Veto Points Revisited: The Role of Party System Institutionalization in Welfare State Change

Abstract

This article argues that the impact of veto points on a government's policy outcomes depends crucially on the degree of institutionalization of the party system. Specifically, the article claims that two dimensions of party system institutionalization – stability of relations between parties and between parties and voters – condition the ability of the opposition to block governments' policy plans through veto points. It showcases this argument by applying the method of causal process tracing to a comparative analysis of health policy reforms in Slovakia (2002–2004) and Hungary (2006–2008).

While the view that ‘politics matters’ for social policy change has long been accepted in comparative welfare state research, the challenging question is ‘how’. In times in which pressures on welfare states are mounting and some welfare states are undergoing significant change while others stick to existing policies, answering the question of how exactly politics shapes variations in welfare policy continues to occupy scholarly minds (see Jakobsson and Kumlin 2017). This article adds to this discussion by asking which political settings enable governments to pass radical change in welfare state programmes. One dominant approach in the existing literature answers this question by pointing to the distribution of power in a political system. It argues that the degree to which power in the system is concentrated or fragmented is a key factor determining a government's chances for transforming existing social policy programmes. The more fragmented – that is, the less concentrated – power is in the political system, the easier it is for the government to enact change (Hallerberg 2011; Huber and Stephens 2001; Immergut 1990; Immergut et al. 2007; Jahn and Müller-Rommel 2010; Keeler 1993; Tsebelis 1999, 2002).

This article builds on this literature by acknowledging that some aspects of political power distribution, such as the institutional arrangement of veto points – points in the policymaking process at which opposition can block policy change (Immergut 1990) – do matter for the government's capacity to enact changes to existing welfare programmes. However, it argues that the impact of veto points on policy change is conditioned by another key feature of democratic politics: the degree of institutionalization of the party system. Defined as the stability of the relationship between parties and between parties and voters (see Tavits 2008), party system institutionalization conditions the ability of the parliamentary opposition to block the government's policy plans through veto points. There are two dimensions of party system institutionalization that condition the opposition's use of veto points. On the one hand, the relative stability of interparty relations influences the likelihood of cooperation between parties and as such is decisive for the opposition's capacity to act in concert against the government. On the other hand, the strength of party–voter ties shapes the opposition's capacity to mobilize voters and involve them in the policymaking process outside the parliamentary arena, thereby increasing (or decreasing) its chances of blocking government plans.

This argument is showcased through a comparative analysis of health policy reforms in Slovakia and Hungary. After the fall of communism, these two countries reformed their healthcare systems, shifting from tax-funded systems to systems based on social health insurance with limited market elements, such as privatization of primary care (Gaál et al. 2011; Szalay et al. 2011). During the 2000s, government in Slovakia (2002–2004) and in Hungary (2006–2008) attempted to introduce more radical reforms aimed at introducing a variety of market-oriented policy changes such as the privatization of healthcare costs through fees for medical services and the establishment of a competitive insurance system. Explanations focusing on institutional veto points fail to explain why, despite more veto opportunities for the opposition, the Slovak government managed to introduce the radical health reforms relatively smoothly, while the Hungarian government faced a reform debacle in spite of its opposition having a very limited veto menu.

The empirical analysis based on process tracing points to the effects of party system institutionalization on the opposition's capacity to block change through veto points. It demonstrates how, in Slovakia, a weakly institutionalized party system marked by loose interparty relations and fragile links between the opposition parties and their voters generated the dynamics of political competition that undermined the capacity of the opposition to exploit the available vetoes. In contrast, interparty cooperation and strong ties between parties and their voters in the context of the more institutionalized party system in Hungary enabled the opposition to use veto points effectively and block government policy plans in spite of having restricted veto opportunities.

The article makes a twofold contribution to the existing literature. First, it adds to the literature on the politics of welfare state change. Early studies of the politics of welfare were concerned mainly with political parties and looked at how party characteristics such as partisanship explain parties' influence on welfare state policy (Castles 1982). Recent literature has taken a more dynamic approach by investigating how competition between parties affects their impact on social policy change (see Häusermann et al. 2013). For example, some of the most recent studies have looked at how intensity of competition for votes acts as a driving factor in parties' policy preferences (Abou-Chadi and Immergut 2019). While these studies underline the importance of the electoral context in which parties operate, they take the party-centred perspective in order to explain how ‘politics matters’ for welfare state change. However, the most recent literature on European party politics emphasizes that the broader environments in which parties operate – that is, party systems – are undergoing significant changes. Unstable voter behaviour, marked by an increasing propensity of voters to shift parties, and new directions in party developments, marked by the emergence and growing electoral support of new parties, suggest changing dynamics within party systems with growing tendencies towards deinstitutionalization (Chiaramonte and Emanuele 2017; Emanuele et al. 2020; Tsatsanis 2018). This article contributes to the literature on the politics of welfare by emphasizing that while ‘parties matter’, it is ultimately the systemic dimension of party politics that may play the most central role in social policy change.

The article also contributes to the literature on the role of institutions and parliamentary oppositions in the policymaking process. As policymaking is traditionally considered a government's ‘business’, thanks to government's key role in the legislative process in parliamentary democracies (Bräuninger and Debus 2009), the existing literature on policy change has focused predominantly on government parties (see Schmidt 1996). In spite of the influence of the veto points theory (Immergut 1990), which emphasizes the role of the institutional powers of the opposition for policymaking, the conditions under which the parliamentary opposition is able to use these powers to influence the policy process have not been subjected to much systematic comparative examination. This article looks precisely at this aspect, analysing how the dynamic of political competition within the parliamentary opposition influences whether governments are able to pass radical policy changes. As such, it contributes to a relatively recent stream of literature that examines the role of the opposition in the policymaking process (De Giorgi and Ilonszki 2018; Tuttnauer 2018).

The article is structured as follows. The first section reviews explanations of welfare state change that focus on the distribution of political power and tests these explanations by comparing Hungarian and Slovak health reforms. The second section elaborates the main argument of the article. The third section describes data and method, and provides an in-depth analysis of the two cases. The last section summarizes the main findings of the article, discusses their implications and suggests directions for future research.

Explaining welfare policy change: the distribution of political power

In a quest to explain how the distribution of political power affects the probability of policy change, previous literature has focused on different aspects of power distribution and their effect on policymaking. One stream of literature has looked more narrowly at the governing aspect of power distribution, focusing on characteristics of power vested with the government, such as size of government (Jahn and Müller-Rommel 2010; Keeler 1993). The size of government is seen as producing a ‘macro-window’ for reform, as it gives the government both the authority needed to pursue its policy programme and empowers it to implement its policy plans (Keeler 1993). Size, measured by the share of legislative seats occupied by the party or parties of the government, is hence considered directly proportionate to a government's capacity for policy change. Governments with higher shares of seats – that is, with a more concentrated political power within the parliament – are expected to be more successful in introducing reforms than governments with lower seat shares.

Another approach within the literature that focuses on the governing aspect of power distribution looked more closely at the distribution of power within the government. This approach argues that it is not simply the size of the government but the number of parties in the governing coalition – labelled as ‘partisan veto players’ – that determines capacity for policy change (Hallerberg 2011; Tsebelis 1999). Because each of the government parties has the ability to block proposals for policy change, the veto player approach implies that the number of parties participating in the government will be inversely related to the capacity for policy change. Consequently, governments with a lower number of veto players will be more likely to introduce policy change than governments with a higher number of these players. Other studies that have analysed the effect of veto players on policy transformation have also found that ideological distance between government parties matters for policy change. Greater distance decreases the government's ability to introduce reforms and vice versa; a smaller distance between ideological stances of the parties in power increases the chances of policy success (Jahn 2011: 47; see also Tsebelis 1999, 2002).

Shifting away from a focus primarily on the distribution of political power vested with or within the government, the veto points approach (Immergut 1990) looks more broadly at the distribution of political power in the policymaking process. Defined through constitutional arrangements, the distribution of power specifies veto points: points in the policymaking process at which the opposition has the power to block policy change. The focus on veto points emphasizes that policy decisions are made not only in the executive arena – by the government – but are also scrutinized by the opposition in other policymaking arenas, such as the legislative – in parliament – or even in the electoral arena, through recourse to referendums. This view on the distribution of political power puts the opposition to government reform in the driving seat, arguing that if the opposition to reform has access to veto points, it can use these points to block the government proposal at different junctures along the decision-making chain. This view of power distribution also adds the concept of sequencing to the understanding of the policymaking process, emphasizing that the probability that a government will be successful in enacting its policy plans depends on the opposition's agreements at several different points along a decision-making chain (Immergut 1990: 396).

The possibility of blocking change through veto points depends not only on fixed constitutional arrangements that define whether the opposition has a formal right to veto legislative proposals in a given arena, but also on the time-specific alignments of political actors (Jahn 2011: 53). This points to the need to combine these two elements in order to create context-specific institutional configurations in which some of the veto points may be inactive, as the right to veto legislation has been formally granted, but political alignments prevent the opposition from accessing those points. Other veto points, however, may be active if electoral results generate alignments that allow the opposition to use those points to block government policy decisions (see Immergut et al. 2007: 7). As a result, the higher number of active veto points increases the oppositions' chances to block policy change.

Limitations of the power distribution approach: health reforms in Slovakia and Hungary

The explanatory power of approaches that focus on power distribution to account for welfare policy change has been proven through numerous studies (e.g. Hallerberg 2011; Huber and Stephens 2001). Providing simple yet powerful and easily testable theoretical propositions, these approaches make it possible to predict social policy outcomes by looking at the size of government, the number of veto players, their ideological distance and the number of veto points available to the opposition.

However, a comparison of health reforms in Slovakia and Hungary reveals a set of empirical anomalies that challenge the power distribution approaches to policy change. As shown in Table 1, in Slovakia the government was smaller and featured a higher number of partisan veto players but nevertheless introduced large-scale health reforms, while in Hungary a larger government with a smaller number of veto players failed to implement its reform plans (Table 1, columns a and b). This difference in policy outputs is also striking in light of the difference in ideological distance between the veto players, since in spite of the significantly smaller distance in Hungary compared to Slovakia, the Hungarian reform was unsuccessful (Table 1, column c). Lastly, the difference in policy outputs is surprising given the opposition's veto opportunities, which shows that even though the Slovak opposition had a larger menu of active veto points than the opposition in Hungary, it did not manage to block government reform (Table 1, column d).

Table 1. Distribution of Political Power in Slovakia and Hungary in the Context of Health Policy Reforms

^a The percentages of the government seat share in the parliament refer to the distribution of power at the times of voting for health reforms, in September 2004 (see section on Slovakia for a more detailed explanation).

Sources: (a) Slovakia: SOSR (2002b), NRSR (2020) and own calculation; Hungary: NVI (2006); (b) Slovakia: NRSR (2020); Hungary: NVI (2006); (c) own calculation of the distance between the two most ideologically distant government parties based on data on parties' left–right positions from Parties, Institutions & Preferences (PIP) Collection (Jahn et al. 2020); (d) Slovakia: Slovak Constitution (1992), SOSR (2002b) and own calculation; Hungary: Hungarian Constitution (1989) and NVI (2006).

In order to account for what appears to be an anomaly we must revisit the veto points approach by reconsidering some of its core assumptions. The veto point approach is built on an assumption that the parliamentary opposition to policy change is a unified actor and as such will act concertedly against government policy plans. The problem with this assumption is that it makes the approach blind to cases in which the opposition is fragmented (see Maeda 2010). The next section lays out a theoretical framework that accounts for these counter-theoretical cases by emphasizing that the unity of the opposition cannot be taken for granted but is, rather, a consequence of the broader structure of the party system. On the basis of this framework the article builds a more specific claim that the opposition's use of veto points is shaped by party system institutionalization, which influences the dynamic of political competition and, as such, is a key factor in policy change.

Party system institutionalization, veto points and policy change

According to scholars of party systems, institutionalization is a critical dimension for the understanding of a party system (Mainwaring and Torcal 2006). To account for differences in the level of party system institutionalization, these scholars look at whether party systems are stable, to what extent they display patterns and how much predictability there is when we compare one party system with another (Casal Bértoa 2015). In terms of how they conceptualize party system institutionalization, scholars fall broadly into two ‘camps’ (see Tavits 2008). One, with more traditional roots, focuses on the demand dimension of the political process – that is, at the party–voter linkages (see Mainwaring and Torcal 2006). Scholars who take this approach assume that volatile electorates are responsible for system instability and argue that the level of institutionalization of the party system depends on the extent to which voter and party alignments are stable and predictable. The other camp looks instead at the supply dimension of the political market – at the relationships between parties (see Casal Bértoa and Mair 2012). This second group of scholars views party system dynamics as an elite-level phenomenon and claims that party system institutionalization is defined by the degree to which the system displays predictable patterns of interaction between parties (Tavits 2008).

Both of these dimensions of party system institutionalization are relevant to a government's capacity to enact policy change. The stability of interparty relations in a political system indicates the readiness of the parties to cooperate and concerns the predictability of party cooperation. As put by Zsolt Enyedi and Fernando Casal Bértoa (2015), depending on the degree to which parties have a tendency to cooperate, party politics can be defined as either an individual or a team sport. The degree to which party politics is a team or an individual sport is crucial for policy change, as the general patterns of interparty cooperation will also influence party dynamics between the parties of the opposition. Cooperation between opposition parties can be decisive in the policymaking process, for example when government policy proposals are voted on in the parliament. Even if the opposition has a de jure possibility to veto government proposals in the legislative arena, its capacity to do so will depend on its de facto ability to cooperate. If parties of the opposition are team players − that is, if they have previous histories of strong interparty bonds in the form of electoral alliances or government coalitions − this increases the likelihood of cooperation, hence increasing the opposition's ability to block change. In contrast, in party systems that feature solo players, the probability that opposition parties will cooperate and challenge government reforms is lower, as government policy proposals can be seen as yet another opportunity for ‘dog eat dog’ competition between the parties in the opposition. This low cooperation potential translates, in turn, into a weak ability to challenge government plans through veto points.

The opposition's ability to use veto points is also shaped by the stability of party–voter linkages. Political systems with weak party–voter linkages are characterized by high electoral volatility, which implies that voters have a high propensity to change their voting behaviour – that is, to shift parties from one election to another (Pedersen 1979). High electoral volatility indicates that parties have weak roots in society (Manwaring and Torcal 2006) and is inherent in party systems featuring the frequent appearance of new parties (Sikk 2005). Electoral volatility as an indicator of parties' anchoring in society conditions the mobilizing capacity of political parties. In the case of opposition parties, this mobilizing capacity can play a decisive role in policymaking settings in which the opposition has limited veto opportunities. For example, this is the case where the opposition lacks the potential to block the government's plans in parliament due to a government majority, but is able to move the decision-making into the electoral arena – by challenging the government's reform in a referendum. In party systems in which parties have stable relationships with their electorate and strong roots in society, opposition parties will have a stronger capacity to mobilize their voters and involve them in the policymaking process, increasing their chances of blocking government plans through a popular vote. In contrast, in systems with weak party–voter linkages, the electoral arena can be a de jure veto point, but its use in practice will depend on parties' mobilizing capacities.

A comparison of Slovak and Hungarian health reforms illustrates how these two dimensions of party system institutionalization influenced the opposition's use of veto points that in turn shaped government capacity to enact policy change. In Slovakia, the opposition to Mikuláš Dzurinda's government, which promoted the 2004 health reform, was emblematic of the country's weakly institutionalized party system. The opposition was characterized by a lack of stable interparty relations, as its parties had no history of party bonds in the form of electoral or governing coalitions. Its parties also featured weak links with voters, either because of high electoral volatility or because they were newcomers on the country's political landscape. The unstable dynamic within the opposition was accentuated by the presence of a new party: Direction – Social Democracy (Smer). In spite of being a newcomer, Smer was a fervent opponent of the health reforms and quickly became the government's main adversary. However, being a newcomer implied that Smer lacked stable relationships with other parties and with the electorate. The lack of stable party relations and linkages with voters translated into a low capacity on the opposition's part to cooperate or mobilize its voters, making it incapable of blocking the health reforms in spite of a relatively large menu of veto opportunities.

In contrast, the case of the Hungarian reforms of 2006 shows how the context of a highly institutionalized party system made it possible for the opposition to block government health plans in spite of its restricted veto powers. Reflecting the broader pattern of party system institutionalization, the opposition was marked by the presence of stable party blocs gathered around the dominant Hungarian Civic Alliance (Fidesz) and firm party–voter linkages, mainly thanks to the Fidesz's strong social roots. As shown below, these two features of the opposition to the Ferenc Gyurcsány government were key to its success in using its meagre veto powers to block radical health reforms in the electoral arena.

Empirical analysis

The selection of the cases for empirical analysis is justified by the methodological rationale of the ‘most similar system design’ (Teune and Przeworski 1970). The scope and the type of the intended policy change in the two countries were similar, as both reforms implied radical market-oriented change to the health system. The governments in the two countries also managed to get legislation based on their reform plans through parliament. Yet, the final outcomes of the healthcare reform paths in the two countries were markedly different – policy reform in Slovakia and its reversal in Hungary.

Data for this study come from the electoral statistics, official policy documents, transcripts of parliamentary debates, legislative bills approved in parliament and newspaper articles, collected for the respective periods of health reforms in the two countries (2002–2004 in Slovakia and 2006–2008 in Hungary). Based on this data, we conducted a qualitative analysis of the two cases of health reforms using process-tracing methodology (Beach and Pedersen 2013). Process tracing allows for the combination of an inductive approach, used to identify elements of policy change (or lack thereof), with a causal-analytical framework, which helps us to analyse which dimensions of party system institutionalization figure in policy change, and how. The inductive approach involved in process tracing allows for an analysis that goes ‘backward from the outcome by sifting through the evidence in an attempt to uncover a plausible sufficient causal mechanism that produced the outcome’ (Beach and Pedersen 2013: 169).

Relying on the process-tracing methodology, we combine a theory-driven framework with inductive analysis in order to identify party system related conditions under which the role of the veto points is significant for policy change. The theoretically driven causal framework (Table 2) identifies the relationship between the characteristics of the party system institutionalization and the distribution of vetoes in the policymaking process, and specifies the impact of this relationship on the policymaking output. This framework constitutes the basis for inductive analysis which traces the reform process in each country, concentrating on how the institutional context of the party system affected the behaviour of the parties of the opposition, in particular the behaviour of the parties that explains the policymaking output.

Table 2. Process-Tracing Framework for the Analysis of Policy Change

Slovakia

Slovakia's healthcare reforms took place in a policymaking context framed by the country's electoral system and constitutional rules. The post-communist electoral laws established a highly proportional electoral system associated with low party system institutionalization, reflected in high levels of parliamentary fragmentation and electoral volatility (Casal Bértoa 2011; Powell and Tucker 2014). According to the 1992 Constitution, the veto powers were conditionally granted to the parliament, which meant that the opposition could block legislation only when government was in minority, to the president of the republic – as his veto could be overturned by parliament's majority – and to the judicial and electoral arena (Slovak Constitution 2002).

Slovakia's 2002 parliamentary elections, which determined the political context of the healthcare reforms, provided paradigmatic examples of the country's weakly institutionalized party system. The party that dominated Slovak politics throughout the 1990s, the Movement for a Democratic Slovakia (HZDS), experienced a significant drop in electoral support, obtaining 19% of the votes; its main opponent, the Slovak Democratic and Christian Union (SDKÚ), came second with 15% (SOSR 2002a).¹ At the same time, the failure of the Slovak National Party (SNS) − the main coalition partner of the HZDS − and of the communist successor party, the Party of the Democratic Left (SDL), to cross the 5% threshold suggested that the electorate was disillusioned with the established parties. The emergence of a new political dynamic was also signalled by the electoral success of two new parties: the left-wing Smer, the SDL's splinter party, and the pro-market Alliance of the New Citizen (ANO). The last but not the least important surprise was that, for the first time after the fall of communism, the unreformed communist-successor party, the Slovak Communist Party (KSS), managed to pass the electoral threshold and entered parliament. After the elections, the SDKÚ took the lead, quickly forming a centre-right government with three other parties: ANO, the Christian Democratic Movement (KDH) and the Party of the Hungarian Coalition (SMK-MKP), with the SDKÚ's leader Dzurinda as prime minister.

Even though the government's electoral support initially resulted in a majority of parliamentary seats (SOSR 2002b),² its composition implied that the successful pursuit of the government agenda might be difficult, due to the relatively high number of veto players and intra-coalitional tension (Haughton and Rybář 2008). Yet, the party composition of Dzurinda's government implied that the opposition was composed of a specific set of parties, including the waning HZDS, the newly formed Smer and the KSS, the post-communist political newcomer. Composed of such different parties, the opposition reflected weak system institutionalization – it was fragmented and characterized by two lines of conflict. One line, which had already emerged during the election campaign, ran between the HZDS and Smer. The election campaign of Smer's young and ambitious leader Robert Fico was seen as aggressive and arrogant: it focused on presenting the party as a new political force transcending the old divisions of Slovak politics (Haughton and Rybář 2008; Henderson 2002), which inevitably generated conflicts with the mainstream yet waning HZDS. Another line of conflict ran between Smer and the KSS. As the new political force on the left, Smer wished to be recognized as a social democratic party, clearly distinct from the communists; it demonstrated this new social democratic identity by adding ‘Social Democracy’ to the party name (Haughton and Rybář 2008).

Shortly after entering office in October 2002, the Dzurinda government presented its larger plan for neoliberal, market-oriented reforms of the public sector (Fisher et al. 2007), which included a plan for health reform. The plan, outlined in a White Paper, co-authored by Minister of Health Rudolf Zajac and titled ‘Health in the Service of Citizens’, justified the need for reform by criticizing the poor state of affairs in the state-dominated healthcare system. It argued that the shift from a state-led to a more competitive, market-oriented model of healthcare provision would stop the rising debt of the health sector, enable its financial stabilization and generate efficiency. The reform plan also provided detailed descriptions of needed changes that implied measures such as the introduction of user fees for medical services, the creation of voluntary health insurance and the transformation of health insurance funds and hospitals into joint-stock companies (Pazitný and Zajac 2004).

As the first step of the reform, on 16 May 2003 the Ministry of Health issued a decree introducing fees of 20 Slovak crowns (0.5 euros) per physician visit and drug prescription, and 50 crowns (1.24 euros) for a day in hospital (Figure 1).³ Even though the fees were relatively small, they immediately spurred hot political debate. The opposition parties criticized the ministry's move but their actions reflected internal division. On 16 June Smer's leader Robert Fico, together with 35 opposition MPs, filed a complaint with the Constitutional Court arguing that the fees were unconstitutional as they threatened the guarantee of the right to free healthcare. About three weeks later, another group of 30 opposition MPs, led by Ján Cuper from the HZDS, filed a separate complaint based on the same argument. In a joint response to these complaints, in May 2004 the court ruled that the fees were not in conflict with the constitution, claiming that the right to free healthcare did not exclude the possibility of requiring payment for some aspects of healthcare provision (Slovak Constitutional Court 2004).

Figure 1. Timeline of Healthcare Reform in Slovakia (2002–4)

In the midst of the debate over the constitutionality of the fees, in April 2004 the government presented parliament with a comprehensive reform package containing six bills (NRSR Bills 2004). The bills, based on the above-mentioned White Paper, envisaged radical changes to the Slovak health system, the most important of which was the transformation of health insurance funds into for-profit, joint-stock companies subject to private commercial law. It also envisaged the establishment of a Healthcare Surveillance Authority that would supervise these companies and monitor their solvency, and introduced sets of provisions that distinguished between services fully covered by health insurance and those that implied co-payments, which could be covered by voluntary private insurance (NRSR Bills 2004).

While the health reform package was comprehensive, its passing in parliament looked anything but certain because of an unexpected reshuffle of the government. Following an intra-coalitional conflict, in early 2004, Dzurinda's SDKÚ lost seven of its MPs, who formed a new party, Free Forum (SF) (Spectator 2004a). Joining the opposition, SF turned Dzurinda's government into a minority government, increasing the opposition's parliamentary share and thus turning the parliament into an effective veto point.⁴ Nevertheless, the opposition's new veto opportunity did not end up being put to use, as relations between its parties became marked by increasing conflict. During the initial parliamentary discussion of the health reform package, all parties announced that they would oppose the reform (NRSR Debate 2004a, 2004b), but over time Smer took the lead. In an effort to present his party as the only real opposition to the reform, Smer's leader Fico announced that if the government passed the health reform, his party would be the one to take it to the Constitutional Court, and presented his party's own plan for health reform (Spectator 2004b). The newly formed opposition party, Free Forum, also refused to support reform, presenting its own alternative plan (Spectator 2004c). The HZDS, however, maintained an ambiguous position: while initially opposing the reform, it later admitted that the party might provide support for the reform package if certain changes were made (Spectator 2004d).

Divisions in the opposition ranks turned what initially looked like an insurmountable problem into a relatively easy task: despite its minority status and intra-coalition conflict, on 21 and 22 September 2004, the government passed the above-cited six laws with majority votes (NRSR Voting Record 2004a, 2004b).⁵ In a continued effort to position itself as the only real opposition to the government, Smer appealed to President Ivan Gašparovič to veto the healthcare laws. Vetoing all six healthcare laws on 18 October, Gašparovič expressed his concern at the consequences of health reform and asked MPs to reconsider two aspects of the reform that he saw as most problematic: the transformation of insurance funds into joint-stock companies and the financing of the proposed Health Care Surveillance Authority with health insurance premiums (Gašparovič 2004).

However, the president's veto was effectively overturned by parliament's majority vote just three days after Gašparovič's proclamation. The six healthcare laws were passed on 21 October for a second time in their original form (NRSR Voting Record 2004c). Voting on the most contested law – Law on Health Insurance Companies and Surveillance – illustrated how fragmentation in the opposition resulted in a lack of voting discipline, which in turn affected the reform outcomes. In addition to the 69 votes from the MPs belonging to the ruling coalition, the law received backing from 12 MPs who were members of the HZDS and the KSS, former members of the SDKÚ who were now members of the new Free Forum and even one vote from an MP belonging to Smer (NRSR Voting Record 2004c).

The lack of voting discipline during the passage of the healthcare bills was a symptom of the weakly institutionalized party system marked by intensified competition among the opposition ranks, which grew stronger as parties were approaching the 2005 regional elections, seen as a rehearsal for the 2006 national elections. Smer, whose popular support was experiencing steady growth (Haughton and Rybář 2008), hoped for major electoral success. Party leader Fico started his campaign by arguing that if his party gained power, it would revise the health reform (Spectator 2005a). Distancing itself further from the KSS as a former communist party and in preparation for the election, the party merged with three other smaller social democratic parties, more clearly positioning itself as a centre-left political force (Haughton and Rybář 2008). At the same time, the HZDS saw regional elections as a chance to regain power, which resulted in increasing hostility towards Smer. The pre-election period was also characterized by more frequent fights between the two parties, often based on personal antagonism between their leaders Mečiar and Fico (Fisher et al. 2007).

In December 2005, opinion polls showed more than 70% of respondents disapproved of the government health reform (Spectator 2005b). This suggests that the reversal of healthcare reform through a possible popular referendum was yet another missed opportunity for an opposition riven by interparty division and marked by a weak social base. However, Smer turned out to be a winning party since, in spite of the relatively modest achievement in the 2005 regional elections, in the 2006 national elections it won the highest share of votes (SOSR 2006), marking the first victory for a left-wing party in Slovakia's national elections since 1920 (Haughton and Rybář 2008).

Hungary

Similar to Slovakia's, Hungary's post-communist constitutional framework assigned conditional veto powers to the parliament and the president, and to the judicial and electoral arena (Hungarian Constitution 1989). Yet, in contrast to Slovakia, in the post-communist period Hungary established a markedly different electoral system based on a weakly proportional formula. This had a strong impact on the number of parties and thus also on the levels of parliamentary fragmentation, which were low and associated with high levels of party system institutionalization (Casal Bértoa 2011; Toka and Popa 2013). As a consequence, by the end of the first transitional decade the country's party system displayed substantial stability, characterized by low voter volatility and the existence of two party blocs dominated by the main political parties: the Hungarian Socialist Party (MSZP), which was a communist successor party, and the Hungarian Civic Alliance (Fidesz) (Casal Bértoa 2011; Powell and Tucker 2014).

The results of the Hungarian 2006 elections confirmed the dominance of the two parties on the country's political scene as the MSZP and Fidesz won 43% and 42% of votes, respectively (NVI 2006). The post-election period resulted in a government formed by the MSZP and its loyal, minor coalition partner, the liberal Alliance of Free Democrats (SZDSZ). The result was a government under the MSZP's leader, Ferenc Gyurcsány, with a low number of veto players and comfortable majority of parliamentary seats (NVI 2006).⁶

The formation of the MSZP–SZDSZ government coalition left Fidesz in opposition together with its own allies, the Christian Democratic People's Party (KDNP) and the Hungarian Democratic Forum (MDF). In contrast to the MSZP, Fidesz had its origins in the underground anti-regime movement active during the last few years of communism. Thanks to the party's distinctive political culture, based on strong anti-communist core and popular mobilizing strategy, during the transitional period Fidesz managed consistently to capitalize on its protest roots and successfully transform itself from a suppressed clandestine movement into one of the country's leading parties, with large voter support (Szabó 2011; Toka and Popa 2013). The Fidesz–KDNP–MDF alliance, like the alliance between the MSZP and the SZDSZ, could be traced back to the 1990s, and the 2006 elections confirmed that Fidesz was clearly the strongest of the three parties, as it had won the highest number of parliamentary seats (NVI 2006).

Quickly after taking office in June 2006, the MSZP–SZDSZ government announced its proposal for health reform laid out in a document titled ‘The Green Book of Hungarian Health Care’ (Hungarian Ministry of Health 2006). The proposal envisioned large-scale reforms that would replace the existing public, state-dominated system with a decentralized, competitive insurance system with private healthcare facilities and user fees for services. It implied the introduction of multiple insurance funds in the form of for-profit, joint-stock companies that would compete for clients in the healthcare insurance market. It also foresaw the transformation of public hospitals into joint-stock companies and the introduction of fees for doctor and hospital visits. Even though the healthcare reform plan was market-oriented, it fitted well with the MSZP's strategy to present itself as not a statist, but a progressive, liberal left party with the social democratic profile willing to break with its socialist past (Toka and Popa 2013: 309). In fact, during the parliamentary debate on the reform proposal, the Minister of Health, Lajos Molnár, placed the user fees at the forefront of the reform, arguing that they would help eradicate informal payments for healthcare services – inherited from socialist times – by converting them into formal charges, reduce the unnecessary demand for healthcare and raise additional revenue for the financially instable insurance sector (Országgyűlés Debate 2006a).

As soon as the government announced its reform plan, the parties of the opposition started vehemently criticizing it. Fidesz and its allies, the KDNP and MDF, jointly attacked the plan as ill-thought-out and unfeasible and promised to reverse the reform as soon as they returned to power (Országgyűlés Debate 2006b, 2006c). Throughout the parliamentary discussion, the parties were most critical of user fees, arguing that rather than eliminating corruption the fees would create barriers in access to health services for the poor (Országgyűlés Debate 2006a). In spite of the opposition's disapproval, the Gyurcsány government's strong majority enabled the passing of three major reforms in December 2006 (Figure 2). The amendment to the Law on Healthcare Insurance introduced user fees for several types of health services: 300 Hungarian forints (1.2 euros) for a visit to a general practitioner (GP) or outpatient specialist, 600 forints (2.4 euros) for a visit to a specialist without GP referral and 300 forints (1.2 euros) for a hospital visit. Another law marked a first step towards managed competition as it established the Health Insurance Supervisory Authority in charge of supervising actors in the health insurance market. The third law allowed for changes in the legal status of hospitals from public, self-governing institutions to joint-stock companies open to privatization. Unsurprisingly, the voting pattern for these legislative changes reflected relatively high cohesion and strong party discipline within both the government and the opposition bloc (Országgyűlés Voting Record 2006a, 2006b, 2006c).⁷

Figure 2. Timeline of Hungarian Health Reform (2006–8)

As early as during the parliamentary debate over the aforementioned laws, the opposition parties resorted to their voter base and veto powers, starting a campaign for a public referendum on the cancellation of user fees (Index 2006). The opposition also turned to its other veto option – the Constitutional Court – asking for a constitutional review, arguing that the health fees ran counter to the constitutionally granted right to health and imposed an undue burden on citizens and as such should be subject to a referendum. However, the constitutional litigation failed as, in April 2007, the court decided that the alleged unconstitutionality of the fees was unfounded, arguing that the constitutional right to health did not guarantee an individual right, but rather established a state duty – subject to broad discretion on the part of policymakers – to regulate healthcare entitlements (Hungarian Constitutional Court 2007).

In the meantime, the government initiated discussion of another important reform aimed at the introduction of a competitive, market-oriented health insurance system. After failing to block the user fees in the judicial arena, the opposition attacked this reform by resorting to the veto opportunities in the executive, appealing to the president, László Sólyom. Even though he did not belong to the party ranks, Sólyom had strong ties to the opposition as Fidesz had nominated him for the presidency and supported his election (Szabó 2011). On 17 December 2007 the government passed the Law on Health Insurance Funds, which established a decentralized, competitive insurance system with 22 insurance funds in the form of joint-stock companies in public-private ownership.⁸ In a statement made 10 days after its introduction, Sólyom asked for the new law to be returned to parliament for reconsideration, arguing that its long-term effects were unpredictable and underlining the need for public approval (Sólyom 2007). However, in February 2008, parliament overruled the president's veto, passing the law for the second time (Országgyűlés Voting Record 2007).

The failure of both the court's and the president's veto strengthened anti-reform sentiment and allowed the opposition to mobilize even stronger public support. Anti-reform mobilization also built on the larger anti-government protest caused by the leaking of a speech by Prime Minister Gyurcsány which suggested that his party's leaders knowingly lied about the state of the country's economy in order to secure the government's second term (Toka and Popa 2013). Drawing on tried-and-true mobilization strategies (Szabó 2011), the opposition parties allied with the trade unions, organizing a wave of nationwide strikes and inciting associations of healthcare professionals and civil groups to stage national-level demonstrations and street rallies (Index 2007, 2008a; Tóth and Neumann 2008). Recognizing that the introduction of user fees had become the most heated and politicized issue of the healthcare reforms, the opposition continued to campaign vigorously for a national referendum on the abolition of fees. As the court finally approved it, the referendum was held on 9 March 2008. With a significant turnout of 50.5% of eligible voters, and by an overwhelming majority of more than 80% of votes, the popular vote abolished the fees (NVI 2008).

Following the referendum's success, the opposition was encouraged and signalled its intention to hold yet another referendum on the remaining healthcare legislation. Fearing that this referendum would have the same outcome as the referendum on user fees, parliament revoked the Law on Health Insurance Funds in May 2008. This created an unparalleled case in the history of Hungarian law-making where the same parliament approved a law twice and then repealed it within the space of six months (Gaál et al. 2011).⁹ The failure of the health reform soon led to friction in the governing coalition and resulted, first, in the resignation of the health minister, then in the SZDSZ leaving the government and, finally, in the resignation of Prime Minister Gyurcsány in April 2009 (Index 2008b; Toka and Popa 2013). The next elections in 2010, unsurprisingly, resulted in a convincing victory for the Fidesz-led coalition, which won enough seats to form the two-thirds majority (NVI 2010) needed to modify the country's legislation, including the constitution.

Conclusion

This article has analysed factors that condition a government's capacity to introduce policy change through comparative process tracing of two episodes of health reforms in Slovakia (2002–2004) and Hungary (2006–2008). This analysis has demonstrated that differences in the capacity of the governments to enact the health reform were the result of differences in the two dimensions of party system institutionalization – the stability of relations between parties and between parties and voters – which shaped the opposition's ability to take advantage of its veto powers in the policymaking process. The inability of the Slovak opposition to use its vetoes in the policymaking process was due to conflicting interparty relations within its ranks and fragile party–voter linkages accentuated by the presence of a new party, Smer. Contrary to the Slovak experience, in Hungary interparty cooperation and strong ties between parties and voters within the opposition, gathered around the powerful Fidesz, allowed for the effective use of the opposition's restricted veto opportunities.

These findings have two important implications. First, they add support to an argument that in order to understand the politics of policy change we should depart from a view of political parties as resourceful organizations autonomous from social structures (see Häusermann et al. 2013). Parties are organized groups focused on the acquisition and exercise of political power and, as political power is a sine qua non for policymaking, parties' activities are critical in shaping policy directions. However, parties ‘do not come alone’: rather they are embedded in settings with established patterns of party competition that shape how parties behave in the policymaking process. Hence, while parties have their own dynamic, this dynamic is simultaneous with the broader dynamic of a party system that, as shown in this article, can either facilitate or impede parties' influence on policy change. Viewing parties as embedded within the broader party system dynamic calls for reconsideration of the still-dominant focus on absolute party qualities such as partisanship, and points to the need to integrate relational qualities of parties – their links with one another and with voters – into the analysis of party influence on welfare state change.

The article's findings also have implications relevant for research on the politics of policy change across Europe. The analysis has focused on the role of party system institutionalization on policymaking in two ‘new’ democracies in Eastern Europe. While previous studies on party systems have contrasted the weak institutionalization in ‘new’ Eastern European democracies, with the strong institutionalization of party systems observed in ‘mature’ Western European democracies, recent research has increasingly called into question this conventional wisdom (Chiaramonte and Emanuele 2017). The most recent study of trends in electoral volatility in Eastern and Western Europe points to a convergence between the two regions (Emanuele et al. 2020), underlining that characteristics of the party systems may become increasingly relevant for the politics of policy change across Europe. Analyses detailing how different aspects of party system stability influence policy change across a larger number of European countries would provide welcome tests of the external validity of the present findings and thus constitute promising avenues for further research.