characterized by an ‘insider-outsider’ divide which generates a potential for political mobilization along these lines (Swank and Betz, Reference Swank and Betz2003: 224). In this context, it makes sense to analyse the impact of PRWPs in consensus democracies, where they can actually take part in government, and Bismarckian welfare systems, where the insider/outsider divide structures welfare state reforms (Häusermann, Reference Häusermann2010). These should be considered as the scope conditions of the analysis. As the focus of the study is still exploratory, I do not produce explicit hypotheses on differences between countries. Variation can take place as much within countries (across time) as across countries. Pensions are analysed as a case in point because they represent the typical ‘traditional’ social insurance scheme which PRWP voters can be reasonably expected to be attached to. By contrast to beneficiaries of other social schemes, elderly people are uniformly seen as the most ‘deserving’ welfare recipients across European countries (Van Oorschot, Reference Van Oorschot2006). Agendas of reform are also fairly comparable across countries, with the issue of the age of retirement figuring prominently in the three countries analysed here. Vote-seeking strategies are defined as the expressed will to satisfy the preferences of the electorate in the domain of welfare, while office-seeking strategies translate into a will to accommodate coalition partners. I define retrenchment as net projected cuts in pension expenditure, essentially through increases in the retirement age.

The analysis draws on a variety of primary and secondary sources. First, I reconstructed reforms and party positions drawing on a media analysis of at least one quality newspaper in each country (the NRC Handelsblad in the Netherlands, Der Standard in Austria, and the Neue Zürcher Zeitung in Switzerland) (see online methodological appendix for details), as well as party documents available on the internet, and parliamentary debates. Second, I used material from 13 interviews conducted between 2007 and 2013 with actors from political parties, interest groups and ministries.Footnote ³ As noted by De Lange and Art (Reference De Lange and Art2011: 1230), access to party elites of PRWPs is fairly difficult for researchers; this was possible in the Swiss and Dutch case, but not in the Austrian case. As the participation in government of the Austrian FPÖ dates back further, this case analysis relies more on secondary literature. In each case study, I present evidence of the trade-off emphasized in the theoretical section.

Austria: from retrenchment to the ‘Social-Patriotic Party’ (2000–06)

In the mid-1990s, the Austrian grand coalition constituted by the SPÖ and the ÖVP proposed a programme of fiscal consolidation which involved a number of cuts in the pension system, for instance through actuarial reductions in early retirement. Jörg Haider’s FPÖ, then in the opposition, spearheaded criticism to this measure, denouncing it as ‘unsocial’ and ‘irresponsible’ (Schludi, Reference Schludi2005: 170). Eventually, a large part of savings measures were watered down due to the resistance of trade unions, who yielded significant influence within the SPÖ. Along similar lines, attempts to increase the early retirement age or increases in the minimum contribution time were blocked in the second half of the 1990s. In 1996, the government commissioned a report on pension reform which proposed a series of far-reaching measures to ensure the financial sustainability of the system, such as a change in the reference salary for the calculation of pensions (from the best 15 to the best 20 years), cuts for people retiring before the mandatory age, a new adjustment formula, and an extension of pension coverage (Schludi, Reference Schludi2005: 175). After vivid opposition from the unions again, only a substantially watered-down version of the reform was passed in 1997. By the late 1990s, it had become clear for the ÖVP that the kind of far-reaching reforms they wanted to pass could simply not be pushed through in a grand coalition with the SPÖ and their links with the unions. In this context, an alliance with the FPÖ could be an opportunity to curtail the power of unions. Even if the FPÖ had a strong working-class base and claimed to represent the ‘little man’, it also advocated lower taxes and deregulation in a way that could appeal to the ÖVP. In the run-up the 1999 elections, Haider increasingly sought to appear as a credible coalition partner for the ÖVP (Luther, Reference Luther2011: 458). Hence, he could potentially act as a support force for retrenchment for the ÖVP.

After the 1999 elections, the ÖVP negotiated a coalition agreement with the FPÖ, who had even outvoted the ÖVP by a few votes. The ÖVP–FPÖ coalition that entered office in 2000 adopted a resolutely more adversarial stance in the domain of welfare reform, and put cost containment at the forefront of its social policy agenda (Obinger and Talos, Reference Obinger and Talos2006). The coalition agreement present by the ÖVP and the FPÖ notably contained an increase of 1.5 years in the age of early retirement for men and women, an increase in the statutory age of retirement for public employees, increased penalties for people retiring sooner than the statutory age of 60 for women and 65 for men, and reforms of widows’ pensions which would remove the lower threshold for benefits (Regierungsprogramm, 2000: 23–24; Schludi, Reference Schludi2005: 178). For the FPÖ, an important point was to curtail the traditional system of corporatist concertation between the two big parties and the social partners, from which it had been completely excluded (Interview 7; Interview 8). The FPÖ notably tried to condition its support to the pension reform to a 40% reduction in the compulsory levy for the chamber of labour, a demand eventually refused by the ÖVP (Schludi, Reference Schludi2005: 180). Finally, an extension of entitlement to severance pay, with the option for all employees to convert it into an individual pension, was a first step towards a fully funded capitalized pension scheme (Schludi, Reference Schludi2005: 182).

While the coalition between the ÖVP and the FPÖ managed to overcome the power of trade unions and pursue a number of retrenchment reforms which would have been impossible in a grand coalition, the electoral impact of these reforms would be significant for the FPÖ. The party lost a string of regional elections after it came to power, which could be interpreted as the consequence of the series of retrenchment reforms it had supported and which hit directly its own electoral clientele of blue-collar workers, many of which returned to the social democrats (Heinisch, Reference Heinisch2003: 110). This caused significant internal strife which would eventually lead many ministers to resign, and the fall of the government. New elections were called in November 2002. These did not change the coalition in power (although its majority was reduced by seven seats), but did change the power configuration therein: the ÖVP jumped from 3rd to 1st place and increased its score by 15%. It gained 27 seats whereas the FPÖ lost 34 seats and two-thirds of its votes. Faced with a harsh defeat, the FPÖ sought to appeal to working-class voters by calling for a 1000 euro minimum wage and tax reductions for low-income earners (Muller et al., Reference Muller, Plasser and Ulram2004: 164).

In April 2003, another far-reaching pension reform was initiated by the cabinet. Retrenchment measures this time included the staged abolition of the early retirement scheme, increased financial penalties for each gap year in contributions, a lower conversion rate for each year of contribution, and a change in the mode of calculation whereby pensions would no longer be determined by the best 20 years of employment, but by the whole working career (Eironline, 2003). As could be expected, these measures yielded formidable protests from trade unions, who organized the biggest wave of strikes since the Second World War, but also within the FPÖ itself, which had become aware of the potential damage of retrenchment reforms for its own electoral standing (DerStandard.at, 2003b). In May 2003, Jörg Haider, who was no longer the party leader but still counted on a number of trustworthy MPs, threatened to withdraw support for the coalition if adaptations to the government pension plans were not made, notably to spare pensions below 1000 EUR (Der Standard, 2003b; Wirtschaftsblatt, 2003). In an interview with Der Standard, he notably argued that the series of electoral defeats that the party had undergone were due to the neoliberal policies pushed by the Minister of Finance from his own party, Karl-Heinz Grasser, and his ‘zero-deficit fetishism’. Retrenchment measures such as the taxation of accident benefits or the pension reform, Haider argued, were too much for the ‘little people’ who constituted the party’s electoral base (DerStandard.at, 2003a). The FPÖ Minister for social affairs, and Vice-Chancellor, Herbert Haupt, also proposed to hold a popular referendum on the reform, which was a way to undermine it in the light of its strong unpopularity (Der Standard, 2003c). Haider even started negotiations with the SPÖ leader Alfred Gusenbauer to block the reform (Der Standard, 2003a). Eventually, the reform went though with a number of slight adaptations and was supported by both parties (Schludi, Reference Schludi2005: 188). However, during this process, the FPÖ used a number of blockade strategies to temper it, MPs in both the lower and upper chamber threatening to block the reform if a softening of cuts was not achieved (Luther, Reference Luther2011: 464–465).

After that, Haider advocated the possibility for workers in strenuous jobs to retire before 60 (Der Standard, 2004). In 2005, internal conflicts with the FPÖ led to the creation of the BZÖ, whose members stayed in government while the remaining FPÖ, now led by H.-C. Strache, joined the opposition. Haider, now leader of the BZÖ, advocated an increase in pensions in spite of the harsh cuts that his party had supported until then (Der Standard, 2006). In the light of the internal struggles which had taken place within its coalition partner and its new opposition against retrenchment, the ÖVP ruled out another coalition with the BZÖ or FPÖ, and chose to engage in a new grand coalition with the social democrats instead after the 2006 elections. Since then, the FPÖ has reoriented its agenda towards a more interventionist agenda ‘targeted at blue-collar workers and welfare recipients’ in terms of welfare policies, along welfare chauvinist lines (Luther, Reference Luther2009: 1052), re-labelling itself the ‘social-national party’ (Soziale Heimatpartei). It notably advocated policies such as payments to compensate pensioners for inflation or reducing VAT on fuel, medication and food (FPÖ, 2008: 7–8).

The Netherlands: defending ‘Henk and Ingrid’ (2010–12)

In October 2009, the Dutch CDA/PvdA/ChristenUnie cabinet led by Jan-Pieter Balkenende announced that it had reached an agreement about an increase in the retirement age from 65 to 66 in 2020, and then to 67 in 2025 (NRC Handelsblad, 2009c). Meanwhile, the right-wing populist PVV was now perceived as a possible government party in the light of its very good score in the 2009 European parliament elections (17%). A poll conducted in 2010 showed that 86% of PVV voters were opposed to an increase of the retirement age to 67, while 41% of CDA and 43% of VVD voters supported it. In an interview with the Telegraaf, the PVV leader Geert Wilders described this measure as ‘asocial’, and retiring at 65 was a ‘non-negotiable’ issue for the PVV should the party take part in future coalition talks (De Telegraaf, 2009). Eventually, the increase in the age of retirement proved to be a much disputed issue even within government parties, and especially the PvdA. As two-thirds of its voters were unconvinced, some backbenchers followed the lead of trade unions in opposing the reform (NRC Handelsblad, 2009b). Internal controversies led to a standstill.

In February 2010, the CDA-PvdA-ChristenUnie cabinet fell due to disagreements on the Dutch engagement in Afghanistan, triggering new elections to be held in June and blocking pension reforms further. As the prospect of its participation in government was reinforced by very favourable polls, Wilders showed a renewed readiness to cooperate with other parties in a number of domains, and sought to demonstrate that the PVV could be a credible coalition partner. While presenting its electoral programme for the Reference Nai, Lloren and Gavilanes2010 elections, Wilders declared that the party was ready to support a minority right-wing cabinet should the PVV not be able to agree on a government programme, and possibly make concessions on a number of policy issues (NRC Handelsblad, 2010e, h). However, the party stated explicitly in its electoral programme that only one issue remained a ‘breaking point’ in eventual cabinet talks: the preservation of the retirement age at 65 (PVV, 2010: 21). Interview partners pointed out that poll results, and the increasing perception that the PVV was competing for votes with the radical-left Socialistische Partij, were the primary driver of this stance (Interview 12; Interview 13). Wilders himself, as a former VVD member, was more of a classical-liberal orientation, and the position of the PVV on welfare was clearly part of a vote-seeking strategy. In the party programme, it showed a resolute ambition to defend the main features of the Dutch welfare state, potentially threatened by mass immigration. Hence, it aimed to change a situation where ‘Henk and Ingrid pay for Ali and Fatima’ (PVV, 2010: 5) by restricting the payment of social transfers, especially child benefits, to people living abroad (PVV, 2010: 23). Besides, it opposed the flexibilization of employment protection and the retrenchment of unemployment benefits (PVV, 2010: 21).

Mark Rutte’s Liberal VVD was the rising force in the run-up to the elections. The Christian-Democratic CDA, who had led the government under the three previous cabinets, was substantially weakened. The question of fiscal consolidation was one of the leading issues of the electoral campaign, and the VVD strongly profiled itself as the champion of deficit reduction, with an electoral programme titled ‘putting the business in order’ (VVD, 2010). Its socio-economic programme was in sharp contrast with that of the PVV. In particular, the spending cuts provided for in the VVD programme were about ten times bigger than those of the PVV, a significant part of which being accounted for by the increase in the age of retirement (CPB, 2010: 23). In spite of these divergences, it also advocated a tightening in immigration policy which could be reconciled with the PVV agenda (VVD, 2010: 6).

At the elections held in June 2010, the VVD came first with 31 seats in the lower chamber, one seat ahead of Job Cohen’s PvdA. The CDA lost 20 seats, while the PVV gained 15, coming third with the best score in its history. As the leader of the first party, Mark Rutte was charged with forming a new government. The day after its electoral triumph, Wilders interestingly announced that the PVV’s only ‘breaking point’ in coalition negotiations was no longer a breaking point, and that he would be ready to abandon retirement at 65 if this was a condition for the participation of the PVV in government (NRC Handelsblad, 2010a). While a number of coalition options with other parties were explored, these were ruled out because of major disagreements about the extent of budget cuts. While the direct participation of the PVV in government was first considered possible, internal opposition within the CDA, the other likely coalition partner, made this solution unviable. Wilders himself was sceptical about leaving the opposition, but there was strong internal pressure to deliver on electoral promises (Interview 12). A particularly tricky issue within the negotiations was the wide differences between the extent of budget cuts that the VVD and the CDA wanted to push through in relation to the PVV agenda. The PVV had based its socio-economic programme on the defence of acquired rights, and its socio-economic agenda was not far apart from the SP. In this context, some commentators argued that the PVV was the only ‘hope for the left’ in dampening the austerity drive pushed by the VVD (NRC Handelsblad, 2010d). After a number of long and protracted negotiation rounds, however, it became clear that the PVV would have to ‘throw its socio-economic agenda overboard’ to take part in government (NRC Handelsblad, 2010b). As pointed out by a member of the PVV coalition negotiation team (Interview 13), the PVV did not seek large concessions in the area of welfare, partly because welfare was not its most important issue, and partly because Wilders himself did not support welfare protection in the first place. It was essentially an electoral strategy (Interview 12; Interview 13). However, it did not commit to support an increase in the age of retirement in the light of its previous electoral promises.

Eventually an agreement was found between the CDA and the VVD about a government programme, and a ‘support programme’ (gedoogakkoord) was agreed between the PVV and the other parties. As a whole, this was considered a fragile arrangement from the start. Together, the VVD, the CDA, and the PVV commanded the minimum possible majority in the lower chamber, with 76 seats, and only a minority in the Eerste Kamer, where the agreement of other parties was necessary. The support programme included a series of detailed measures about immigration, public safety, health care for the elderly, and a short general statement about budget balance, but with no explicit commitment on the part of the PVV towards socio-economic policy (Rijksoverheid, 2010a). The age of retirement only figured in the coalition agreement between the VVD and the CDA, whereby it would be raised not to 67 but to 66 years in 2020, following the guidelines of an agreement that social partners had found in the run-up to the elections (Rijksoverheid, 2010b: 43). Thereafter, the government committed to elaborate a solution to tie the age of retirement to life expectancy. The amount of basic pension benefits was to be linked to real wages and no longer to collectively labour agreements, resulting in a real increase in pensions for low wages (NRC Handelsblad, 2010c). In the light of the promises made by Wilders about this non-negotiable issue, this was a concession made to the PVV, since both the CDA and the VVD supported the increase to 67 (NRC Handelsblad, 2010f).

The tensions between the cabinet and the PVV about pensions emerged soon after. The PVV announced that it did not see the need for a reform of the pension system beyond 66, while the VVD Minister for Social Affairs argued that the further adaptation of the system, notably with an increase of the age of retirement to 67, was unavoidable. In the Spring of 2011, the cabinet submitted a law proposal to increase the retirement age to 66 by 2020 while social partners were again negotiating on this issue. It announced that it would seek support from other parties if the PVV stuck to its position (NRC Handelsblad, 2010g). In June 2011, the cabinet signed an agreement (pensioenakkoord) over the increase of the pension age with the leadership of trade unions and employers. The agreement was disputed within the trade unions, and the PVV heavily criticized it (NRC Handelsblad, 2011b). Despite forcefully criticizing the agreement as based on ‘ignorance and wishful thinking’, the PVV was challenged by the SP to withdraw its support for the government if the retirement age was increased, but it did not (Debat Tweede Kamer, 2011: min. 67). In this context, the cabinet started negotiating support with the main opposition party, the PvdA, with some concessions for early retirement and low incomes (NRC Handelsblad, 2011a). Negotiations were protracted, however.

In late 2011, the Dutch economic context worsened as the country officially entered a recession. The cabinet sought to pursue further budget cuts of about 14 billion EUR to compensate for the loss of revenues and conform to the EU 3% deficit rule. Accordingly, it initiated negotiations between the coalition parties and the PVV to agree on a package of austerity measures. Conflicts within the PVV erupted, and one MP left the parliamentary group, undermining the very tight majority that the PVV provided for the government. As the cabinet was determined to pursue budget cuts more quickly than anticipated, the idea to increase the age of retirement before 2020 was discussed in budget negotiations (NRC Handelsblad online, 2012). Wilders could not support this measure in the light of his electoral promises. As disagreements within the PVV could not be overcome, this eventually led Wilders to pull out of budget negotiations in April 2012, triggering the fall of the government (Interview 12). Wilders justified withdrawal from the negotiations by the severe cuts that pensioners would have to bear, arguing that he ‘could look his voters in the eye’ because ‘this [austerity] package is not in the interest of the PVV voter’ (NRC Handelsblad online, 2012). In the following elections, the PVV lost nine seats and went down to 5.3%, losing a third of its vote share. Eventually, the VVD and the CDA reached an agreement on the increase of the retirement age with smaller centre parties (ChristenUnie, D66, GroenLinks, and SGP) and with the opposition of the PVV, SP, and PvdA in June 2012. This agreement would be taken over by the following VVD–PvdA coalition after the 2012 elections.

Switzerland: having one’s cake and eating it

In June 1995, a reform of the pension system was accepted by 60.7% of Swiss voters (Vox, 1995). This 10th reform of the law on old-age pensions (AHV) provided for an increase in the age of retirement for women from 62 to 64 on the one hand, and a number of measures promoting gender equality on the other, notably the principle of ‘splitting’ whereby pension benefits would be split equally among couples so that one partner, usually women, were no longer dependent on their partner for their pension. As analysed by Häusermann et al. (Reference Häusermann, Mach and Papadopoulos2004: 44) this successful compromise was a result of a ‘social-liberal coalition’ between social democrats and modernizing sections within right-wing parties (the Liberal FDP and the Christian-Democratic CVP). The reform divided the social democrats and trade unions, but only the latter decided to challenge the law in a referendum (Le Nouveau Quotidien, 1994a). At first, the Swiss people’s Party opposed the principle of splitting (Le Nouveau Quotidien, 1993a), and championed the same age of retirement for men and women at 65 (Le Nouveau Quotidien, 1993b). In its 1993 party assembly, the President of the Swiss People’s Party had declared that it would oppose any further extension of the welfare system; welfare had ‘introduced a culture of assistance’ and the country needed to re-learn to save (Le Nouveau Quotidien, 1994b). Eventually, however, it supported the reform with the other parties on account of the fiscal consolidation measures it achieved (Nationalrat, 1994, 1961). At the time, the party was still in line with other right-wing parties, but the more hardliner stance championed by the Zurich-based leader Christoph Blocher – regarding immigration, the relationship with the European Union and fiscal retrenchment – were starting to emerge. Hence, in 2000, the party assembly approved the general principles of a far-reaching programme of reform providing for a phasing out of the pay-as-you-go system down to a basic level and the expansion of individualized personal pensions, coupled with an increase in the mandatory age of retirement to 68 (Le Temps, 2000a). This agenda represented a substantial shift in relation to the party’s previous positions in this domain, and was surely a risky strategy vis-à-vis its own clientele of pensioners. After these proposals were vividly criticized in the press, the party tried to temper them substantially (Le Temps, 2000b).

In the federal elections of 1999, the SVP increased its parliamentary representation in the lower chamber by 15 seats (out of 200), thereby becoming the largest parliamentary fraction after the social democrats. It outvoted the FDP by two seats for the first time. For centre-right parties, the SVP was emerging as a strong competitor. This not only involved competition for votes, but also competition for funding by firms and business groups, which play an important role in Swiss politics (Le Temps, 2001). In many ways, this led centre-right parties, and particularly the FDP, to move further to the right and assume a less accommodating stance towards the left and trade unions (Interview 4; 5; 9).

It is in this context that the 11th revision of the law on state pensions took place, and it would yield a clearly different outcome from the previous one. The initial project of the government provided for an equalization of the retirement age for men and women to 65, combined with the increase in VAT to fund pensions and the earmarking of 400 million Swiss francs for the flexibilization of early retirement, allowing for people on low incomes, and possibly in manual jobs, to retire earlier (Curia Vista, 2004). This was understood as a compromise which could rally left and right. Right wing parties, however, supported the increase in the age of retirement but categorically opposed the facilitation of early retirement, which they presented as yet another expansion of spending. They championed the idea that early retirement should imply linear cuts in benefits instead. Parliamentary negotiations were long and protracted, and extended until the 2003 elections, in which the CVP and especially the PRD lost a substantial share of their voters (respectively 7% and 14%) to the SVP (Selb and Lachat, Reference Selb and Lachat2004: 15). After the elections, the SVP managed to place Blocher as the second SVP member in the Federal Council in place of the Christian Democrats. The SVP and Blocher in particular were very opposed to both the increase in VAT and the facilitation of early retirement ‘for the rich or for the poor’ which would create further costs (Bulletin Officiel du Conseil National, 2003: 1513). The party displayed the most intransigent position vis-à-vis social adaptations to the reform (Interview 1; Interview 9). Eventually, the reform was accepted without the early retirement part with the support of all right-wing parties and opposed by the left and trade unions, which challenged it at referendum. In the referendum held in May 2004, the reform was massively rejected by voters, with only 32.1% accepting it. Interestingly, while 56% of FDP voters supported it, 59% of SVP voters refused the reform in spite of the fact that the party leadership had been the most determined to retrench (Engeli, Reference Engeli2004: 2). A similar pattern could be observed in a subsequent popular vote on a change in the conversion rate of occupational pensions in 2010, where the SVP championed retrenchment on the grounds of fiscal consolidation but only 27% of its own voters supported the change (Nai et al., Reference Nai, Lloren and Gavilanes2010: 2).

In the following 2007 elections, the SVP made substantial electoral gains, achieving 29% of votes. Advances were especially significant among low-income voters (Lutz, Reference Lutz2008: 13). After the SVP had become the most powerful parliamentary group in the lower chamber, a new attempt to reform pensions was initiated. Similar issues were on the table, namely the increase in the age of retirement to 65 for both sexes, the flexibilization of the age of retirement, and adjustments in the indexation modalities to inflation, which would be tied to the financial situation of the pension fund. As the government was conscious that a reform involving only cuts would stand little chance in a referendum, the facilitation of early retirement with smaller penalties for low incomes was supposed to buy off support from the left and trade unions to the increase in the age of retirement. While the left opposed the increase in the age of retirement, the SVP and the Liberals categorically opposed more spending to facilitate early retirement of low incomes.

After a protracted negotiation process where a fragile compromise combining the increase of the age of retirement with the creation of a fund to facilitate early retirement was found by centre-right parties, the project was eventually buried in the lower Chamber, where both the SVP and the Social Democrats voted it down. Social democrats denounced the weakness of compensation for the increase in the age of retirement, and the SVP officially denounced an ‘irresponsible’ extension of the welfare state through early retirement (Pressedienst SVP, 2010). As indicated by interview partners, however, the refusal of the reform by the SVP was also the result of concerns regarding its electoral consequences. Party elites were conscious that the electoral base rarely followed the positions of the party in the domain of welfare (Interview 10; Interview 11). As new elections were looming the following year, it was dangerous for the party to position itself once gain in favour of retrenchment and possibly to lose another referendum. The riskiness of welfare issues was clearly explained by a former MP:

In the 2011 federal elections, the SVP lost eight seats after its split from the PBD. However, it did not alter its position on welfare reform in a substantial manner.

Synthesis and conclusion

The empirical evidence presented here generally supports the argument that PRWPs face a trade-off between office and votes in the area of welfare. This does not mean that they are always electorally sanctioned for their support for retrenchment, as the Swiss case indicates, but internal tensions between the welfare preferences of the electorate and the welfare policies that these parties have to support or champion while in office are present in all three cases. In Austria, the FPÖ enabled a number of retrenchment reforms which hurt the core working-class electorate of the party (betraying voters) and led to internal splits and opposition to the retrenchment plans of its coalition partner. The participation of the FPÖ in government certainly facilitated retrenchment, as the ÖVP was able to push through reforms which had been impossible to implement with the SPÖ and its links with the unions, but it also led to heavy electoral losses for the FPÖ. The party then shifted to a more pro-welfare stance.

In the Netherlands, Geert Wilders’s PVV sought to find a balance between votes and office, first by presenting the retirement age as a ‘non-negotiable issue’, then arguing it was negotiable after all (betraying voters), and eventually refusing to sign off on retrenchment and pulling out of government (betraying its coalition partners). While the VVD in particular thought it could enjoy more leeway for retrenchment with the support of the PVV (as the ÖVP had done), these hopes were undermined by the vote-seeking strategy of the PVV which caused a policy deadlock on pension reform. In Switzerland, the main political tension has consisted in the strong pro-retrenchment stance of SVP party elites and the systematic anti-retrenchment position of SVP voters (betraying voters). Even if the SVP has not been directly electorally sanctioned for its stance on welfare which differs from its voters, it has faced systematic defeats in referendums about pensions, which can be interpreted as a form of specific sanction as opposed to the diffuse sanctions in elections faced by the other parties in Austria and the Netherlands. The ambiguous stance of the SVP in the last pension reform studied here, however, can be considered equivalent to ‘betraying’ its coalition partners driven by vote-seeking concerns. As a whole, the strengthening of the SVP has led to deadlock in Switzerland because the kind of pro-outsider quid-pro-quo that enabled pension reforms in the 1990s no longer enjoys a sufficient base of support with a pro-insider party yielding much more power. Moreover, the legislative support of the SVP for retrenchment is not really effective because its electoral base does not follow the elite. In all cases, there is a clear trade-off between office or policy on the one hand, and votes on the other.

While I did not produce explicit hypotheses about when parties prioritize office over votes, the tentative pattern which emerges is that PRWPs predictably privilege pro-welfare vote-seeking strategies during electoral campaigns (with the exception of the SVP), office-seeking strategies during coalition formation and at the beginning of their tenure, and then turn to vote-seeking strategies again (and a more welfare chauvinist stance) when they realize the potential electoral costs of their choices. It is difficult to explain when changes occur, partly because party strategies within PRWPs depend to a larger extent than other parties on – sometimes erratic – individual choices by a charismatic leader (Mudde, Reference Mudde2007: 271). PRWPs tend to have more centralized decision-making structures than other parties (Bornschier, Reference Bornschier2010: 3). In general, however, a greater emphasis on vote-seeking strategies in the area of welfare seems to emerge, either by emphasizing welfare protectionism (in Austria and the Netherlands) or continuing to conceal their position on this issue (in Switzerland), thereby making it more difficult to pursue office-seeking strategies in a credible way. In many ways, this is in line with De Lange’s (Reference De Lange2007) idea of a ‘move to the centre’, also corroborated by interview partners in the Netherlands (Interview 12, 2013).

It is important to outline the limits and caveats of the analysis with respect to the policy sector and the countries covered. Pensions were a case in point to outline the trade-off because of their typical pro-insider orientation. However, further research would be needed to see whether this trade-off applies in other areas of welfare protection such as unemployment benefits. Preliminary evidence shows that the stance of PRWPs about unemployment benefits varies: both elites and voters of the SVP have supported retrenchment in Swiss unemployment protection in recent years because it was framed as a source of abuse by ‘undeserving’ recipients, while the PVV has opposed reductions in unemployment benefits. Similarly, further research is needed to assess whether the type of welfare state regime is really a scope condition for the existence of the trade-off: patterns observed in Scandinavian countries such as Denmark or Norway tend to indicate similar developments. Context matters here, because defending the welfare status quo in Norway is different from doing so in Switzerland. With respect to the external validity of the argument, it is also striking to note that the gap between ‘grassroots’ and ‘elite’ preferences about welfare emphasized here is also pointed out by Williamson et al. (Reference Williamson, Skocpol and Coggin2011: 32–33) in their analysis of the Tea Party in the United States, a feature that is similar to the Swiss case. Hence, there is a large scope to assess the relevance the trade-off analysed here for right-wing populist organizations beyond the contexts discussed.