The determinants of parliamentary opposition behaviour in the Hungarian legislative process

Transcription

1 Réka Várnagy, assistant professor, 1 Gabriella Ilonszki, Professor, 2 The determinants of parliamentary opposition behaviour in the Hungarian legislative process First draft, please do not quote! Paper presented for the XXII. World Congress of Political Science Madrid, 8-12 July Corvinus University of Budapest, 8 Fővám tér, Budapest, H-1093, Hungary 2 Corvinus University of Budapest, 8 Fővám tér, Budapest, H-1093, Hungary 1

2 Introduction In the institutional approach the Hungarian legislative process is dominated by the government as the power of the opposition, the parliamentary opportunity structures of the opposition has been highly limited as documented by academic studies (Ilonszki, 2007, Zubek, 2011). The logic of Hungarian politics also implies the division of government and opposition: until 2010 a growing polarization and rigid block politics characterized the political arena where the two sides followed a strict left-right divide. Strong bloc politics has been reflected in the high Rice-index (the voting unity) of the Hungarian PPGs 3, that is very close, occasionally even exceeds the average unity measured in consolidated democracies, and much higher than in other East Central European countries (Dalton-Farrel McAllistair, 2011:197). With no parties ever crossing the coalition formation line between the two blocks (Enyedi Casal Bertoá, 2011) and a high coalition unity 4 (Ilonszki-Jáger 2010) we expect to witness a conflictive opposition behavior in the parliamentary arena. However, previous research (Jáger-Várnagy-Ilonszki, 2011) has pointed out a surprisingly high consensual voting pattern in the Hungarian parliament which provokes a question of why: what are the driving forces of consensus in a conflicting political context? Studying the legislative process and analyzing roll-call data has been dominated by US research where the driving force behind the legislative logic is the behavior of individual legislators. In the Hungarian case the above described features of strong bipolarization and high voting unity of PPGs make the party and the parliamentary party group the basis of the analysis in understanding the legislative process. Moreover, we argue that it is the government versus opposition dynamic that determines this process. Indeed, in face of block politics and strong party unity it is a real challenge to understand the background of consensual legislation. Why do opposition parties support bills at all? And why is support missing in other instances? Among the explanatory factors we distinguish between political-institutional and bill-related explanations. Out of the political institutional explanations the consequences of the electoral system and the type of government have to be considered. Research on electoral systems in this context draws the attention to the potential pork barrel behaviour of MPs elected in an 3 91,0-98,25 for the coalition PPGs, with slightly lower values for opposition PPGs (Ilonszki-Jáger, 2010). 4 Even reaching 100 % under the Medgyessy government, and lowest during the Gyurcsány I government, as the coalition conflicts already appeared, but still kept at 89,81%. 2

3 environment where weak party discipline and strong constituency presence prevail. This may conclude in parliamentary behaviour where party lines are crossed for the sake of constituency service and legislative consensus is created on other than party divisions (Mayhew, 1974) This approach is extended by the notion of the personal vote removing constituency service and thus parliamentary vote out of the exclusively electoral dimension (Fiorina, 1997). Hungary, as a mixed-member electoral system, in principle, might provide examples of this type of behaviour. Former research has proved however that there is no obvious mandate divide, pork barrel behaviour cannot be proven and in the legislative dimension of parliamentary work the personal vote does not exist, even local politicians who also serve as MPs do maintain their party belonging and discipline thus we cannot except consensus building in legislature on these grounds (Ishiyama, 2000, Thames, 2001, Várnagy, 2008, Papp, 2011). As for the potential impact of the government structure the Hix and Noury (2011) distinguish three types of legislative voting: the floor agenda model where legislative coalitions are built issue by issue, the cartel agenda model where a group of agenda controllers (more particularly the government majority) has power to control the agenda and thus influence the vote, and the Westminster model where the government monopoly results in an exclusive government-opposition divide. Their comparative, 16 country study proves that even among the conditions of parliamentary government different legislative conditions might prevail depending on the government type. For example, out of the post-communist countries the minority government in the Czech Republic ( ) produced the floor agenda model and a majority coalition in Poland ( ) produced the cartel agenda model. Our question in this respect is the following: is the Hungarian political system with its obvious majoritarian bias, bipolarisitation and disciplined strong parties an easy case of the Westminster model- or due to other systemic (institutional) and political factors some flexibility and variation do prevail? As for the bill related explanations there is a rich literature (e.g. Döring, 1995ab, Döring and Hallenberg, 2004, Tsebelis, 2002) which proves that different bill characteristics (from its timing through the number of actors involved to the nature/saliency of the bill) will determine whether a bill does have a fair chance to get accepted and more particularly whether consensuality can develop around it. From the Hungarian perspective the strong agenda control of the government and the constitutional requirement to legislate with a 2/3rd 3

4 qualified majority on certain policy issues do particularly circumscribe the chances of bills. However, previous research has suggested that policy-orientation of bills does not have an explanatory power in terms of consensual voting nor does EU-related legislation in itself as the bills often encompass a complex set of issues relating to various policy-fields. In the analysis below we shall thus explore only a limited scope of bill-related issues, namely the initiator, the required majority and the amendments to understand the dynamics of the legislative process. As the question on consensual voting implies, we will focus on the legislative process and base our research on roll-call data assuming that a critical opposition right is to influence the policy-making process and its outcome in the Parliament. Along with the importance of influencing the policy-making process we assume that legislative debates also contribute to the mediating role of the Parliament which has came to the forefront ((Mansfeldova 2011: 128) in times when political conflicts are more and more sharp. Our focus on consensus thus not only reflects the assumption that consensual decision-making yields better legislative results 5 but also analyses the mediating capacity of the Parliament itself. The framework of opposition activities in the Hungarian parliament Following upon Blondel s argument we can distinguish between three powers of parliaments: the preventive power, which concerns the agenda-setting rules with special attention to informal negotiations, the power of initiation which concerns the ability to put forward bills (especially private members bills) and the power of reaction which reflects the ability to modify bills in the legislative process (Blondel, 1970). In the Hungarian legislative process, the government, the standing committees 6 and single MPs are entitled to initiate bills without quantitative restrictions. Yet, bills proposed by MPs only get on the order book of the parliament, thus they are only dealt with, if the designated committee decides so by simple 5 As:Tindale, Cameda and Hinsz note Groups are seen as superior to individuals as decision-making entities for at least two reasons. First, groups can represent a larger and more diverse set of perspectives, constituencies, etc. Thus, they tend to be seen as more fair by providing voice or input from a greater portion of the body for which the decision is made (Darley & Tyler, this volume; Folger, 1977; Thibaut & Walker, 1975; Tyler & Smith, 1998). However, groups are also perceived as better than individuals at making important decisions (Paulus, Dzindolet, Poletes, & Comacho, 1993). The idea that two heads are better than one is widespread and typically accurate, based on the empirical record (Hastie, 1986; Hill, 1982; Kerr, MacCoun, & Kramer, 1996a ; Laughlin & Ellis, 1986). (Tindale-Cameda-Hinsz, 2003:382)- 6 The President of the State is also granted the right to propose bills, but traditionally does not practice this right. 4

5 majority vote 7. In contrast, committee and government bills do not need the approval of the designated committee for being placed on the order book. In practice this results in strong filtering of the opposition bills as the membership of the standing committees reflects the size of the parliamentary party groups, thus allows strong agenda control for the government as long as it can rule the majority of MPs. Conversely, to be on the order book does not equal getting on the floor. The agenda is adopted by the plenary session with simple majority, but the proposal for the agenda is prepared by the House Committee 8, where the leader of each PPG has one vote and they are supposed to reach consensus about the planned agenda, hence opposition preferences can be expressed. On the other hand, when consensus cannot be reached, the Speaker of the House is entitled to propose the agenda. As the Speaker is elected by the majority of the MPs, the position is always filled by the government parties, therefore at the end of the day, the opposition has little influence on which bills get on the floor. As Zubek already concluded, the negative agenda control of the Hungarian governments is very strong, resulting mainly from the concentrated party system (Zubek 2011). The agenda setting power of the government is thus important, the parliament became the prisoner of a highly centralized government (Ilonszki - Olson 2011b:243).The negative agenda control of the government can be grasped also in the decreasing number of accepted bills that require 2/3 rd majority. When it comes to the final voting we have to distinguish between two types of bills: most laws are passed by simple majority, however two-thirds majority of the MPs present is required to pass (or amend) laws with constitutional force such as laws on the electoral system or the local government system 9. With time being one of the scarcest resources in parliaments (Döring, 1995a), only bills that have the chance to pass are introduced to the floor and thus when the government needs the support of opposition parties to legislate, negotiations between different parliamentary actors are a must. Table 1 reflects the number of bills introduced and accepted which require 2/3 rd majority in seven different years (including five parliamentary cycles). The increase in 2010 and 2011 can be explained by the government controlling for 2/3 rd majority in the Parliament which enables them to legislate 7 The Standing Orders make it possible for PPG leaders to call for a vote on the plenary to place rejected bills on the order book six times each session, but chances are pretty low to have a different outcome than in the committee, as the ratio of the MPs from the different PPGs is the same. 8 Members of the House Committee are the Speaker and Vice-speakers of the House and the leaders of the PPGs, but only the leaders of PPGs have voting write. 9 The new Fundamental Law that entered into force the 1 st January, 2012 lists 32 main themes that require a twothird majority and call these as cardinal laws. 5

6 upon qualified majority bills without the consent of opposition making preventive legislative control unnecessary. Table 1: The number and percentage of accepted bills requiring 2/3 rd majority Year of voting 2/3 majority required no yes Total 1996 N % 82,8% 17,2% 100,0% 2000 N % 87,2% 12,8% 100,0% 2005 N % 93,6% 6,4% 100,0% 2008 N % 96,6% 3,4% 100,0% 2009 N % 93,0% 7,0% 100,0% 2010 N % 65,6% 34,4% 100,0% 2011 N % 64,7% 35,3% 100,0% Total N % 82,0% 18,0% 100,0% Table 1 shows that the proportion of bills that require a 2/3rd majority support declined from 1996 to 2008, stabilized 2009 and increased in the two more recent years as mentioned above only due to the particular majority condition in the current House. Blondel highlights that the preventive power of parliaments is hard to measure as the negative agenda control and the reasons behind it are in many cases invisible (Blondel, 1970:199) but in terms of legislative proposals the power of initiations can be captured. Table 2 shows the overall picture about the government opposition connections in terms of legislative proposals. The first tendency we should observe is the growing absolute number of bills proposed and accepted which signals a growing pressure to legislate on the Hungarian Parliament. It is especially true for the last period of where only 1,5 years are taken into consideration (out of the four year cycle), yet the number of proposed bills has already reached 2/3 of the bills submitted in the previous period. Analyzing the type of initiator, a clear governmental dominance is visible since the government submits more than 50% of the bills in all parliamentary cycles. Its success rate also evident as it reaches 90% in all 6

7 governments which counts for growing percentage of all passed legislation. This tendency already points out the restricted possibilities of the opposition: while opposition bills stabilized around 23-25% their success rate is rather low, 4-6%. Committees are more successful in proposing and passing bills but their share of the legislative work is rather low around 2-4%. Government MPs do also enjoy a higher success rate ranging from 24% to 50% in previous parliamentary cycles and mounting up to 76% in the period. The current parliamentary cycle seems to be an outlier in other dimensions as well: the number of bills proposed by government MPs is already more in absolute terms than in any other parliamentary periods reaching 28% of all proposed bills. Along with their high success rate we can suspect that MPs can behave as governmental messengers introducing bills in accord with the government. On the other hand opposition MPs did not seem to succeed: while in absolute terms they have handed in almost as many proposals as in previous cycles (216 in and 236 in ) only 1 was accepted. This fact points to a hostile governmentopposition relation that is not entirely new to the Parliament though: data in table 2 supports our assumption about the strong polarization of government and opposition since the common introduction of bills is a very rare occasion although it does show a very modest peak in which can be explained by the emergence of a Socialist minority government supported externally by the Liberals who did submit proposals together with the government. 7

8 Table 2: The number and percentage of submitted and passed legislative proposals according to the type of initiator in the different parliamentary cycles Initiator Submitted Passed Submitted Passed Submitted Passed N % from all N % from all % from source N % from all N % from all % from source N % from all N % from all % from source Government Committees gov opp ,5 1 MPs gov.+ opp* ,5 50 all MPs Total Initiator Submitted Passed Submitted Passed Submitted Passed N % N % % N % N % % N % N % % from from from from from from from from from all all source all all source all all source Government Committees gov opp gov.+ opp* MPs all MPs Total * full data are not available for the period Source: Ilonszki-Jáger,2008 and own counting based on the website of the Hungarian Parliament ( 8

9 The data in table 2 suggests that the preventive power of the Hungarian legislature is exercised by the government and while the opposition does have the ability to introduce bills which it indeed does, its ability to succeed is quite limited. Nevertheless, as Cox (2000) has pointed out proposing bills is not the only way to influence legislative output; amendments can considerably change the content of bills which Blondel called reactive power. In the Hungarian legislative process amendments can be initiated by the same actors as the bills themselves, but of course the proposer cannot amend its own bill. If there were amendments proposed during the general debate, they can be adjusted further by amendments to amendments during the debate in detail. Both type of amendments fall under the same regulation, to be considered they need to gain the support of one-third of the members of the designated committee 10, what ensures minority, consequently the opposition a chance to get amendments on the floor. As a third type, amendments before the final vote can be also distinguished. Theoretically, these amendments should fix the incoherence resulting from the different amendments and can only be submitted by the Constitutional Committee and by the proposer of the bill. In practice, the government often uses this tool as last minute adjustment for various reasons. In our analysis we collected all types of amendments in one category, as the opposition has the same rights and chances. Committees have substantial rights in the whole legislative process; they have a detailed discussion about the bills falling in their policy field 11, their report along with the minority committee opinion is discussed in the plenary as well, and as described above they also have an important selective effect on the tabled amendments. Yet their actual power went through a decline. With the strengthening of the divided party system cross-party cooperation decreased even within the committees, what lead to diminished deliberation and predictable majority decisions resembling to that of the plenary voting. Specialized committees tend to favor policy discussion contraire to the battles of politics, where interparty cooperation is fostered (Andeweg Nijzink 1985). As a consequence, strong committees are connected to working parliaments, which have more influence on the legislative output (Norton 1984). Yet, when the bill reaches the final vote the only thing that matters is whether the government is able to gain the support of the majority for passing its bill or not. In the case of majority coalition 10 Except of the amendments proposed by the committees themselves. 11 As the bills often crosscut policy fields, it is quite typical that several standing committees are considering the same bill. 9

10 governments this depends on two factors: PPG voting unity and coalition voting unity, which are both high in the Hungarian setting. To sum up, even if the institutional setup supports the opposition at several stages of the legislative process, due to the stable majoritarian party system disciplined party groups developed dividing the Hungarian political arena into two blocks, what results in strong government dominance of the legislative process and outcome (Ilonszki - Olson 2011a). Theoretical considerations and research hypotheses The opportunities of opposition to prevent, induce or even to influence the legislative process are rather limited, so the puzzle concerning its legislative activity becomes twofold: first, the fundamental question is that in such a restrictive and confrontative setting why do opposition MPs bother at all with introducing bills and amendments? Certainly, introducing bills is not only a valuable activity in terms of legislative capacity but also in terms of visibility and communication since parliamentary activities do have a face value : for parties they reflect the values and preferences of political parties and the alternatives they suggest, while for the MPs they send signals to the constituencies and also to their parties about their loyalty, their work capacity and political aims. But then a second question arises: even if opposition actors seem to bother with parliamentary work, why do they act consensually with the government? What are the factors that motivate opposition MPs or PPGs to vote along with the government that seemingly rejects the idea of cooperation? In order to elaborate further the determinants of consensual voting, first consensus in itself should be defined. As already mentioned above we have to distinguish between two types of bills: those passed by simple majority and those passed by two thirds of the MPs present which are called laws with constitutional force. Thus in terms of consensus for bills with constitutional force a qualified majority is required which generally presupposes a consensus between government and opposition the need for consensus orientation was one of the driving forces behind the introduction of qualified majority bills. In certain cases though the government in itself is capable of legislating upon constitutional affairs as is the case with the 2010 Fidesz-KDNP government that possesses more than two-thirds of parliamentary mandates. In a political approach it is more suitable to define consensus as an agreement 10

11 between the government and opposition. The strict definition of consensus implies total agreement meaning that bills can only be termed as consensual if they were accepted unanimously by all PPGs (or even by all MPs) in the parliament. A broader definition points to the importance of any opposition party agreeing and voting for a bill together with the government while the most permissive understanding of consensus implies that the largest governmental party votes for the given bill with one party of the opposition. While all definitions have their limits, the application of the most permissive definition seems to be appropriate in the Hungarian case since the bipolarisation of politics enhances the importance of any kind of agreement. Thus we regard a bill consensual if it is supported by the largest governmental party and one party of the opposition. This approach however conceals the logic of consensus in certain cases: when the left-right divide does not only separate the opposition and the government but the opposition as well, the consensus between one opposition party and a government party does not necessarily imply a consensus overarching the left-right divide. In order to detect the ideological logic of consensus for each bill the PPGs voting in support are coded in the database. Also the application of the permissive approach impedes the analysis of consensus between coalition partners thus distinct variable is entered into the database for each bill reflecting consensus between coalition partners. When consensus does happen an important political-institutional explanation can be offered by the type of government in charge. Miller and Stecker (2008) have demonstrated that in the German political context where the Budesrat can act as a veto player in the legislation process, parliamentary work is characterized by a more consensual approach. Following Following Hix and Noury (2011) we hypothesize that in case of minority governments the legislative decision-making is characterized by a higher rate of consensus (H1). The rational behind is simple: when a government is not capable of legislating alone, it seeks allies before entering the parliamentary arena. In the Hungarian case the ally itself is worth of analysis, since as we described above it is quite different when the government looks for a potential partner on its side of the political divide or on the other. Thus in order to better understand the functioning of the minority government an analysis of the supporting side is carried out. Among the bill-related explanations first we consider the initiator of the bill, suggesting that the rate of consensus will be lower in case of the governments initiatives (H2). Due to the government s agenda-setting power only those bills introduced by the opposition are scheduled that enjoy the support of the government. As for the government bills, the 11

12 opposition does not have the same preventive power, thus we suppose that they will express their discontent at the time of voting. While the opposition s preventive and initiative power is limited, it is capable of reacting to government bills through amendments. In our third hypothesis (H3) we assume a positive relation between the number of accepted opposition amendments and the level of consensus. Relating to the reaction power of the opposition the legislative process itself is important: the length of debate is supposed to be related to consensual voting since during a longer debate the opposition has more opportunities to influence the content of the bill. According to Hungarian legislation the legislative process can be accelerated through legislation by extraordinary procedure when the bill spend less time in the plenary and the debate time is limited. Through the extraordinary procedure, the time required to pass a law is reduced to one-third. In our fourth hypothesis we suppose a negative relation between the application of an extraordinary procedure and consensual voting (H4). Empirical analysis and results The analysis of the legislative process is based on a roll-call voting database. While available roll-call voting data in the Hungarian Parliament captures voting on the individual level (with the exact number of yes/no/abstention votes), the data is gathered on the party level in our database. The unit of analysis is the bill and our database compromises of 850 accepted bills from four parliamentary cycles including six different governments, although one of them (Gyurcsány II.) only features 10 accepted bills, which is a rather low number of cases. Still as this government is not analyzed in a case-study approach and its characteristics does not differ from the others in terms of independent variables, the government is included in the research In the dataset the characteristics of the critical factors in the legislative process are captured for each bill: the initiator (type of initiator committee, single MP, group of MPs, government; whether it is an opposition initiator), the amendments (the number of all amendments, the number of opposition amendments and the number of accepted opposition amendments), the special procedures applied (whether it was debated in an expedited procedure and whether it was published quickly in the Official Journal) and the final voting (how PPGs voted in favor or rejected -, whether the bill was accepted by consensual voting and whether the coalition voted together). 12

13 In our research we focus on six different governments the Horn government which was a Socialist-Liberal qualified surplus coalition government from ; the first Orbán government which was a right-wing surplus coalition government from ; the first Gyurcsány minimum winnining coalition government from , which was a Socialist- Liberal government that was formed after the resignation of the elected Prime Minister, Medgyessy; the second Gyurcsány government from , again a minimum winning coalition; the third Gyurcsány government that turned into a minority government after the Liberals left the coalition, and due to the resignation of Gyurcsány was reorganized under the Bajnai government in and finally the second Orbán government, which is a qualified surplus coalition again (see table 3). Table 3: Governments in our analysis Election date Date Name PM of Coalition type No of coalition MPs* % of coalition MPs Government Parties Opposition Parties 8 and 29 May Gyula Horn Center- left Qualified surplus majority ,3 MSZP, SZDSZ FIDESZ FKGP KDNP MDF 10 and 24 May Viktor Orbán I Conservative Surplus majority FIDESZ, FKGP MDF MIÉP MSZP SZDSZ na Ferenc Gyurcsány I. Center-left Mimimum winning coalition MSZP-SZDSZ MDF FIDESZ 9 and 23 April, Ferenc Gyurcsány II. Center-left Minimum winning coalition MSZP-SZDSZ FIDESZ KDNP MDF 13

14 na Ferenc Gyurcsány III. Left One party minority MSZP FIDESZ KDNP MDF SZDSZ na Gordon Bajnai Left One party minority MSZP FIDESZ KDNP MDF SZDSZ Conservative FIDESZ Jobbik 11 and 25 April Viktor Orbán II Qualified surplus majority KDNP LMP MSZP * The definition of the current government is a cause of concern. Some claim following Lijphart (1999) that since the two parties in the government (Fidesz and KDNP) are so closely allied this cannot be called a coalitions. Indeed, the KDNP was able to return to parliamentary politics after its failure only when it signed an agreement with the Fidesz concerning electoral cooperation and both in 2006 and 2010 the two parties ran common candidates and electoral lists. Still, we follow the approach that if the executive office is shared by different political parties, and each party has at least one designated representative that enjoys voting rights in the cabinet, we can talk of a coalition (Müller - Bergman - Strom, 2008:6). At the same time the power structure of the large government party and the general political environment clearly shows that the KDNP has an inferior position, the image is that of a sham coalition. Still, in the analysis we regard the government as a coalition. For each observed parliamentary cycle a year in the middle of the cycle (so not consecutive to elections or previous to elections) was chosen as reference year and all accepted bills were coded to the database. As our research interest focuses on the second Fidesz government in detail, the case selection here captured a longer period from the day of entering into office in 2010 to the end of It is important to note that this selection prevented us from comparing absolute numbers since the longer period of the second Fidesz government could 14

15 distort raw data. Thus while Ns might be included in our tables only relative numbers should be taken into account. For each government we collected the final votes of adopted bills of a selected year in the government s lifecycle. Our selected years - as samples for the concerned governments - are 1996, 2000, 2005, 2008, 2009, 2010 and We chose the middle period of the governments office terms in the first three cases, and we added, what includes the end of the Gyurcsány III government and the first half of the Bajnai government as minority governments of 2009 and finally we collected all available information (year 2010 and 2011) for the current Orbán government. The first step of examining consensual decision-making is the mapping of the overall trends of consensus. Data in table 4 confirms the findings of previous research namely that consensual decision-making is an important characteristic of the Hungarian legislative process reaching 62% on average ranging from 42% to 94%. Table 4: Consensual voting in different governments Name of Prime Consensual voting Minister non-consensual consensual Total Horn N % 45,5% 54,5% 100,0% Orbán N % 46,2% 53,8% 100,0% Gyurcsány I N % 58,0% 42,0% 100,0% Gyurcsány II N % 20,0% 80,0% 100,0% Gyurcsány III N % 6,6% 93,4% 100,0% Bajnai N % 16,2% 83,8% 100,0% Orbán II N % 45,6% 54,4% 100,0% Total N % 39,4% 60,6% 100,0% The data in table 4 already suggest that our first hypothesis stands true as minority governments (in bold) have the highest rates of consensual voting. The finding is further strengthened by correlation analysis as the relation between government type and consensual 15

16 voting is significant (Pearsons correlatin index=0,319, significant at the 0,01 level). However, in order to look deeper into the dynamics of minority governments and on the legislative voting model they apply, we have to consider the allies that support governments. First, the analysis of coalition voting reinforces our knowledge about high coalition unity since there are only 16 instances where coalition partners did not vote together. The high coalition unity suggests that we might look for a left-right divide in the voting patterns. The correlation analysis partly confirms this assumption: the bivariate correlation index of support for bills between parties (in pairs) shows a significant positive relation between the voting patterns of PPGs in the same ideological field while it shows a significant negative relation between the voting patterns of PPGs in opposite ideological fields (see annex 2). So for example the two biggest parties from the left and the right, the MSZP and the Fidesz have a correlation index of -0,514 while the Socialist and the Liberal partners, the MSZP and the SZDSZ have a correlation index of 0,512. This positive relation suggests that even in the cases when the Socialist government was in minority it was the Liberals who supported them steadily from the outside. Our findings prove this suggestion as 87% of the consensual bills accepted by minority governments were supported by the Liberals, while only 40% of bills were supported by the biggest opposition party, Fidesz. Thus minority governments simply conserved the old pattern of voting with the ex-coalition partner supporting the government from the outside and the opposition party agreeing only on a limited number of bills. However, the correlation analysis also appoints to an important defect in our assumption namely that the left-right divide does not seem to be decisive in the 2010 parliamentary cycle since neither Jobbik s, nor LMP s voting patterns follow a steady path. The radical right-wing party, Jobbik seems to be the most important outlier: the Pearsons Correlation index is 0,366 with the left-liberal party, LMP and 0,207 with the socialist party, MSZP. Indeed if we look at the rate of consensual voting between Jobbik and the government, we can see that they vote together in 47% of all cases, while LMP votes together with the government in 26% of all cases and MSZP in 14% of all cases. These voting patterns suggest that a cartel model of voting behavior emerged in the Hungarian Parliament with the government controlling the agenda. The model persisted until 2010 when new parties entering the arena tried to break the traditional government-opposition divide and aimed at building an issue-by-issue model of legislative voting. What are the reasons behind this change of attitude? One possible answer is the parties ideological 16

17 composition since Jobbik is on the same side of the left-right scale as the government even if it is at the radical side of it so it seems obvious that there is a wider range of policies where agreement is viable. Polarization indexes support this assumption as while the distance between the MSZP and Fidesz grew, the LMP is situated closer to the Fidesz on a left-right scale and Jobbik being the closest to the government 12. Enyedi and Benoit (2011) also point out that the polarization of Hungarian politics reached its peak before the 2010 elections with the new Parliament introducing a less polarized environment (2011:26-27). The importance of decreasubg polarization is also supported by our evidence since consensual voting prevails in policy-making areas only whereas in the case of 2/3 rd majority bills the consensus-rate drops sharply. While decreasing polarization certainly explains part of the consensus, a puzzle still remains: opposition parties tend to use extra-parliamentary tools to express their discontent with the government more and more often. One striking example is when opposition parties left the parliamentary floor to demonstrate their disagreement with the new Constitution or another when MPs from LMP decided to demonstrate in front of the Parliament. In search of an answer several plausible explanations arise: one the one hand we can explore exogenous factors such as the type of parties that play the parliamentary game. According to Gunther and Diamond the two new parties of the parliament can be characterized as movement parties (2003: ) that engage in direct political action more frequently. Another factor to be taken into consideration is the face-value of voting versus more direct political action: the later is clearly more visible, easier to sell on the media and thus have greater potential to reach and influence voters. It seems plausible that different explanations stand true for the two new opposition parties: while consensus can be ideologically-driven in the case of Jobbik, the LMP might try to use consensus in the legislative arena and protest out of it to build the image of new-politics which is a tactic addressed to voters instead of the government Among the bill-related explanations for consensual voting we first consider the initiator of the bill. Table 5 offers an overview of the accepted bills according to the type of initiators and shows that the analyzed years follow the general pattern of initiation since the majority of bills are introduced by the government. The low number of Private Member bills and committee bills indicates the low initiative power of the Parliament although the second 12 According to Bíró Nagy the PPGs are situated on a left-right scale with 0 being extreme right and 10 being extreme left: MSZP-8,36, LMP-5,86,FIDESZ-KDNP-2,66, JOBBIK-1,61 (2012:3). 17

18 Orbán government is different in this respect since the number of bills introduced as Private Member bills and even more by a group of MPs is on the rise. The fact that single MPs and group of MPs are frequent alternatives to the government signals the government s intention to impose its dominance not only on the Parliament but in a sense on itself by avoiding coordination with ministries and/or trade unions and other political actors. The government type seems decisive in this respect as the coalition-governments and minority governments implied a need to bargain at least with coalition-partners, while the second Fidesz government which is rather a sham coalition does not face intra-government challengers: a unified pattern prevails. It seems thus that while the agenda-setting power can easily overrule opposition minorities, the minor coalition partner position can also be weakened through agenda-control. This finding is further emphasized by the fact that in we found a 100% coalition unity that is extraordinary even in the Hungarian Parliament where coalition unity has been traditionally high. Table 5: The number of bills according to the type of initiator Considering the initiator of the bill, we suggested that the rate of consensus will be lower in case of the governments initiatives (H2). The data confirms this hypothesis as the rate of nonconsensual bills is the highest among government initiated ones reaching 43% while among Private Member bills it is 25% and committee initiated ones it is 38%. While the hypothesis is confirmed it still does not offer an explanation for the consensual behavior of opposition (which does vote for more than half of government initiated bills). An additional explanatory factor might lie in the reactive power of opposition which can amend those bills. 18

19 The third hypothesis is based on the number of accepted opposition amendments suggesting a positive correlation with consensual decision-making. However, the correlation between the number of accepted opposition amendments shows a negative but not significant correlation with consensual voting which is a rather puzzling relationship. One possible explanation might be derived from the overview of the use of amendments in the legislative process: first, only half of the amendments are proposed by the opposition which signals that this tool is as much used by the government as by the opposition with the intention of influencing the legislative outcome. The government s control over the legislative process seems to bloc the amendments as well: the majority of amendments introduced by the opposition does not even reach the floor but is refused at the committee level. Thus the accepted opposition amendments are not numerous or convincing enough to result in consensual voting. The explanation is underlined by the fact that the number of opposition amendments and consensual voting is negatively correlated (Pearsons correlation index=-0,196 significant at the 0,01 level) as well as the number of amendments introduced (Pearsons correlation index=- 0,122 significant at the 0,01 level). Table 6.: The number and mean value of opposition amendments introduced in the observed governments Name of prime minister Mean N Std. Deviation Horn 32, ,11062 Orbán 26, ,23936 Gyurcsány I 10, ,52986 Gyurcsány II 11, ,34242 Gyurcsány III 12, ,37553 Bajnai 7, ,49172 Orbán II 19, ,90716 Total 18, ,38812 Table 6 suggests that parliamentary actors have recognized that introducing amendments is costly while results low benefits since the number of amendments have more or less steadily decreased since the first observed government (Horn government, 1996). The second Fidesz government is an outlier in this respect as the mean value increased which can be explained by the fact that the new opposition parties are in the need of visibility and thus are more active in all legislative dimensions. The activity does not translate directly to success: according to data about the second Fidesz government while the opposition introduces 73% of all 19

20 amendments its share of accepted amendments is only 6% (the government and the committees have the highest success rate: the government introduces 18% of all amendments and it accounts for 41% of all accepted amendments while the committees account for 9% of all introduced amendments but 40% of all accepted ones). Finally the last bill related explanation offered connects the timeframe of the legislative process and the access to the floor (in terms of plenary debate) to consensual voting through the analysis of extraordinary procedure assuming a negative relation. First, we have to note that extraordinary procedures used to be applied quite rarely and it is the second Orbán government that started to use the procedure more frequently: out of the 672 bills accepted by extraordinary procedures 225 took place in the second Orbán government. Thus when we observe the presupposed negative relation between consensual voting and the application of special procedure (Pearsons correlation index=-0,111 significant at the 0,01 level) we mainly reflect upon the recent government activities. Conclusion Our research has reinforced the well-known fact that opposition parties are rather suppressed in the legislative process, in general majority decisions taken by the governing parties do not take their preferences into account. Hence their chance to influence legislative outcome is very low in the case of simple majority bills and only turns decisive in case of qualified majority bills. However, the power of opposition to influence qualified majority bills does not translate to a decrease in consensual decision-making in case of qualified majority but rather manifests itself in a decreasing number of qualified decision-making (until 2010). This trend suggests that a major filtering effect as the government does not table bills that the opposition won t approve in order to optimize the legislative capacity of the parliament. This filtering effect suggests that in the Hungarian case the preventive power of parliament is exercised exclusively by the government who is only willing to consider initiatives that have been agreed on previously. This finding suggests that in order to understand the legislative process research should look for explanatory factors out of the process itself and turn attention to the critical factors shaping the pre-legislative negotiations. The fact that the majority of bills are initiated by the government or by government MPs suggests that the power of initiation is also dominated by governmental actors. Together with 20

21 a weak reactive power the opportunity structure of opposition is weak in the institutional as well as in the political sense. Until 2010 these phenomena resulted in a cartel model of legislative voting with the government and the opposition playing their role in a conflictive environment. In this period consensual decision-making seems to be motivated by factors outside of the legislative process. One factor that has explanatory power in the model we investigated is the type of government although deeper analysis discovered that minority governments did not introduce new patterns of voting behaviour as the decisive role of the left-right divide was maintained. In Hungary the party blocks vote together even if they do not share government positions and as a consequence minority governments only open legislative opportunities to parties in their own block who are willing to support them from the outside. Throughout the analysis distinctive features of the current, Orbán II government emerged. With the government controlling for 68% of seats the need to cooperate with the opposition is minimal and it seems that political will does not make up for the lack of institutional pressures. While the current opposition tried to break the strictly adversarial relationship between the government and the opposition by introducing a consensus-oriented attitude based on an issue-by-issue legislative model, the government does not seem to accommodate this behaviour. As a result the two new parties are looking for new ways to build their image: the LMP tries to use its grass-root features to motivate direct communication, while Jobbik struggles with opposing the government from the same side of the political spectrum slowly overwriting the long-standing left-right divide. The increasing opposition activities taking place outside of the parliamentary arena are a sign of a weakening institution. Minority rights are important not only to enable the opposition to influence decision-making and to control the government but also to channel disagreement towards the floor where conflicts can be played out (even if not always solved). This is exactly what the Hungarian Parliament has lost: it has lost its capacity to manage political conflict. This has a destabilizing effect as well as it results in poor political performance. References Andeweg, R. B. Nijzink, H. W. (1985): The Netherlands: Cabinet Committees in a Coalition Cabinet. In: T. Mackie - B. Hogwood eds.: Unlocking the Cabinet; cabinet structures in comparative perspective, London/Beverly Hills/New Delhi: Sage 21

22 Blondel, Jean (1970): Legislative behavior: some steps towards a Cross-National Measurement. In: Government and Opposition, Vol. 5. No. 1, pp Cox, Gary W. (2000): On the Effects of Legislative Rules. In: Legislative Studies Quarterly, XXV/2, pp Dalton, Russel J. David M. Farrell Ian McAllister (2011) Political Parties and Democratic Linkage. How Parties Organize Democracy. Oxford University Press Döring, Herbert (1995a): Fewer Though Presumably More Conflictual Bills: Parliamentarry Government as a Monopolist. In Döring, Herbert ed.: Parliaments and Majority Rule in Western Europe. Berlin: Campus Verlag. Döring, Herbert (1995b): Time as a Scarce Resource: Government Control of the Agenda. In Döring, Herbert ed.: Parliaments and Majority Rule in Western Europe. Berlin: Campus Verlag. Döring, Herbert and Mark Hallerberg eds (2004): Patterns of Parliamentary Behavior Passage of Legislation Across Western Europe. Ashgate: Aldershot Enyedi, Zsolt Fernando Casal Bértoa (2011) Patterns of Party Competition in: Paul Lewis and Radoslaw Markowski eds. Europeanising Party Politics? Comparative Perspectives on Central and Eastern Europe after Enlargement. Manchester UP Enyedi Zsolt Kenneth Benoit (2011): Kritikus választás A Magyar pártrendszer átrendeződése a bal-jobb dimenzióban. in: Enyedi Zsolt Tardos Róbert Szabó András (eds): Új képlet. A 2010-es választások Magyarországon. DMKA, Budapest, pp Fiorina, Morris (1997): Congress: Keystone of the Washington establishment New Haven: Yale University Press Gunther, Richard Larry Diamond (2003): Species of Political Parties. A New Typology. Party Politics 9(2): Hix, Simon and Abdul Noury (2011) Government Opposition or Left-Right? The Institutional Determinants of Voting in Legislatures. paper accessed on July 10, 2011 Ilonszki, Gabriella Olson, David M. (2011a): Questions about Legislative Institutional Change and Transformation in Eastern and East Central Europe: Beyond the Initial Decade. In: The Journal of Legislative Studies, Vol 17, No.2, pp