Same Species, Different Breed: The Conditional Effect of Legislator Activities in Parliament on Re-Selection in a Mixed-Member Electoral System

Same Species, Different Breed: The Conditional Effect of Legislator Activities in Parliament on... Abstract The article investigates whether or not the amount of work legislators carry out in parliament affects their chances of re-selection in a mixed-member electoral system. A unique dataset collecting Hungarian Members of Parliaments’ (MPs) publicly available electoral, socio-demographic and parliamentary activity data between 1998 and 2010 is analysed. The study concludes that activity in parliament positively influences re-selection chances when selectors decide on party list nominations. Additionally, MPs whose parties do not count on as future Single Member District (SMD) candidates benefit from parliamentary work to a greater extent than prospective SMD candidates. At the same time, extra work in parliament does not bring MPs closer to SMD nominations. Results confirm that selectors evaluate MPs along how well they carry out the tasks dedicated to their groups. Students of legislators’ behaviour differ regarding the activities they study, the factors with which they explain the variance in these activities, the consequences which they attribute to the activities and the methods they apply to uncover causal relationships. However, the majority of authors seems to agree with Mayhew’s oft-cited starting point, and considers Members of Parliament (MPs) as ‘single-minded seekers of re-election’ (Mayhew, 1974). Consequently, most of the things representatives do in the present are attributed to the MPs’ individual desire for future re-election. At the same time, students of candidate selection point out that under centralised selection rules and with a high number of safe seats, emphasis shifts from re-election to re-selection (Hazan and Rahat, 2010; Atmor et al., 2011). Additionally, as Strøm (1997) argues, MPs act strategically to reach their goals in the hierarchical order of their feasibility: in order to be re-elected, one has to be re-selected first. The situation becomes even more complicated in mixed-member electoral systems when parties may nominate candidates in Single Member Districts (SMDs) and at the same time closed party lists. In the first case, re-selection is dependent upon the potential for re-election, inviting strategies that balance between promoting one’s ability to attract extra votes and advertising party loyalty (see also Russo, 2011). In the second case, due to the inability of the voters to change the order of candidates on the list, the vote gathering potential of individual MPs is negligible, thus loyalty will be enforced at the stage of candidate selection. Relatively little attention has been paid to the extent of parliamentary work and its effects on the MPs’ electoral fortunes. When developing the portfolio of their activities, balancing between constituency and parliamentary work, legislators face a strategic decision. On the one hand, constituency work earns more visibility among constituents, thus allocating resources to it is desirable for vote gathering purposes. On the other hand, parliamentary work must be done for a functioning parliament without any substantial electoral benefit in terms vote-seeking, which invites only little effort from MPs. From the viewpoint of the parties, however, both are essential. Parties need both MPs who carry out constituency service—and through that stabilise their support in the district, and keep voters satisfied—and MPs who are active in parliament independent of their constituency ties—speak on the floor, submit questions and sponsor bills. The starting point of the article is that selectors aim at selecting representatives who are expected to perform best in the task dedicated to their positions. Very simply put, in a mixed-member context, list candidates are expected to focus more on parliamentary work than SMD candidates, and parties may use prior performance in parliament to evaluate the MPs’ future potential. In this study the focus is on re-selection chances in a mixed-member electoral system. More specifically, I askif there is a connection between the volume of activities legislators carry out in parliament and their candidacy at the next elections. In other words: are MPs encouraged to working harder in order to reach their goals, and make the first steps toward re-election? I select Hungary as case and analyse electoral data from 1998 to 2010. The article finds that activity in parliament positively influences re-selection chances when selectors decide on party list nominations. Additionally, MPs whose parties do not count on as future SMD candidates benefit from parliamentary work to a greater extent than prospective SMD candidates. Furthermore, extra work in parliament does not bring MPs closer to SMD nominations. Results confirm that selectors evaluate MPs along how well they carry out the tasks dedicated to their group. 1. Theoretical argument There are many factors on various levels (i.e. macro, legal, institutional, party) that influence what types of candidates are selected (Siavelis and Morgenstern, 2008). We may distinguish between formal party rules and informal norms that guide the implementation of these rules (Helmke and Levitsky, 2004). The overwhelming majority of the literature on candidate selection deals with the former, especially its centralisation and democratisation (see e.g. Field and Siavelis, 2008; Hazan and Rahat, 2010; Atmor et al., 2011). Lately, little attention has been paid to candidate characteristics and activities that potentially increase the likelihood of selection, and that we may also consider the informal drives of candidate selection. The question here is, what kinds of candidates do parties select and why? It is safe to say that candidates are not equal: some are more likely to be selected than others (Katz, 2001). Similarly, one can assume that legislators are also not identical in terms of re-selection chances. The most important aspect of candidate selection (besides its formal rules) is the candidate’s loyalty to the selectorate (i.e. the body or collections of individuals that selected the candidate), which may also be the central determinant of candidate behaviour (Siavelis and Morgenstern, 2008). Candidate strength is an additional strong determinant of nominations: parties tend to select locally (Gallagher, 1980) or nationally (Siavelis, 2002) well-known politicians, like incumbents, local politicians, ministers or party leaders. Parties may even create rules with respect to seniority, incumbency (Helmke and Levitsky, 2004) or gender (Krook, 2010). As candidacy also depends on candidate behaviour, once candidates are elected, (s)election rules continue to affect the extent to which they think about re-selection, which in turn will affect their decisions about behaviour in parliament (Siavelis and Morgenstern, 2008). Thus, legislators who want to re-enter the electoral arena have to make strategic decisions about how they manage their time and other resources when in office. It has been shown on many occasions that in a mixed-member electoral system, MP behaviour differs across mandate type. Because of the diverging mechanisms of accountability, SMD MPs are more likely to engage in district representation, while list MPs are more party-centred (Lancaster and Patterson, 1990; Gallagher and Holliday, 2003; Heitshusen et al., 2005; Curtice and Shively, 2009). SMD MPs may engage in two types of activities: work in the constituency and work in parliament (Glazer and Wattenberg, 1996). As to the former, they hold office hours, carry out ombudsman-like activities and deal with problems arising in the district as well as respond to individual petitions (Fenno, 1978; Cain et al., 1987; Norton and Wood, 1990; Norris, 1997). Work in parliament consists of sponsoring bills, drafting questions or speaking on the floor. To maximise the share of votes in the SMDs at the next election, SMD MPs are expected to carry out constituency service, which leaves less space for parliamentary work. However, MPs do not necessarily need to trade-off between constituency and parliament work. The local focus of bill initiation (Crisp and Ingall, 2002; Marangoni and Tronconi, 2011), membership in certain committees (Stratmann and Baur, 2002; Manow, 2013), speeches (Hill and Hurley, 2002), parliamentary questioning (Martin, 2011a) and the willingness to defect the party line at roll-call (Carey, 2007; Tavits, 2009) can reflect general concerns about issues that are potentially important for the constituency, bringing parliamentary work closer to constituency service. Nevertheless, constituency service on location is always more visible than in parliament, and it also creates greater satisfaction for MPs with their jobs (Norton, 2002), which provides greater incentives to upgrade their constituency presence and economise on parliament work. Also, constituency work is more highly evaluated in countries where work in parliament provides little electoral benefit (Martin, 2011b), because of—for instance—the poor visibility of parliamentary work. As parliamentary work is not essential to a job well done in the case of SMD MPs, selectors are expected not to take parliamentary activities into account at SMD nominations. H1: Selectors do not take prior parliamentary activity into account when making the decision about SMD nominations. Contrarily, as list MPs do not have office-related obligations towards the constituencies, they should be focusing more on parliamentary activities, tasks within the party organisation and representing the party in public. As SMD MPs are more occupied with constituency work, the larger part of the tasks in parliament naturally falls onto list MPs. Although parliamentary work may not result in direct electoral benefits such as extra votes gathered by direct constituency work, the mere intensity with which MPs engage in parliamentary activities may work as a simple proxy of their efforts to establish the reputation of a hard-working representative, who is willing to do the ‘legwork’ without any clear compensation. Parties gain important information on legislators simply by looking at their track records, without having even to look at the content of their activities. Bill sponsorship, (Bowler, 2010; Bräuninger et al., 2012), the frequency of floor speeches (Anderson et al., 2003) and tabling questions (Lazardeux, 2005; Rasch, 2009) or simple attendance indicate that they work just as hard—or harder—as other members. Nevertheless, list MPs are also not alike from at least one important perspective: their prior record as SMD candidates. As list MPs are also very often SMD candidates at the previous elections, parties may have different plans for them than serving just as ‘second class representatives’ (Ward, 1998). It is very likely that they are the future SMD candidates of the party, not to mention that they can work as ‘shadows’ to SMD MPs (Papp, 2016), which helps maintain the connection between the party and the constituency. Therefore, list MPs who were nominated in an SMD at the previous elections are expected to engage in extra activities, such as constituency work. For the rest of the list MPs, parliamentary work is the major task. H2: At party list nominations, parliamentary activities play a greater role in the case of MPs formerly not nominated in SMDs, than in the case former SMD candidates. 2. The Hungarian case During the period under investigation (1998–2010), Hungary has a three-tier electoral system with unlinked electoral tiers. A total number of 386 legislators are elected from 176 SMDs (SMD tier), 20 regions and the national level (regional and national list tiers). Both list tiers apply closed party lists, and—especially the national tier—are designed to compensate the often disproportional results of the SMD tier. Parties may nominate candidates on multiple electoral tiers at the same time. This practice is quite widespread in Hungary: between 1990 and 2010, almost 40% of the candidates were nominated on more than one tier. In the Hungarian parliament, House rules entitle individual legislators to sponsor bills, make brief speeches on the plenary during debates and submit questions. There are four types of questions in the Országgyűlés: interpellations, written, oral and direct questions. These types differ with regards to the form of the response as well as party control. Interpellations, oral and direct questions must be answered at the plenary session, while written questions require written response. With the exception of direct questions, all questions have to be submitted in writing prior to the session. With regards to party control over the frequency with which MPs may engage in the different types of activities, generally speaking, activities that have time limits, and may receive more publicity are controlled by the PPG-leadership to a greater extent than those not limited either in time or in numbers. In the Hungarian parliament, MPs may submit as many written questions as they want, while oral forms (including speeches, interpellations, direct and oral questions) are strictly limited by House rules applying quite restrictive time constraints. Consequently, written question are subject to less party control than oral activities. Hungarian politics was characterised by the competition of two blocks lead by MSZP1 and Fidesz2 until the 2010 elections, when two new parties (LMP and Jobbik) emerged and MSZP went into a dramatic decline. In the case of the elections under investigation here, the majority of SMD seats went to either Fidesz or MSZP. Candidate selection in Hungarian parties is not a well-studied topic. There is relatively little hard evidence on party strategies, especially in a longitudinal fashion. However, from the little that we have, we may re-construct candidate selection as being organised by the party centres. This is no surprise as Field and Siavelis (2008) argue that in countries after transition selection of candidates is likely to become centralised by the reason of political uncertainty, lower party organisation and strategically complex electoral systems. In Hungary the party centres select, interview and/or approve candidates (Marjai, 2012, p. 48) either on the regional or on the national scale. There are several additional circumstances that—despite the existence of the SMD tier—make Hungarian politics party centred. These characteristics are hardly unique to Hungary. Along with centralised candidate selection, strong partisan attachment of voters (Enyedi and Tóka, 2007),3 increasing polarisation (Enyedi and Benoit, 2011; Körösényi, 2013) and strong party discipline (Ilonszki and Jáger, 2008; Ilonszki, 2012)4 all characterise new democracies. In this sense, as sharing the same logic, the Hungarian case may be considered as a typical case for new democracies. As to mixed member electoral systems, if hypotheses derived from the literature on established democracies can be verified using a new democracy as a case, it is to be expected that results hold for older democracies as well. 3. Data and variables The analysis builds upon a unique dataset that collects Hungarian MPs’ publicly available electoral, socio-demographic and parliamentary activity data. The data cover three consecutive parliaments from 1998 to 2010. As we follow a relatively large number of individuals (large N) throughout a relatively short period of time (small T), the data can be handled as panel. Since there are also MPs who did not participate in every single election under investigation, the panel is fairly unbalanced. Throughout this study t will denote the year of re-selection (2002, 2006 and 2010). MPs are elected at election t−1 (1998, 2002 and 2006), serve between t−1 and t and are re-selected (or not) at t. The dataset includes 1216 observations for 683 individual MPs. Three hundred and twenty-two MPs only served once during the three electoral terms under investigation, 189 served twice and 172 three times. 3.1 Dependent variables Throughout the analysis I use two dependent variables, which measure whether or not the MP was re-selected in an (1) SMD and on a (2) party list at election t. Seventy-three percent of MPs running at the next election was nominated in one of the constituencies, while 80.4% of the MPs secured a position on the party lists. 3.2 Independent variable: activity in parliament As outlined earlier, MPs may engage in a wide range of activities. They attend plenary sessions; take the floor and make speeches, sponsor bills, submit and table various types of questions. With the exception of attendance, information on all these activities is available on the official website of the Hungarian parliament. Utilizing the data available, I create a variable measuring average MP activity by taking the mean value of the absolute number of bills sponsored by the MP, the number of questions and the number of speeches within the respective electoral term.5 Although these activities may not be comparable in terms of their costs (time, human resource, etc. requirements), neither the level of control by the parties, an average value indicates how much the MP makes herself visible in the parliamentary work.6 3.3 Control variables To obtain the net effect of parliamentary activities on re-selection, additional variables must be controlled for. As SMD mandates are more prestigious positions in the Hungarian parliament, I expect that SMD MPs have a greater chance of re-selection than list MPs. Thus, I include mandate type obtained at election t−1. Additionally, I also account for tenure as the number of electoral terms served as an MP at the time of election t. Taking the low levels of legislative turnover into consideration (Ilonszki and Edinger, 2007), I assume that a more extensive legislative experience will increase the chances of re-nomination. Filling in various political positions is also expected to help MPs getting re-selected. I test if mayors, ministers, party leaders, parliamentary office holders and parliamentary committee chairs have better chances in the race for re-nomination. Including these variables (all dummies) is also key because it is expected that these MPs may take the floor more frequently than backbench MPs, thus controlling only for the effect of parliamentary activities could absorb the effect of the position variables, which would result in a significant effect for activities when there is none.7 Hungarian electoral rules permit multiple candidacy, namely that the same individual may be nominated at different tiers of the electoral system. As multiple candidacy may function as a ‘safety net,’ parties may nominate important candidates on more tiers at the same time to maximise election chances. Thus, simultaneous candidacy on multiple tiers is expected to correlate with re-selection in an SMD as well as on a party list. Moving on, taking the party effect into account regarding the three elections under investigation is difficult by the reason of changing electoral coalitions.8 However, to get as close to the party differences as possible I include two variables into the models: the dominant party effect captures the difference between the two dominant parties at the time (MSZP and Fidesz) and all the others,9 while the government effect accounts for the differences parties may have in their selection strategies by the reason of their government versus opposition status. The unmeasured variable, the MP’s position within the party, affects both re-selection chances and the frequency with which the party allows the MP to take the floor in parliament. Assuming a strong correlation between the MP’s position within the party and her position on the party lists, with controlling for party list positions at the current election (as a proxy for the MP’s position within the party),10 I hope to mitigate the problem of endogeneity in the model caused by the interrelatedness of parliamentary activity and the MP’s position within the party, the effect of which would otherwise appear in the error term.11 Furthermore, I control for the lagged dependent variables to see if SMD and party list candidacy at election t−1 affects candidacy at election t. And finally, I also include a time variable to capture the differences between the three elections. 4. Results As the dataset has panel properties, to model the effect of activity in parliament, I use mixed-effects regressions. SMD and party list candidacy being binary variables, mixed effects logit regression is the proper choice.12 In all models, standard errors are clustered around individual MPs. I only discuss effects that are significant on a 5% level. Table 1 displays the results of the regression models explaining whether or not the MP was nominated in an SMD at election t. Table 1 Mixed-effects binary logit regressions explaining SMD nominations at the next elections Model 1 Model 2 Model 3 Activity in parliament 0.002 (0.002) −0.000 (0.003) 0.000 (0.003) Former candidacy  SMD candidate (t−1) 1.623 (0.235)*** 2.307 (0.345)*** 1.821 (0.280)***  List candidate (t−1) 0.034 (0.216) −0.611 (0.366)* 0.308 (0.299) Control variables  SMD MP (t) 1.604 (0.295)*** 0.447 (0.443) 1.529 (0.354)***  Regional list MP (t) −0.336 (0.229) −0.897 (0.322)*** 0.397 (0.277)  Tenure (t) −0.392 (0.100)*** −0.344 (0.141)** −0.361 (0.116)***  Mayor (t) 1.065 (0.341)*** 0.265 (0.436) 0.638 (0.550)  Minister (t) 1.541 (0.508)*** 13.265 (614.290) 0.346 (0.604)  Party leader (t) −0.034 (0.203) −0.446 (0.304) −0.379 (0.242)  Parliamentary positions (t) 0.318 (0.407) 0.074 (0.665) −0.165 (0.435)  Committee chair (t) 0.264 (0.197) 0.109 (0.298) −0.169 (0.245)  Dominant party (t) −0.204 (0.240) −1.557 (0.483)*** −1.203 (0.365)***  Government party (t) −0.062 (0.172) 0.785 (0.270)*** 0.466 (0.237)**  Regional list candidate (t) 2.005 (0.266)***  National list candidate (t) 0.114 (0.203)  Relative position on the regional list (t) −4.340 (0.823)***  Relative position on the national list (t) −0.469 (0.433)  Time −0.146 (0.035)*** 0.025 (0.041) −0.009 (0.037)  Constant 290.751 (71.903)*** −46.646 (83.265) 18.293 (75.080) Random effects parameter  SD of the MP-level intercept 0.807 (0.389) 0.355 (0.981) 0.001 (4.615)  N 1216 633 547  Wald Chi2 163.84*** 74.64*** 118.48***  Log likelihood −572.75 −228.428 −261.029 Model 1 Model 2 Model 3 Activity in parliament 0.002 (0.002) −0.000 (0.003) 0.000 (0.003) Former candidacy  SMD candidate (t−1) 1.623 (0.235)*** 2.307 (0.345)*** 1.821 (0.280)***  List candidate (t−1) 0.034 (0.216) −0.611 (0.366)* 0.308 (0.299) Control variables  SMD MP (t) 1.604 (0.295)*** 0.447 (0.443) 1.529 (0.354)***  Regional list MP (t) −0.336 (0.229) −0.897 (0.322)*** 0.397 (0.277)  Tenure (t) −0.392 (0.100)*** −0.344 (0.141)** −0.361 (0.116)***  Mayor (t) 1.065 (0.341)*** 0.265 (0.436) 0.638 (0.550)  Minister (t) 1.541 (0.508)*** 13.265 (614.290) 0.346 (0.604)  Party leader (t) −0.034 (0.203) −0.446 (0.304) −0.379 (0.242)  Parliamentary positions (t) 0.318 (0.407) 0.074 (0.665) −0.165 (0.435)  Committee chair (t) 0.264 (0.197) 0.109 (0.298) −0.169 (0.245)  Dominant party (t) −0.204 (0.240) −1.557 (0.483)*** −1.203 (0.365)***  Government party (t) −0.062 (0.172) 0.785 (0.270)*** 0.466 (0.237)**  Regional list candidate (t) 2.005 (0.266)***  National list candidate (t) 0.114 (0.203)  Relative position on the regional list (t) −4.340 (0.823)***  Relative position on the national list (t) −0.469 (0.433)  Time −0.146 (0.035)*** 0.025 (0.041) −0.009 (0.037)  Constant 290.751 (71.903)*** −46.646 (83.265) 18.293 (75.080) Random effects parameter  SD of the MP-level intercept 0.807 (0.389) 0.355 (0.981) 0.001 (4.615)  N 1216 633 547  Wald Chi2 163.84*** 74.64*** 118.48***  Log likelihood −572.75 −228.428 −261.029 Notes: Entries are logistic regression coefficients with cluster robust standard errors in parentheses. Standard errors are clustered around individual MPs. *** p < 0.01, ** p < 0.05, * p < 0.1. Table 1 Mixed-effects binary logit regressions explaining SMD nominations at the next elections Model 1 Model 2 Model 3 Activity in parliament 0.002 (0.002) −0.000 (0.003) 0.000 (0.003) Former candidacy  SMD candidate (t−1) 1.623 (0.235)*** 2.307 (0.345)*** 1.821 (0.280)***  List candidate (t−1) 0.034 (0.216) −0.611 (0.366)* 0.308 (0.299) Control variables  SMD MP (t) 1.604 (0.295)*** 0.447 (0.443) 1.529 (0.354)***  Regional list MP (t) −0.336 (0.229) −0.897 (0.322)*** 0.397 (0.277)  Tenure (t) −0.392 (0.100)*** −0.344 (0.141)** −0.361 (0.116)***  Mayor (t) 1.065 (0.341)*** 0.265 (0.436) 0.638 (0.550)  Minister (t) 1.541 (0.508)*** 13.265 (614.290) 0.346 (0.604)  Party leader (t) −0.034 (0.203) −0.446 (0.304) −0.379 (0.242)  Parliamentary positions (t) 0.318 (0.407) 0.074 (0.665) −0.165 (0.435)  Committee chair (t) 0.264 (0.197) 0.109 (0.298) −0.169 (0.245)  Dominant party (t) −0.204 (0.240) −1.557 (0.483)*** −1.203 (0.365)***  Government party (t) −0.062 (0.172) 0.785 (0.270)*** 0.466 (0.237)**  Regional list candidate (t) 2.005 (0.266)***  National list candidate (t) 0.114 (0.203)  Relative position on the regional list (t) −4.340 (0.823)***  Relative position on the national list (t) −0.469 (0.433)  Time −0.146 (0.035)*** 0.025 (0.041) −0.009 (0.037)  Constant 290.751 (71.903)*** −46.646 (83.265) 18.293 (75.080) Random effects parameter  SD of the MP-level intercept 0.807 (0.389) 0.355 (0.981) 0.001 (4.615)  N 1216 633 547  Wald Chi2 163.84*** 74.64*** 118.48***  Log likelihood −572.75 −228.428 −261.029 Model 1 Model 2 Model 3 Activity in parliament 0.002 (0.002) −0.000 (0.003) 0.000 (0.003) Former candidacy  SMD candidate (t−1) 1.623 (0.235)*** 2.307 (0.345)*** 1.821 (0.280)***  List candidate (t−1) 0.034 (0.216) −0.611 (0.366)* 0.308 (0.299) Control variables  SMD MP (t) 1.604 (0.295)*** 0.447 (0.443) 1.529 (0.354)***  Regional list MP (t) −0.336 (0.229) −0.897 (0.322)*** 0.397 (0.277)  Tenure (t) −0.392 (0.100)*** −0.344 (0.141)** −0.361 (0.116)***  Mayor (t) 1.065 (0.341)*** 0.265 (0.436) 0.638 (0.550)  Minister (t) 1.541 (0.508)*** 13.265 (614.290) 0.346 (0.604)  Party leader (t) −0.034 (0.203) −0.446 (0.304) −0.379 (0.242)  Parliamentary positions (t) 0.318 (0.407) 0.074 (0.665) −0.165 (0.435)  Committee chair (t) 0.264 (0.197) 0.109 (0.298) −0.169 (0.245)  Dominant party (t) −0.204 (0.240) −1.557 (0.483)*** −1.203 (0.365)***  Government party (t) −0.062 (0.172) 0.785 (0.270)*** 0.466 (0.237)**  Regional list candidate (t) 2.005 (0.266)***  National list candidate (t) 0.114 (0.203)  Relative position on the regional list (t) −4.340 (0.823)***  Relative position on the national list (t) −0.469 (0.433)  Time −0.146 (0.035)*** 0.025 (0.041) −0.009 (0.037)  Constant 290.751 (71.903)*** −46.646 (83.265) 18.293 (75.080) Random effects parameter  SD of the MP-level intercept 0.807 (0.389) 0.355 (0.981) 0.001 (4.615)  N 1216 633 547  Wald Chi2 163.84*** 74.64*** 118.48***  Log likelihood −572.75 −228.428 −261.029 Notes: Entries are logistic regression coefficients with cluster robust standard errors in parentheses. Standard errors are clustered around individual MPs. *** p < 0.01, ** p < 0.05, * p < 0.1. The results of Model 1 indicate no connection between the intensity of parliamentary activities and the probability of re-selection in SMD, which confirms H1. Nevertheless, the distinguished role of incumbency is observed, as being a constituency representative increases the probability of SMD nomination by 0.39.13 Additionally, not only SMD candidacy at the previous elections boosts the likelihood of re-selection by quite a large margin (0.46), but also mayors and ministers find it easier to be re-nominated. With regards to multiple candidacy, MPs who are selected also on regional party lists become SMD candidates with a significantly greater probability. The same cannot be said for national list candidates, which indicates that parties use regional party lists as ‘safety net’ for SMD candidates in case they fail to bring in the district. This is further supported when relative list positions are included into Model 1. Here, we find (see Models 2 and 3) that candidates higher on the regional list (i.e. taking a lower value on the relative list position variable) are significantly more likely to be selected as SMD candidates. The difference in the probabilities of selection for two MPs at the two extremes of the relative list positions variable is 0.54. Although there is no significant effect attached to relative positions on the national party list, its negative sign reveals similar tendencies. Last but not least, contrary to the expectations tenure has a significant negative effect on SMD nominations. Although parties tend to ‘stick to their people’ in the given districts (see the effect of SMD incumbency), at the same time they seem to give an advantage to less experienced MPs as well. The difference in probabilities between the least and most experienced MPs is 0.24 which suggests a considerable disadvantage for the most senior legislators, and is probably a consequence of the retirement of the most experienced MPs. Turning to party list nominations, we see that parliamentary activities play a significant role (see Table 2, Model 4). The effect size, however, is quite small. A 10 unit change in the activity variable increases the chance of list nomination by 0.014. Additionally, the probability of selection on a party lists starts from a quite high level: even with an activity score of 0, MPs have a 0.78 chance of reselection on one of the party lists. With an average level of activity, MPs are able to bring up the re-selection probability by only 0.035. This indicates that MPs have to work exceptionally hard to improve their re-selection chances. Table 2 Mixed-effects binary logit regressions explaining nomination on party lists at the next elections Model 4 Model 5 Activity in parliament 0.014 (0.004)*** 0.043 (0.012)*** Former candidacy  SMD candidate (t−1) −0.603 (0.303)** −0.109 (0.334)  List candidate (t−1) 0.812 (0.293)*** 0.783 (0.287)*** Interaction term  Activity * SMD candidate (t−1) −0.035 (0.012)*** Control variables  SMD MP (t) 0.070 (0.367) 0.036 (0.358)  Regional list MP (t) 0.307 (0.283) 0.320 (0.281)  Tenure (t) 0.020 (0.117) 0.046 (0.115)  Mayor (t) −0.299 (0.408) −0.287 (0.406)  Minister (t) −0.877 (0.618) −1.027 (0.620)  Party leader (t) 0.589 (0.296)** 0.586 (0.292)**  Parliamentary positions (t) 0.734 (0.594) 0.784 (0.593)  Committee chair (t) 0.489 (0.280)* 0.475 (0.277)*  Dominant party (t) 1.944 (0.395)*** 1.937 (0.398)***  Government party (t) −0.604 (0.235)** −0.502 (0.232)**  SMD candidate (t) 4.558 (0.631)*** 4.516 (0.630)***  Time −0.104 (0.043)** −0.112 (0.040)**  Constant 206.317 (87.32)** 222.612 (89.279)** Random effects parameter  SD of the MP-level intercept 1.069 (0.516) 0.989 (0.541)  N 1216 1216  Wald Chi2 70.99*** 69.38***  Log likelihood −370.870 −365.36 Model 4 Model 5 Activity in parliament 0.014 (0.004)*** 0.043 (0.012)*** Former candidacy  SMD candidate (t−1) −0.603 (0.303)** −0.109 (0.334)  List candidate (t−1) 0.812 (0.293)*** 0.783 (0.287)*** Interaction term  Activity * SMD candidate (t−1) −0.035 (0.012)*** Control variables  SMD MP (t) 0.070 (0.367) 0.036 (0.358)  Regional list MP (t) 0.307 (0.283) 0.320 (0.281)  Tenure (t) 0.020 (0.117) 0.046 (0.115)  Mayor (t) −0.299 (0.408) −0.287 (0.406)  Minister (t) −0.877 (0.618) −1.027 (0.620)  Party leader (t) 0.589 (0.296)** 0.586 (0.292)**  Parliamentary positions (t) 0.734 (0.594) 0.784 (0.593)  Committee chair (t) 0.489 (0.280)* 0.475 (0.277)*  Dominant party (t) 1.944 (0.395)*** 1.937 (0.398)***  Government party (t) −0.604 (0.235)** −0.502 (0.232)**  SMD candidate (t) 4.558 (0.631)*** 4.516 (0.630)***  Time −0.104 (0.043)** −0.112 (0.040)**  Constant 206.317 (87.32)** 222.612 (89.279)** Random effects parameter  SD of the MP-level intercept 1.069 (0.516) 0.989 (0.541)  N 1216 1216  Wald Chi2 70.99*** 69.38***  Log likelihood −370.870 −365.36 Notes: Entries are logistic regression coefficients with cluster robust standard errors in parentheses. Standard errors are clustered around individual MPs. *** p < 0.01, ** p < 0.05, * p < 0.1. Table 2 Mixed-effects binary logit regressions explaining nomination on party lists at the next elections Model 4 Model 5 Activity in parliament 0.014 (0.004)*** 0.043 (0.012)*** Former candidacy  SMD candidate (t−1) −0.603 (0.303)** −0.109 (0.334)  List candidate (t−1) 0.812 (0.293)*** 0.783 (0.287)*** Interaction term  Activity * SMD candidate (t−1) −0.035 (0.012)*** Control variables  SMD MP (t) 0.070 (0.367) 0.036 (0.358)  Regional list MP (t) 0.307 (0.283) 0.320 (0.281)  Tenure (t) 0.020 (0.117) 0.046 (0.115)  Mayor (t) −0.299 (0.408) −0.287 (0.406)  Minister (t) −0.877 (0.618) −1.027 (0.620)  Party leader (t) 0.589 (0.296)** 0.586 (0.292)**  Parliamentary positions (t) 0.734 (0.594) 0.784 (0.593)  Committee chair (t) 0.489 (0.280)* 0.475 (0.277)*  Dominant party (t) 1.944 (0.395)*** 1.937 (0.398)***  Government party (t) −0.604 (0.235)** −0.502 (0.232)**  SMD candidate (t) 4.558 (0.631)*** 4.516 (0.630)***  Time −0.104 (0.043)** −0.112 (0.040)**  Constant 206.317 (87.32)** 222.612 (89.279)** Random effects parameter  SD of the MP-level intercept 1.069 (0.516) 0.989 (0.541)  N 1216 1216  Wald Chi2 70.99*** 69.38***  Log likelihood −370.870 −365.36 Model 4 Model 5 Activity in parliament 0.014 (0.004)*** 0.043 (0.012)*** Former candidacy  SMD candidate (t−1) −0.603 (0.303)** −0.109 (0.334)  List candidate (t−1) 0.812 (0.293)*** 0.783 (0.287)*** Interaction term  Activity * SMD candidate (t−1) −0.035 (0.012)*** Control variables  SMD MP (t) 0.070 (0.367) 0.036 (0.358)  Regional list MP (t) 0.307 (0.283) 0.320 (0.281)  Tenure (t) 0.020 (0.117) 0.046 (0.115)  Mayor (t) −0.299 (0.408) −0.287 (0.406)  Minister (t) −0.877 (0.618) −1.027 (0.620)  Party leader (t) 0.589 (0.296)** 0.586 (0.292)**  Parliamentary positions (t) 0.734 (0.594) 0.784 (0.593)  Committee chair (t) 0.489 (0.280)* 0.475 (0.277)*  Dominant party (t) 1.944 (0.395)*** 1.937 (0.398)***  Government party (t) −0.604 (0.235)** −0.502 (0.232)**  SMD candidate (t) 4.558 (0.631)*** 4.516 (0.630)***  Time −0.104 (0.043)** −0.112 (0.040)**  Constant 206.317 (87.32)** 222.612 (89.279)** Random effects parameter  SD of the MP-level intercept 1.069 (0.516) 0.989 (0.541)  N 1216 1216  Wald Chi2 70.99*** 69.38***  Log likelihood −370.870 −365.36 Notes: Entries are logistic regression coefficients with cluster robust standard errors in parentheses. Standard errors are clustered around individual MPs. *** p < 0.01, ** p < 0.05, * p < 0.1. To test the second hypothesis (H2), namely that previous SMD candidacy changes the effect of activities, I control for the interaction of activity and SMD candidacy at the previous election (Model 5). The interaction is significant, and as shown in Figure 1, reveals that in the cases of both former SMD candidates and those who only have been list candidates at the previous election activity has a positive effect on the probability of re-selection on a party list. Nevertheless, the two effects are not identical. In the case of former SMD candidates (dashed line), the curve is steeper indicating that the effect is larger in the case of those who were only nominated on party lists—and thus the party may not count on them as future SMD MPs. Hence, the parliamentary activities of prospective SMD candidates does not count as much at list nominations than that of list-only candidates, which confirms H2. Figure 1 View largeDownload slide The predicted probability of re-selection on a party-list. Figure 1 View largeDownload slide The predicted probability of re-selection on a party-list. 5. Conclusion The starting point of this study was that selectors aim at selecting legislators who are willing to carry out activities expected from their groups. In a mixed-member electoral system, SMD MPs are encouraged to work in their constituencies, while list MPs should focus more on parliamentary work. However, list MPs are not alike either: some are expected to run in SMDs at the next elections, which may change the balance between constituency and parliamentary work. Accordingly, it was expected that (H1) selectors do not take prior parliamentary activity into account when making the decision about SMD nominations, and that (H2) at party list nominations, parliamentary activities play a greater role in the case of MPs formerly not nominated in SMDs, than in the case former SMD candidates. Both hypotheses of the article have been confirmed: the mixed-effects regression models did not indicate the significant effect of activity in parliament on SMD re-nominations, but in the case of party list nomination, within the group of those formerly not having been selected in SMDs, more active legislators manage to improve their re-selection chances. These results indicate that even if only to a moderate extent parties evaluate MPs along how well they carry out the tasks dedicated to their groups. Nevertheless, the effect size indicates that MPs have to work enormously hard to considerably better re-selection prospects. The low effect size raises the question whether—despite the significant results—MPs are encouraged at all to working harder in parliament to achieve re-selection. Apart from parliamentary activities, parties value SMD incumbency and local political background on the SMD tier, which is nothing unexpected. Both, incumbency and local background increase local visibility, and are key for personal vote-seeking (Scheiner, 2005; Shugart et al., 2005; Tavits, 2010). The effect of parliamentary experience, however, is not what could be expected: due to the retirement of the most experienced MPs, the importance of seniority is not confirmed in the context of SMD re-selection. It has to be mentioned that in all models presented in this study, the random effects parameters suggest quite a large variation in the MP-level intercepts. This indicates that there are considerable differences between MPs and these affect both SMD and list tier re-selection. MPs are far from being equal in the race for re-nomination: parties clearly give advantage to some instead of others. Identifying all factors is beyond the scope of this study though. Nevertheless, as the majority of formal (hence measurable) factors are already taken into account, one could theorise that informal intra-party connections play key roles in nominations, which makes the selection process rather situational, therefore less predictable in a model setting. These informal positions, however are very hard to measure, thus their effects add to the error term within the models, leading to endogeneity related to omitted variables. Although prior list positions appear to be a solid proxy for this, we still have to face measurement challenges. The key term seems to be loyalty to the party. However, parties are notoriously united when it comes to roll-call (Ilonszki and Jáger, 2008), leaving little space for the researcher to categorise MPs on the basis of voting behaviour. The loyalty that matters here probably originates from the MPs’ personal connections to the party and its leadership. Although there are some attempts to sort out interpersonal relationships within parties (Tunkis, 2017), no systematic effort has gone into disentangling the web of intra-party connections. Intra-party personal connections are key to understanding selection patterns, and the pursuit of overcoming measurement difficulties gives us plenty to think about in the future. Funding This work was supported by the National Research, Development and Innovation Office (NKFIH) [PD 115747]. The author is a recipient of the János Bolyai Research Scholarship. Conflict of Interest The author declares that there is no conflict of interest regarding the publication of this article. Footnotes 1 Magyar Szocialista Párt; Hungarian Socialist Party. 2 Fidesz—Magyar Polgári Szövetség; Fidesz—Hungarian Civic Alliance. 3 Enyedi and Todosijevic (2009) show that the general level and the changing patterns of partisanship within the Hungarian electorate is characteristic to that of other Eastern European countries. 4 High levels of party unity are also very common in new democracies (Field, 2013). 5 The main independent variable (activity) was also computed using the standardized values of the frequency of the different activity types to control for the different magnitude in standard deviations. Running the models using the standardized version as independent variable does not change results. For the sake of easier interpretation, I use the raw activity number in the models presented in this article. 6 I do not take roll-call data and committee work into account. With regard to the first, the frequency of votes and voting costs is roughly the same across all MPs, not generating substantial differences between individuals. As to work in committees, although Hungarian MPs spend a considerable amount of time with tasks related to legislation in committees, committee work does not appear as individual activity; therefore, it is hard to measure it on the level of individual MPs. The number of committees in which the MP works may be indicative of the MPs’ workload related to committees. However, as it can only take a limited amount of (low) values, it does not add much to the average number of activities in terms of its variance. Including the number of committees into the models does not change results either. 7 Obviously, this could cause a problem of multicollinearity which is monitored throughout modelling. In none of the reported models goes the value of VIF above 2.73. The average value of VIF across all models varies between 1.2 and 1.4. 8 In Hungary, parties may jointly nominate SMD candidates and/or lists. During the period under investigation, parties joined and left electoral coalitions, which makes it difficult to separate the party effect. 9 The Dominant party variable also captures PPG-size effects. Legislators of larger PPGs can afford to work less, because the mandatory work distributes between a larger number of MPs. In small PPGs, however, a smaller number of MPs are responsible for the parliamentary work, which makes them end up with a larger workload. 10 As there is a rather substantive variation in the length of party lists, the absolute list position of the MPs on the lists is a misleading indicator of the MP’s intra-party position, and is difficult to compare across parties, or elections. Relative list position, on the other hand, takes into account the length of the party lists (i.e. the number of candidates on the party list), thus offers a more suitable measure to capture the candidate’s actual standing relative to other candidates of the party. The relative list position variable is calculated in the following way: Relative list position =List positionList length. Obviously, the higher the candidate climbs on the party list (i.e. the lower the rank), the smaller the value of the relative list position. 11 Here, I have to add that the MPs ideological distance from their parties is also considered a relevant factor in influencing re-selection chances. The rationale behind this is that the closer MPs are to their parties on the ideological spectrum, the better their positions at re-nomination. However, no suitable measures are available for the whole time period that could control for differences in ideology. Roll-call votes being strictly controlled by the party leadership throughout the whole period under investigation (Ilonszki and Jáger, 2008) are not good indicators of intra-party differences, while MP surveys are—apart from their sporadic availability—look at differences from the MPs’ viewpoints instead of how gatekeepers perceive these differences. Nevertheless, it can also be speculated that party list positions at the previous elections correlate to ideological differences between the party, in a sense that MPs higher on the party lists are closer to the ideological position of the party leadership. Furthermore, list positions incorporate other important factors that characterize intra-party relations making list positions a better indicator of these relations than ideological distance. 12 Apart from multicollinearity discussed earlier, the models have been checked for outliers. Standardized residuals were used to identify outliers. In the models, a few cases have been identified as having residuals larger than 2. Removing these observations do not change model results, thus they remain included. 13 Values in the text are the differences in predicted probabilities. References Anderson W. D. , Box-Steffensmeier J. M. , Sinclair-Chapman V. ( 2003 ) ‘The Keys to Legislative Success in the U.S. House of Representatives’ , Legislative Studies Quarterly , 28 , 357 – 386 . Google Scholar CrossRef Search ADS Atmor N. , Hazan R. Y. , Rahat G. ( 2011 ) ‘Candidate Selection’. In Colomer J. M. (ed.), Personal Representation. The Neglected Dimension of Electoral Systems , Colchester , ECPR Press , pp. 21 – 36 . Bowler S. ( 2010 ) ‘Private Members Bills in the UK Parliament: Is There an “Electoral Connection”?’ , The Journal of Legislative Studies , 16 , 476 – 494 . Google Scholar CrossRef Search ADS Bräuninger T. , Brunner M. , Däubler T. ( 2012 ) ‘Personal Vote-Seeking in Flexible List Systems: How Electoral Incentives Shape Belgian MPs’ Bill Initiation Behaviour’ , European Journal of Political Research , 51 , 607 – 645 . Google Scholar CrossRef Search ADS Cain B. , Ferejohn J. , Fiorina M. P. ( 1987 ) The Personal Vote: Constituency Service and Electoral Independence , Cambridge, MA , Harvard University Press . Google Scholar CrossRef Search ADS Carey J. M. ( 2007 ) ‘Competing Principals, Political Institutions, and Party Unity in Legislative Voting’ , American Journal of Political Science , 51 , 92 – 107 . Google Scholar CrossRef Search ADS Crisp B. , Ingall R. E. ( 2002 ) ‘Institutional Engineering and the Nature of Representation: Mapping the Effects of Electoral Reform in Colombia’ , American Journal of Political Science , 46 , 733 – 748 . Google Scholar CrossRef Search ADS Curtice J. , Shively P. ( 2009 ) ‘Who Represents Us Best? One Member or Many?’. In Klingemann H.-D. (ed.) The Comparative Study of Electoral Systems , Oxford , Oxford University Press , pp. 171 – 192 . Enyedi Z. , Bojan T. ( 2009 ) ‘Adversarial politics, civic virtues and party identification in Eastern and Western Europe’. In John B. , Paolo B. (eds) Political parties and partisanship: Social identities and individual attitudes , Oxon , Routledge , pp. 246 – 284 . Enyedi Zs. , Benoit K. ( 2011 ) ‘Kritikus Választás 2010. A Magyar Pártrendszer Átrendeződése a Bal-jobb Dimenzióban’. In Enyedi Z. , Szabó A. , Tardos R. (eds) Új Képlet. A 2010-es Választások Magyarországon , Budapest , DKMKA , pp. 17 – 42 . Enyedi Zs. , Tóka G. ( 2007 ) ‘The Only Game in Town: Party Politics in Hungary’. In Webb P. , White S. (eds) Party Politics in New Democracies , Oxford , Oxford University Press , pp. 147 – 178 . Google Scholar CrossRef Search ADS Fenno R. F. ( 1978 ) Home Style: House Members in Their Districts , Glenview, IL , Longman . Field B. N. ( 2013 ) ‘Resolute Leaders and “Cardboard Deputies”: Parliamentary Party Unity in the New Spanish Democracy’ , South European Society and Politics , 18 , 355 – 374 . Google Scholar CrossRef Search ADS Field B. N. , Siavelis P. M. ( 2008 ) ‘Candidate Selection Procedures in Transitional Polities: A Research Note’ , Party Politics , 14 , 620 – 639 . Google Scholar CrossRef Search ADS Gallagher M. ( 1980 ) ‘Candidate Selection in Ireland: The Impact of Localism and the Electoral System’ , British Journal of Political Science , 10 , 489 – 503 . Google Scholar CrossRef Search ADS Gallagher M. , Holliday I. ( 2003 ) ‘Electoral Systems, Representational Roles and Legislator Behaviour: Evidence from Hong Kong’ , New Zealand Journal of Asian Studies , 5 , 107 – 120 . Glazer A. , Wattenberg M. P. ( 1996 ) ‘How Will Term Limits Affect Legislative Work?’ [online]. In Grofman B. (ed.), Legislative Term Limits: Public Choice Perspectives , The Netherlands , Springer , pp. 37 – 46 . Retrieved from http://link.springer.com/chapter/10.1007/978-94-009-1812-2_3. Google Scholar CrossRef Search ADS Hazan R. Y. , Rahat G. ( 2010 ) Democracy within Parties: Candidate Selection Methods and their Political Consequences , Oxford , Oxford University Press . Google Scholar CrossRef Search ADS Heitshusen V. , Young G. , Wood D. M. ( 2005 ) ‘Electoral Context and MP Constituency Focus in Australia, Canada, Ireland, New Zealand, and the United Kingdom’ , American Journal of Political Science , 49 , 32 – 45 . Google Scholar CrossRef Search ADS Helmke G. , Levitsky S. ( 2004 ) ‘Informal Institutions and Comparative Politics: A Research Agenda’ , Perspectives on Politics , 2 , 725 – 740 . Google Scholar CrossRef Search ADS Hill K. Q. , Hurley P. A. ( 2002 ) ‘Symbolic Speeches in the U.S. Senate and Their Representational Implications’ , The Journal of Politics , 64 , 219 – 231 . Google Scholar CrossRef Search ADS Ilonszki G. ( 2012 ) ‘Role Stability in the Context of Institutional and Positional Change’. In Blomgren M. , Rozenberg O. (eds) Parliamentary Roles in Modern Legislatures , London , Routledge , pp. 162 – 183 . Ilonszki G. , Edinger M. ( 2007 ) ‘MPs in Post-Communist and Post-Soviet Nations: a Parliamentary Elite in the Making’ , The Journal of Legislative Studies , 13 , 142 – 163 . Google Scholar CrossRef Search ADS Ilonszki G. , Jáger K. ( 2008 ) ‘Erős Kormány—Gyenge Parlament? A Törvényhozási Kapacitás as Törvényhozási Teljesítmény Politikai Szociológiai Összefüggései’ , Századvég , 2008 , 119 – 142 . Katz R. S. ( 2001 ) ‘The Problem of Candidate Selection and Models of Party Democracy’ , Party Politics , 7 , 277 – 296 . Google Scholar CrossRef Search ADS Körösényi A. ( 2013 ) ‘Political Polarization and Its Consequences on Democratic Accountability’ , Corvinus Journal of Sociology and Social Policy , 4 , 3 – 30 . Krook M. L. ( 2010 ) Quotas for Women in Politics: Gender and Candidate Selection Reform Worldwide , Oxford , Oxford University Press . Lancaster T. D. , Patterson W. D. ( 1990 ) ‘Comparative Pork Barrel Politics’ , Comparative Political Studies , 22 , 458 – 477 . Google Scholar CrossRef Search ADS Lazardeux S. ( 2005 ) ‘“ Une Question Ecrite, Pour Quoi Faire?” The Causes of the Production of Written Questions in the French Assemblée Nationale’ , French Politics , 3 , 258 – 281 . Google Scholar CrossRef Search ADS Manow P. ( 2013 ) ‘Mixed Rules, Different Roles? An Analysis of the Typical Pathways into the Bundestag and of MPs’ Parliamentary Behaviour’ , The Journal of Legislative Studies , 19 , 287 – 308 . Google Scholar CrossRef Search ADS Marangoni F. , Tronconi F. ( 2011 ) ‘When Territory Matters: Parliamentary Profiles and Legislative Behaviour in Italy (1987–2008)’ , The Journal of Legislative Studies , 17 , 415 – 434 . Google Scholar CrossRef Search ADS Marjai E. ( 2012 ) ‘Candidate Selection Patterns in the 2010 Hungarian Parliamentary Elections’, MA Thesis, Budapest, Central European University. Martin S. ( 2011a ) ‘Parliamentary Questions, the Behaviour of Legislators, and the Function of Legislatures: an Introduction’ , The Journal of Legislative Studies , 17 , 259 – 270 . Google Scholar CrossRef Search ADS Martin S. ( 2011b ) ‘Electoral Institutions, the Personal Vote, and Legislative Organization’ , Legislative Studies Quarterly , 36 , 339 – 361 . Google Scholar CrossRef Search ADS Mayhew D. R. ( 1974 ) Congress: the Electoral Connection , 2nd edn , New Haven, CT , Yale University Press . Norris P. ( 1997 ) ‘The Puzzle of Constituency Service’ , The Journal of Legislative Studies , 3 , 29 – 49 . Google Scholar CrossRef Search ADS Norton P. ( 2002 ) ‘Introduction: Linking Parliaments and Citizens’. In Norton P. (ed.) Parliaments and Citizens in Western Europe , London , Frank Cass , pp. 1 – 18 . Norton P. , Wood D. ( 1990 ) ‘Constituency Service by Members of Parliament: Does It Contribute to a Personal Vote?’ , Parliamentary Affairs , 43 , 196 – 208 . Google Scholar CrossRef Search ADS Papp Zs. ( 2016 ) ‘Shadowing the Elected: Mixed-Member Incentives to Locally Oriented Parliamentary Questioning’ , The Journal of Legislative Studies , 22 , 216 – 236 . Google Scholar CrossRef Search ADS Rasch B. E. ( 2009 ) ‘Opposition Parties, Electoral Incentives and the Control of Government Ministers: Parliamentary Questioning in Norway’. In Ganghof S. , Hönnige C. , Stecker C. (eds), Parlamente, Agendasetzung und Vetospieler , VS Verlag für Sozialwissenschaften , pp. 199 – 214 . Google Scholar CrossRef Search ADS Russo F. ( 2011 ) ‘The Constituency as a Focus of Representation: Studying the Italian Case through the Analysis of Parliamentary Questions’ , The Journal of Legislative Studies , 17 , 290 – 301 . Google Scholar CrossRef Search ADS Scheiner E. ( 2005 ) Democracy without Competition in Japan: Opposition Failure in a One-Party Dominant State , Cambridge , Cambridge University Press . Google Scholar CrossRef Search ADS Shugart M. S. , Valdini M. E. , Suominen K. ( 2005 ) ‘Looking for Locals: Voter Information Demands and Personal Vote-Earning Attributes of Legislators under Proportional Representation’ , American Journal of Political Science , 49 , 437 – 449 . Google Scholar CrossRef Search ADS Siavelis P. ( 2002 ) ‘The Hidden Logic of Candidate Selection for Chilean Parliamentary Elections’ , Comparative Politics , 34 , 419 – 438 . Google Scholar CrossRef Search ADS Siavelis P. M. , Morgenstern S. ( 2008 ) ‘Political Recruitment and Candidate Selection in Latin America: A Framework for Analysis’. In Siavelis P. M. , Morgenstern S. (eds) Pathways to Power: Political Recruitment and Candidate Selection in Latin America , University Park, PA , Pennsylvania State University Press , pp. 3 – 37 . Stratmann T. , Baur M. ( 2002 ) ‘Plurality Rule, Proportional Representation, and the German Bundestag: How Incentives to Pork-Barrel Differ across Electoral Systems’ , American Journal of Political Science , 46 , 506 – 514 . Google Scholar CrossRef Search ADS Strøm K. ( 1997 ) ‘Rules, Reasons and Routines: Legislative Roles in Parliamentary Democracies’, Müller W. C. , Saalfeld T. (eds) Members of Parliament in Western Europe. Roles and Behaviour , London , Frank Cass , pp. 155 – 174 . Tavits M. ( 2009 ) ‘The Making of Mavericks Local Loyalties and Party Defection’ , Comparative Political Studies , 42 , 793 – 815 . Google Scholar CrossRef Search ADS Tavits M. ( 2010 ) ‘Effect of Local Ties on Electoral Success and Parliamentary Behaviour: the Case of Estonia’ , Party Politics , 16 , 215 – 235 . Google Scholar CrossRef Search ADS Tunkis P. J. ( 2017 ) ‘The Ties that Bind: Do Group Associations among Legislators Matter for Political Parties?’ , Problems of Post-Communism , 64 , 79 – 93 . Google Scholar CrossRef Search ADS Ward L. J. ( 1998 ) ‘Second-Class MPs? New-Zealand’s Adaptation to Mixed-Member Parliamentary Representation’ , Political Science , 1998 , 127 . © The Author(s) 2018. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For Permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Parliamentary Affairs Oxford University Press

Same Species, Different Breed: The Conditional Effect of Legislator Activities in Parliament on Re-Selection in a Mixed-Member Electoral System

Parliamentary Affairs , Volume Advance Article – Jan 31, 2018

Loading next page...
 
/lp/ou_press/same-species-different-breed-the-conditional-effect-of-legislator-zOQIiXrjSg
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For Permissions, please e-mail: journals.permissions@oup.com
ISSN
0031-2290
eISSN
1460-2482
D.O.I.
10.1093/pa/gsx049
Publisher site
See Article on Publisher Site

Abstract

Abstract The article investigates whether or not the amount of work legislators carry out in parliament affects their chances of re-selection in a mixed-member electoral system. A unique dataset collecting Hungarian Members of Parliaments’ (MPs) publicly available electoral, socio-demographic and parliamentary activity data between 1998 and 2010 is analysed. The study concludes that activity in parliament positively influences re-selection chances when selectors decide on party list nominations. Additionally, MPs whose parties do not count on as future Single Member District (SMD) candidates benefit from parliamentary work to a greater extent than prospective SMD candidates. At the same time, extra work in parliament does not bring MPs closer to SMD nominations. Results confirm that selectors evaluate MPs along how well they carry out the tasks dedicated to their groups. Students of legislators’ behaviour differ regarding the activities they study, the factors with which they explain the variance in these activities, the consequences which they attribute to the activities and the methods they apply to uncover causal relationships. However, the majority of authors seems to agree with Mayhew’s oft-cited starting point, and considers Members of Parliament (MPs) as ‘single-minded seekers of re-election’ (Mayhew, 1974). Consequently, most of the things representatives do in the present are attributed to the MPs’ individual desire for future re-election. At the same time, students of candidate selection point out that under centralised selection rules and with a high number of safe seats, emphasis shifts from re-election to re-selection (Hazan and Rahat, 2010; Atmor et al., 2011). Additionally, as Strøm (1997) argues, MPs act strategically to reach their goals in the hierarchical order of their feasibility: in order to be re-elected, one has to be re-selected first. The situation becomes even more complicated in mixed-member electoral systems when parties may nominate candidates in Single Member Districts (SMDs) and at the same time closed party lists. In the first case, re-selection is dependent upon the potential for re-election, inviting strategies that balance between promoting one’s ability to attract extra votes and advertising party loyalty (see also Russo, 2011). In the second case, due to the inability of the voters to change the order of candidates on the list, the vote gathering potential of individual MPs is negligible, thus loyalty will be enforced at the stage of candidate selection. Relatively little attention has been paid to the extent of parliamentary work and its effects on the MPs’ electoral fortunes. When developing the portfolio of their activities, balancing between constituency and parliamentary work, legislators face a strategic decision. On the one hand, constituency work earns more visibility among constituents, thus allocating resources to it is desirable for vote gathering purposes. On the other hand, parliamentary work must be done for a functioning parliament without any substantial electoral benefit in terms vote-seeking, which invites only little effort from MPs. From the viewpoint of the parties, however, both are essential. Parties need both MPs who carry out constituency service—and through that stabilise their support in the district, and keep voters satisfied—and MPs who are active in parliament independent of their constituency ties—speak on the floor, submit questions and sponsor bills. The starting point of the article is that selectors aim at selecting representatives who are expected to perform best in the task dedicated to their positions. Very simply put, in a mixed-member context, list candidates are expected to focus more on parliamentary work than SMD candidates, and parties may use prior performance in parliament to evaluate the MPs’ future potential. In this study the focus is on re-selection chances in a mixed-member electoral system. More specifically, I askif there is a connection between the volume of activities legislators carry out in parliament and their candidacy at the next elections. In other words: are MPs encouraged to working harder in order to reach their goals, and make the first steps toward re-election? I select Hungary as case and analyse electoral data from 1998 to 2010. The article finds that activity in parliament positively influences re-selection chances when selectors decide on party list nominations. Additionally, MPs whose parties do not count on as future SMD candidates benefit from parliamentary work to a greater extent than prospective SMD candidates. Furthermore, extra work in parliament does not bring MPs closer to SMD nominations. Results confirm that selectors evaluate MPs along how well they carry out the tasks dedicated to their group. 1. Theoretical argument There are many factors on various levels (i.e. macro, legal, institutional, party) that influence what types of candidates are selected (Siavelis and Morgenstern, 2008). We may distinguish between formal party rules and informal norms that guide the implementation of these rules (Helmke and Levitsky, 2004). The overwhelming majority of the literature on candidate selection deals with the former, especially its centralisation and democratisation (see e.g. Field and Siavelis, 2008; Hazan and Rahat, 2010; Atmor et al., 2011). Lately, little attention has been paid to candidate characteristics and activities that potentially increase the likelihood of selection, and that we may also consider the informal drives of candidate selection. The question here is, what kinds of candidates do parties select and why? It is safe to say that candidates are not equal: some are more likely to be selected than others (Katz, 2001). Similarly, one can assume that legislators are also not identical in terms of re-selection chances. The most important aspect of candidate selection (besides its formal rules) is the candidate’s loyalty to the selectorate (i.e. the body or collections of individuals that selected the candidate), which may also be the central determinant of candidate behaviour (Siavelis and Morgenstern, 2008). Candidate strength is an additional strong determinant of nominations: parties tend to select locally (Gallagher, 1980) or nationally (Siavelis, 2002) well-known politicians, like incumbents, local politicians, ministers or party leaders. Parties may even create rules with respect to seniority, incumbency (Helmke and Levitsky, 2004) or gender (Krook, 2010). As candidacy also depends on candidate behaviour, once candidates are elected, (s)election rules continue to affect the extent to which they think about re-selection, which in turn will affect their decisions about behaviour in parliament (Siavelis and Morgenstern, 2008). Thus, legislators who want to re-enter the electoral arena have to make strategic decisions about how they manage their time and other resources when in office. It has been shown on many occasions that in a mixed-member electoral system, MP behaviour differs across mandate type. Because of the diverging mechanisms of accountability, SMD MPs are more likely to engage in district representation, while list MPs are more party-centred (Lancaster and Patterson, 1990; Gallagher and Holliday, 2003; Heitshusen et al., 2005; Curtice and Shively, 2009). SMD MPs may engage in two types of activities: work in the constituency and work in parliament (Glazer and Wattenberg, 1996). As to the former, they hold office hours, carry out ombudsman-like activities and deal with problems arising in the district as well as respond to individual petitions (Fenno, 1978; Cain et al., 1987; Norton and Wood, 1990; Norris, 1997). Work in parliament consists of sponsoring bills, drafting questions or speaking on the floor. To maximise the share of votes in the SMDs at the next election, SMD MPs are expected to carry out constituency service, which leaves less space for parliamentary work. However, MPs do not necessarily need to trade-off between constituency and parliament work. The local focus of bill initiation (Crisp and Ingall, 2002; Marangoni and Tronconi, 2011), membership in certain committees (Stratmann and Baur, 2002; Manow, 2013), speeches (Hill and Hurley, 2002), parliamentary questioning (Martin, 2011a) and the willingness to defect the party line at roll-call (Carey, 2007; Tavits, 2009) can reflect general concerns about issues that are potentially important for the constituency, bringing parliamentary work closer to constituency service. Nevertheless, constituency service on location is always more visible than in parliament, and it also creates greater satisfaction for MPs with their jobs (Norton, 2002), which provides greater incentives to upgrade their constituency presence and economise on parliament work. Also, constituency work is more highly evaluated in countries where work in parliament provides little electoral benefit (Martin, 2011b), because of—for instance—the poor visibility of parliamentary work. As parliamentary work is not essential to a job well done in the case of SMD MPs, selectors are expected not to take parliamentary activities into account at SMD nominations. H1: Selectors do not take prior parliamentary activity into account when making the decision about SMD nominations. Contrarily, as list MPs do not have office-related obligations towards the constituencies, they should be focusing more on parliamentary activities, tasks within the party organisation and representing the party in public. As SMD MPs are more occupied with constituency work, the larger part of the tasks in parliament naturally falls onto list MPs. Although parliamentary work may not result in direct electoral benefits such as extra votes gathered by direct constituency work, the mere intensity with which MPs engage in parliamentary activities may work as a simple proxy of their efforts to establish the reputation of a hard-working representative, who is willing to do the ‘legwork’ without any clear compensation. Parties gain important information on legislators simply by looking at their track records, without having even to look at the content of their activities. Bill sponsorship, (Bowler, 2010; Bräuninger et al., 2012), the frequency of floor speeches (Anderson et al., 2003) and tabling questions (Lazardeux, 2005; Rasch, 2009) or simple attendance indicate that they work just as hard—or harder—as other members. Nevertheless, list MPs are also not alike from at least one important perspective: their prior record as SMD candidates. As list MPs are also very often SMD candidates at the previous elections, parties may have different plans for them than serving just as ‘second class representatives’ (Ward, 1998). It is very likely that they are the future SMD candidates of the party, not to mention that they can work as ‘shadows’ to SMD MPs (Papp, 2016), which helps maintain the connection between the party and the constituency. Therefore, list MPs who were nominated in an SMD at the previous elections are expected to engage in extra activities, such as constituency work. For the rest of the list MPs, parliamentary work is the major task. H2: At party list nominations, parliamentary activities play a greater role in the case of MPs formerly not nominated in SMDs, than in the case former SMD candidates. 2. The Hungarian case During the period under investigation (1998–2010), Hungary has a three-tier electoral system with unlinked electoral tiers. A total number of 386 legislators are elected from 176 SMDs (SMD tier), 20 regions and the national level (regional and national list tiers). Both list tiers apply closed party lists, and—especially the national tier—are designed to compensate the often disproportional results of the SMD tier. Parties may nominate candidates on multiple electoral tiers at the same time. This practice is quite widespread in Hungary: between 1990 and 2010, almost 40% of the candidates were nominated on more than one tier. In the Hungarian parliament, House rules entitle individual legislators to sponsor bills, make brief speeches on the plenary during debates and submit questions. There are four types of questions in the Országgyűlés: interpellations, written, oral and direct questions. These types differ with regards to the form of the response as well as party control. Interpellations, oral and direct questions must be answered at the plenary session, while written questions require written response. With the exception of direct questions, all questions have to be submitted in writing prior to the session. With regards to party control over the frequency with which MPs may engage in the different types of activities, generally speaking, activities that have time limits, and may receive more publicity are controlled by the PPG-leadership to a greater extent than those not limited either in time or in numbers. In the Hungarian parliament, MPs may submit as many written questions as they want, while oral forms (including speeches, interpellations, direct and oral questions) are strictly limited by House rules applying quite restrictive time constraints. Consequently, written question are subject to less party control than oral activities. Hungarian politics was characterised by the competition of two blocks lead by MSZP1 and Fidesz2 until the 2010 elections, when two new parties (LMP and Jobbik) emerged and MSZP went into a dramatic decline. In the case of the elections under investigation here, the majority of SMD seats went to either Fidesz or MSZP. Candidate selection in Hungarian parties is not a well-studied topic. There is relatively little hard evidence on party strategies, especially in a longitudinal fashion. However, from the little that we have, we may re-construct candidate selection as being organised by the party centres. This is no surprise as Field and Siavelis (2008) argue that in countries after transition selection of candidates is likely to become centralised by the reason of political uncertainty, lower party organisation and strategically complex electoral systems. In Hungary the party centres select, interview and/or approve candidates (Marjai, 2012, p. 48) either on the regional or on the national scale. There are several additional circumstances that—despite the existence of the SMD tier—make Hungarian politics party centred. These characteristics are hardly unique to Hungary. Along with centralised candidate selection, strong partisan attachment of voters (Enyedi and Tóka, 2007),3 increasing polarisation (Enyedi and Benoit, 2011; Körösényi, 2013) and strong party discipline (Ilonszki and Jáger, 2008; Ilonszki, 2012)4 all characterise new democracies. In this sense, as sharing the same logic, the Hungarian case may be considered as a typical case for new democracies. As to mixed member electoral systems, if hypotheses derived from the literature on established democracies can be verified using a new democracy as a case, it is to be expected that results hold for older democracies as well. 3. Data and variables The analysis builds upon a unique dataset that collects Hungarian MPs’ publicly available electoral, socio-demographic and parliamentary activity data. The data cover three consecutive parliaments from 1998 to 2010. As we follow a relatively large number of individuals (large N) throughout a relatively short period of time (small T), the data can be handled as panel. Since there are also MPs who did not participate in every single election under investigation, the panel is fairly unbalanced. Throughout this study t will denote the year of re-selection (2002, 2006 and 2010). MPs are elected at election t−1 (1998, 2002 and 2006), serve between t−1 and t and are re-selected (or not) at t. The dataset includes 1216 observations for 683 individual MPs. Three hundred and twenty-two MPs only served once during the three electoral terms under investigation, 189 served twice and 172 three times. 3.1 Dependent variables Throughout the analysis I use two dependent variables, which measure whether or not the MP was re-selected in an (1) SMD and on a (2) party list at election t. Seventy-three percent of MPs running at the next election was nominated in one of the constituencies, while 80.4% of the MPs secured a position on the party lists. 3.2 Independent variable: activity in parliament As outlined earlier, MPs may engage in a wide range of activities. They attend plenary sessions; take the floor and make speeches, sponsor bills, submit and table various types of questions. With the exception of attendance, information on all these activities is available on the official website of the Hungarian parliament. Utilizing the data available, I create a variable measuring average MP activity by taking the mean value of the absolute number of bills sponsored by the MP, the number of questions and the number of speeches within the respective electoral term.5 Although these activities may not be comparable in terms of their costs (time, human resource, etc. requirements), neither the level of control by the parties, an average value indicates how much the MP makes herself visible in the parliamentary work.6 3.3 Control variables To obtain the net effect of parliamentary activities on re-selection, additional variables must be controlled for. As SMD mandates are more prestigious positions in the Hungarian parliament, I expect that SMD MPs have a greater chance of re-selection than list MPs. Thus, I include mandate type obtained at election t−1. Additionally, I also account for tenure as the number of electoral terms served as an MP at the time of election t. Taking the low levels of legislative turnover into consideration (Ilonszki and Edinger, 2007), I assume that a more extensive legislative experience will increase the chances of re-nomination. Filling in various political positions is also expected to help MPs getting re-selected. I test if mayors, ministers, party leaders, parliamentary office holders and parliamentary committee chairs have better chances in the race for re-nomination. Including these variables (all dummies) is also key because it is expected that these MPs may take the floor more frequently than backbench MPs, thus controlling only for the effect of parliamentary activities could absorb the effect of the position variables, which would result in a significant effect for activities when there is none.7 Hungarian electoral rules permit multiple candidacy, namely that the same individual may be nominated at different tiers of the electoral system. As multiple candidacy may function as a ‘safety net,’ parties may nominate important candidates on more tiers at the same time to maximise election chances. Thus, simultaneous candidacy on multiple tiers is expected to correlate with re-selection in an SMD as well as on a party list. Moving on, taking the party effect into account regarding the three elections under investigation is difficult by the reason of changing electoral coalitions.8 However, to get as close to the party differences as possible I include two variables into the models: the dominant party effect captures the difference between the two dominant parties at the time (MSZP and Fidesz) and all the others,9 while the government effect accounts for the differences parties may have in their selection strategies by the reason of their government versus opposition status. The unmeasured variable, the MP’s position within the party, affects both re-selection chances and the frequency with which the party allows the MP to take the floor in parliament. Assuming a strong correlation between the MP’s position within the party and her position on the party lists, with controlling for party list positions at the current election (as a proxy for the MP’s position within the party),10 I hope to mitigate the problem of endogeneity in the model caused by the interrelatedness of parliamentary activity and the MP’s position within the party, the effect of which would otherwise appear in the error term.11 Furthermore, I control for the lagged dependent variables to see if SMD and party list candidacy at election t−1 affects candidacy at election t. And finally, I also include a time variable to capture the differences between the three elections. 4. Results As the dataset has panel properties, to model the effect of activity in parliament, I use mixed-effects regressions. SMD and party list candidacy being binary variables, mixed effects logit regression is the proper choice.12 In all models, standard errors are clustered around individual MPs. I only discuss effects that are significant on a 5% level. Table 1 displays the results of the regression models explaining whether or not the MP was nominated in an SMD at election t. Table 1 Mixed-effects binary logit regressions explaining SMD nominations at the next elections Model 1 Model 2 Model 3 Activity in parliament 0.002 (0.002) −0.000 (0.003) 0.000 (0.003) Former candidacy  SMD candidate (t−1) 1.623 (0.235)*** 2.307 (0.345)*** 1.821 (0.280)***  List candidate (t−1) 0.034 (0.216) −0.611 (0.366)* 0.308 (0.299) Control variables  SMD MP (t) 1.604 (0.295)*** 0.447 (0.443) 1.529 (0.354)***  Regional list MP (t) −0.336 (0.229) −0.897 (0.322)*** 0.397 (0.277)  Tenure (t) −0.392 (0.100)*** −0.344 (0.141)** −0.361 (0.116)***  Mayor (t) 1.065 (0.341)*** 0.265 (0.436) 0.638 (0.550)  Minister (t) 1.541 (0.508)*** 13.265 (614.290) 0.346 (0.604)  Party leader (t) −0.034 (0.203) −0.446 (0.304) −0.379 (0.242)  Parliamentary positions (t) 0.318 (0.407) 0.074 (0.665) −0.165 (0.435)  Committee chair (t) 0.264 (0.197) 0.109 (0.298) −0.169 (0.245)  Dominant party (t) −0.204 (0.240) −1.557 (0.483)*** −1.203 (0.365)***  Government party (t) −0.062 (0.172) 0.785 (0.270)*** 0.466 (0.237)**  Regional list candidate (t) 2.005 (0.266)***  National list candidate (t) 0.114 (0.203)  Relative position on the regional list (t) −4.340 (0.823)***  Relative position on the national list (t) −0.469 (0.433)  Time −0.146 (0.035)*** 0.025 (0.041) −0.009 (0.037)  Constant 290.751 (71.903)*** −46.646 (83.265) 18.293 (75.080) Random effects parameter  SD of the MP-level intercept 0.807 (0.389) 0.355 (0.981) 0.001 (4.615)  N 1216 633 547  Wald Chi2 163.84*** 74.64*** 118.48***  Log likelihood −572.75 −228.428 −261.029 Model 1 Model 2 Model 3 Activity in parliament 0.002 (0.002) −0.000 (0.003) 0.000 (0.003) Former candidacy  SMD candidate (t−1) 1.623 (0.235)*** 2.307 (0.345)*** 1.821 (0.280)***  List candidate (t−1) 0.034 (0.216) −0.611 (0.366)* 0.308 (0.299) Control variables  SMD MP (t) 1.604 (0.295)*** 0.447 (0.443) 1.529 (0.354)***  Regional list MP (t) −0.336 (0.229) −0.897 (0.322)*** 0.397 (0.277)  Tenure (t) −0.392 (0.100)*** −0.344 (0.141)** −0.361 (0.116)***  Mayor (t) 1.065 (0.341)*** 0.265 (0.436) 0.638 (0.550)  Minister (t) 1.541 (0.508)*** 13.265 (614.290) 0.346 (0.604)  Party leader (t) −0.034 (0.203) −0.446 (0.304) −0.379 (0.242)  Parliamentary positions (t) 0.318 (0.407) 0.074 (0.665) −0.165 (0.435)  Committee chair (t) 0.264 (0.197) 0.109 (0.298) −0.169 (0.245)  Dominant party (t) −0.204 (0.240) −1.557 (0.483)*** −1.203 (0.365)***  Government party (t) −0.062 (0.172) 0.785 (0.270)*** 0.466 (0.237)**  Regional list candidate (t) 2.005 (0.266)***  National list candidate (t) 0.114 (0.203)  Relative position on the regional list (t) −4.340 (0.823)***  Relative position on the national list (t) −0.469 (0.433)  Time −0.146 (0.035)*** 0.025 (0.041) −0.009 (0.037)  Constant 290.751 (71.903)*** −46.646 (83.265) 18.293 (75.080) Random effects parameter  SD of the MP-level intercept 0.807 (0.389) 0.355 (0.981) 0.001 (4.615)  N 1216 633 547  Wald Chi2 163.84*** 74.64*** 118.48***  Log likelihood −572.75 −228.428 −261.029 Notes: Entries are logistic regression coefficients with cluster robust standard errors in parentheses. Standard errors are clustered around individual MPs. *** p < 0.01, ** p < 0.05, * p < 0.1. Table 1 Mixed-effects binary logit regressions explaining SMD nominations at the next elections Model 1 Model 2 Model 3 Activity in parliament 0.002 (0.002) −0.000 (0.003) 0.000 (0.003) Former candidacy  SMD candidate (t−1) 1.623 (0.235)*** 2.307 (0.345)*** 1.821 (0.280)***  List candidate (t−1) 0.034 (0.216) −0.611 (0.366)* 0.308 (0.299) Control variables  SMD MP (t) 1.604 (0.295)*** 0.447 (0.443) 1.529 (0.354)***  Regional list MP (t) −0.336 (0.229) −0.897 (0.322)*** 0.397 (0.277)  Tenure (t) −0.392 (0.100)*** −0.344 (0.141)** −0.361 (0.116)***  Mayor (t) 1.065 (0.341)*** 0.265 (0.436) 0.638 (0.550)  Minister (t) 1.541 (0.508)*** 13.265 (614.290) 0.346 (0.604)  Party leader (t) −0.034 (0.203) −0.446 (0.304) −0.379 (0.242)  Parliamentary positions (t) 0.318 (0.407) 0.074 (0.665) −0.165 (0.435)  Committee chair (t) 0.264 (0.197) 0.109 (0.298) −0.169 (0.245)  Dominant party (t) −0.204 (0.240) −1.557 (0.483)*** −1.203 (0.365)***  Government party (t) −0.062 (0.172) 0.785 (0.270)*** 0.466 (0.237)**  Regional list candidate (t) 2.005 (0.266)***  National list candidate (t) 0.114 (0.203)  Relative position on the regional list (t) −4.340 (0.823)***  Relative position on the national list (t) −0.469 (0.433)  Time −0.146 (0.035)*** 0.025 (0.041) −0.009 (0.037)  Constant 290.751 (71.903)*** −46.646 (83.265) 18.293 (75.080) Random effects parameter  SD of the MP-level intercept 0.807 (0.389) 0.355 (0.981) 0.001 (4.615)  N 1216 633 547  Wald Chi2 163.84*** 74.64*** 118.48***  Log likelihood −572.75 −228.428 −261.029 Model 1 Model 2 Model 3 Activity in parliament 0.002 (0.002) −0.000 (0.003) 0.000 (0.003) Former candidacy  SMD candidate (t−1) 1.623 (0.235)*** 2.307 (0.345)*** 1.821 (0.280)***  List candidate (t−1) 0.034 (0.216) −0.611 (0.366)* 0.308 (0.299) Control variables  SMD MP (t) 1.604 (0.295)*** 0.447 (0.443) 1.529 (0.354)***  Regional list MP (t) −0.336 (0.229) −0.897 (0.322)*** 0.397 (0.277)  Tenure (t) −0.392 (0.100)*** −0.344 (0.141)** −0.361 (0.116)***  Mayor (t) 1.065 (0.341)*** 0.265 (0.436) 0.638 (0.550)  Minister (t) 1.541 (0.508)*** 13.265 (614.290) 0.346 (0.604)  Party leader (t) −0.034 (0.203) −0.446 (0.304) −0.379 (0.242)  Parliamentary positions (t) 0.318 (0.407) 0.074 (0.665) −0.165 (0.435)  Committee chair (t) 0.264 (0.197) 0.109 (0.298) −0.169 (0.245)  Dominant party (t) −0.204 (0.240) −1.557 (0.483)*** −1.203 (0.365)***  Government party (t) −0.062 (0.172) 0.785 (0.270)*** 0.466 (0.237)**  Regional list candidate (t) 2.005 (0.266)***  National list candidate (t) 0.114 (0.203)  Relative position on the regional list (t) −4.340 (0.823)***  Relative position on the national list (t) −0.469 (0.433)  Time −0.146 (0.035)*** 0.025 (0.041) −0.009 (0.037)  Constant 290.751 (71.903)*** −46.646 (83.265) 18.293 (75.080) Random effects parameter  SD of the MP-level intercept 0.807 (0.389) 0.355 (0.981) 0.001 (4.615)  N 1216 633 547  Wald Chi2 163.84*** 74.64*** 118.48***  Log likelihood −572.75 −228.428 −261.029 Notes: Entries are logistic regression coefficients with cluster robust standard errors in parentheses. Standard errors are clustered around individual MPs. *** p < 0.01, ** p < 0.05, * p < 0.1. The results of Model 1 indicate no connection between the intensity of parliamentary activities and the probability of re-selection in SMD, which confirms H1. Nevertheless, the distinguished role of incumbency is observed, as being a constituency representative increases the probability of SMD nomination by 0.39.13 Additionally, not only SMD candidacy at the previous elections boosts the likelihood of re-selection by quite a large margin (0.46), but also mayors and ministers find it easier to be re-nominated. With regards to multiple candidacy, MPs who are selected also on regional party lists become SMD candidates with a significantly greater probability. The same cannot be said for national list candidates, which indicates that parties use regional party lists as ‘safety net’ for SMD candidates in case they fail to bring in the district. This is further supported when relative list positions are included into Model 1. Here, we find (see Models 2 and 3) that candidates higher on the regional list (i.e. taking a lower value on the relative list position variable) are significantly more likely to be selected as SMD candidates. The difference in the probabilities of selection for two MPs at the two extremes of the relative list positions variable is 0.54. Although there is no significant effect attached to relative positions on the national party list, its negative sign reveals similar tendencies. Last but not least, contrary to the expectations tenure has a significant negative effect on SMD nominations. Although parties tend to ‘stick to their people’ in the given districts (see the effect of SMD incumbency), at the same time they seem to give an advantage to less experienced MPs as well. The difference in probabilities between the least and most experienced MPs is 0.24 which suggests a considerable disadvantage for the most senior legislators, and is probably a consequence of the retirement of the most experienced MPs. Turning to party list nominations, we see that parliamentary activities play a significant role (see Table 2, Model 4). The effect size, however, is quite small. A 10 unit change in the activity variable increases the chance of list nomination by 0.014. Additionally, the probability of selection on a party lists starts from a quite high level: even with an activity score of 0, MPs have a 0.78 chance of reselection on one of the party lists. With an average level of activity, MPs are able to bring up the re-selection probability by only 0.035. This indicates that MPs have to work exceptionally hard to improve their re-selection chances. Table 2 Mixed-effects binary logit regressions explaining nomination on party lists at the next elections Model 4 Model 5 Activity in parliament 0.014 (0.004)*** 0.043 (0.012)*** Former candidacy  SMD candidate (t−1) −0.603 (0.303)** −0.109 (0.334)  List candidate (t−1) 0.812 (0.293)*** 0.783 (0.287)*** Interaction term  Activity * SMD candidate (t−1) −0.035 (0.012)*** Control variables  SMD MP (t) 0.070 (0.367) 0.036 (0.358)  Regional list MP (t) 0.307 (0.283) 0.320 (0.281)  Tenure (t) 0.020 (0.117) 0.046 (0.115)  Mayor (t) −0.299 (0.408) −0.287 (0.406)  Minister (t) −0.877 (0.618) −1.027 (0.620)  Party leader (t) 0.589 (0.296)** 0.586 (0.292)**  Parliamentary positions (t) 0.734 (0.594) 0.784 (0.593)  Committee chair (t) 0.489 (0.280)* 0.475 (0.277)*  Dominant party (t) 1.944 (0.395)*** 1.937 (0.398)***  Government party (t) −0.604 (0.235)** −0.502 (0.232)**  SMD candidate (t) 4.558 (0.631)*** 4.516 (0.630)***  Time −0.104 (0.043)** −0.112 (0.040)**  Constant 206.317 (87.32)** 222.612 (89.279)** Random effects parameter  SD of the MP-level intercept 1.069 (0.516) 0.989 (0.541)  N 1216 1216  Wald Chi2 70.99*** 69.38***  Log likelihood −370.870 −365.36 Model 4 Model 5 Activity in parliament 0.014 (0.004)*** 0.043 (0.012)*** Former candidacy  SMD candidate (t−1) −0.603 (0.303)** −0.109 (0.334)  List candidate (t−1) 0.812 (0.293)*** 0.783 (0.287)*** Interaction term  Activity * SMD candidate (t−1) −0.035 (0.012)*** Control variables  SMD MP (t) 0.070 (0.367) 0.036 (0.358)  Regional list MP (t) 0.307 (0.283) 0.320 (0.281)  Tenure (t) 0.020 (0.117) 0.046 (0.115)  Mayor (t) −0.299 (0.408) −0.287 (0.406)  Minister (t) −0.877 (0.618) −1.027 (0.620)  Party leader (t) 0.589 (0.296)** 0.586 (0.292)**  Parliamentary positions (t) 0.734 (0.594) 0.784 (0.593)  Committee chair (t) 0.489 (0.280)* 0.475 (0.277)*  Dominant party (t) 1.944 (0.395)*** 1.937 (0.398)***  Government party (t) −0.604 (0.235)** −0.502 (0.232)**  SMD candidate (t) 4.558 (0.631)*** 4.516 (0.630)***  Time −0.104 (0.043)** −0.112 (0.040)**  Constant 206.317 (87.32)** 222.612 (89.279)** Random effects parameter  SD of the MP-level intercept 1.069 (0.516) 0.989 (0.541)  N 1216 1216  Wald Chi2 70.99*** 69.38***  Log likelihood −370.870 −365.36 Notes: Entries are logistic regression coefficients with cluster robust standard errors in parentheses. Standard errors are clustered around individual MPs. *** p < 0.01, ** p < 0.05, * p < 0.1. Table 2 Mixed-effects binary logit regressions explaining nomination on party lists at the next elections Model 4 Model 5 Activity in parliament 0.014 (0.004)*** 0.043 (0.012)*** Former candidacy  SMD candidate (t−1) −0.603 (0.303)** −0.109 (0.334)  List candidate (t−1) 0.812 (0.293)*** 0.783 (0.287)*** Interaction term  Activity * SMD candidate (t−1) −0.035 (0.012)*** Control variables  SMD MP (t) 0.070 (0.367) 0.036 (0.358)  Regional list MP (t) 0.307 (0.283) 0.320 (0.281)  Tenure (t) 0.020 (0.117) 0.046 (0.115)  Mayor (t) −0.299 (0.408) −0.287 (0.406)  Minister (t) −0.877 (0.618) −1.027 (0.620)  Party leader (t) 0.589 (0.296)** 0.586 (0.292)**  Parliamentary positions (t) 0.734 (0.594) 0.784 (0.593)  Committee chair (t) 0.489 (0.280)* 0.475 (0.277)*  Dominant party (t) 1.944 (0.395)*** 1.937 (0.398)***  Government party (t) −0.604 (0.235)** −0.502 (0.232)**  SMD candidate (t) 4.558 (0.631)*** 4.516 (0.630)***  Time −0.104 (0.043)** −0.112 (0.040)**  Constant 206.317 (87.32)** 222.612 (89.279)** Random effects parameter  SD of the MP-level intercept 1.069 (0.516) 0.989 (0.541)  N 1216 1216  Wald Chi2 70.99*** 69.38***  Log likelihood −370.870 −365.36 Model 4 Model 5 Activity in parliament 0.014 (0.004)*** 0.043 (0.012)*** Former candidacy  SMD candidate (t−1) −0.603 (0.303)** −0.109 (0.334)  List candidate (t−1) 0.812 (0.293)*** 0.783 (0.287)*** Interaction term  Activity * SMD candidate (t−1) −0.035 (0.012)*** Control variables  SMD MP (t) 0.070 (0.367) 0.036 (0.358)  Regional list MP (t) 0.307 (0.283) 0.320 (0.281)  Tenure (t) 0.020 (0.117) 0.046 (0.115)  Mayor (t) −0.299 (0.408) −0.287 (0.406)  Minister (t) −0.877 (0.618) −1.027 (0.620)  Party leader (t) 0.589 (0.296)** 0.586 (0.292)**  Parliamentary positions (t) 0.734 (0.594) 0.784 (0.593)  Committee chair (t) 0.489 (0.280)* 0.475 (0.277)*  Dominant party (t) 1.944 (0.395)*** 1.937 (0.398)***  Government party (t) −0.604 (0.235)** −0.502 (0.232)**  SMD candidate (t) 4.558 (0.631)*** 4.516 (0.630)***  Time −0.104 (0.043)** −0.112 (0.040)**  Constant 206.317 (87.32)** 222.612 (89.279)** Random effects parameter  SD of the MP-level intercept 1.069 (0.516) 0.989 (0.541)  N 1216 1216  Wald Chi2 70.99*** 69.38***  Log likelihood −370.870 −365.36 Notes: Entries are logistic regression coefficients with cluster robust standard errors in parentheses. Standard errors are clustered around individual MPs. *** p < 0.01, ** p < 0.05, * p < 0.1. To test the second hypothesis (H2), namely that previous SMD candidacy changes the effect of activities, I control for the interaction of activity and SMD candidacy at the previous election (Model 5). The interaction is significant, and as shown in Figure 1, reveals that in the cases of both former SMD candidates and those who only have been list candidates at the previous election activity has a positive effect on the probability of re-selection on a party list. Nevertheless, the two effects are not identical. In the case of former SMD candidates (dashed line), the curve is steeper indicating that the effect is larger in the case of those who were only nominated on party lists—and thus the party may not count on them as future SMD MPs. Hence, the parliamentary activities of prospective SMD candidates does not count as much at list nominations than that of list-only candidates, which confirms H2. Figure 1 View largeDownload slide The predicted probability of re-selection on a party-list. Figure 1 View largeDownload slide The predicted probability of re-selection on a party-list. 5. Conclusion The starting point of this study was that selectors aim at selecting legislators who are willing to carry out activities expected from their groups. In a mixed-member electoral system, SMD MPs are encouraged to work in their constituencies, while list MPs should focus more on parliamentary work. However, list MPs are not alike either: some are expected to run in SMDs at the next elections, which may change the balance between constituency and parliamentary work. Accordingly, it was expected that (H1) selectors do not take prior parliamentary activity into account when making the decision about SMD nominations, and that (H2) at party list nominations, parliamentary activities play a greater role in the case of MPs formerly not nominated in SMDs, than in the case former SMD candidates. Both hypotheses of the article have been confirmed: the mixed-effects regression models did not indicate the significant effect of activity in parliament on SMD re-nominations, but in the case of party list nomination, within the group of those formerly not having been selected in SMDs, more active legislators manage to improve their re-selection chances. These results indicate that even if only to a moderate extent parties evaluate MPs along how well they carry out the tasks dedicated to their groups. Nevertheless, the effect size indicates that MPs have to work enormously hard to considerably better re-selection prospects. The low effect size raises the question whether—despite the significant results—MPs are encouraged at all to working harder in parliament to achieve re-selection. Apart from parliamentary activities, parties value SMD incumbency and local political background on the SMD tier, which is nothing unexpected. Both, incumbency and local background increase local visibility, and are key for personal vote-seeking (Scheiner, 2005; Shugart et al., 2005; Tavits, 2010). The effect of parliamentary experience, however, is not what could be expected: due to the retirement of the most experienced MPs, the importance of seniority is not confirmed in the context of SMD re-selection. It has to be mentioned that in all models presented in this study, the random effects parameters suggest quite a large variation in the MP-level intercepts. This indicates that there are considerable differences between MPs and these affect both SMD and list tier re-selection. MPs are far from being equal in the race for re-nomination: parties clearly give advantage to some instead of others. Identifying all factors is beyond the scope of this study though. Nevertheless, as the majority of formal (hence measurable) factors are already taken into account, one could theorise that informal intra-party connections play key roles in nominations, which makes the selection process rather situational, therefore less predictable in a model setting. These informal positions, however are very hard to measure, thus their effects add to the error term within the models, leading to endogeneity related to omitted variables. Although prior list positions appear to be a solid proxy for this, we still have to face measurement challenges. The key term seems to be loyalty to the party. However, parties are notoriously united when it comes to roll-call (Ilonszki and Jáger, 2008), leaving little space for the researcher to categorise MPs on the basis of voting behaviour. The loyalty that matters here probably originates from the MPs’ personal connections to the party and its leadership. Although there are some attempts to sort out interpersonal relationships within parties (Tunkis, 2017), no systematic effort has gone into disentangling the web of intra-party connections. Intra-party personal connections are key to understanding selection patterns, and the pursuit of overcoming measurement difficulties gives us plenty to think about in the future. Funding This work was supported by the National Research, Development and Innovation Office (NKFIH) [PD 115747]. The author is a recipient of the János Bolyai Research Scholarship. Conflict of Interest The author declares that there is no conflict of interest regarding the publication of this article. Footnotes 1 Magyar Szocialista Párt; Hungarian Socialist Party. 2 Fidesz—Magyar Polgári Szövetség; Fidesz—Hungarian Civic Alliance. 3 Enyedi and Todosijevic (2009) show that the general level and the changing patterns of partisanship within the Hungarian electorate is characteristic to that of other Eastern European countries. 4 High levels of party unity are also very common in new democracies (Field, 2013). 5 The main independent variable (activity) was also computed using the standardized values of the frequency of the different activity types to control for the different magnitude in standard deviations. Running the models using the standardized version as independent variable does not change results. For the sake of easier interpretation, I use the raw activity number in the models presented in this article. 6 I do not take roll-call data and committee work into account. With regard to the first, the frequency of votes and voting costs is roughly the same across all MPs, not generating substantial differences between individuals. As to work in committees, although Hungarian MPs spend a considerable amount of time with tasks related to legislation in committees, committee work does not appear as individual activity; therefore, it is hard to measure it on the level of individual MPs. The number of committees in which the MP works may be indicative of the MPs’ workload related to committees. However, as it can only take a limited amount of (low) values, it does not add much to the average number of activities in terms of its variance. Including the number of committees into the models does not change results either. 7 Obviously, this could cause a problem of multicollinearity which is monitored throughout modelling. In none of the reported models goes the value of VIF above 2.73. The average value of VIF across all models varies between 1.2 and 1.4. 8 In Hungary, parties may jointly nominate SMD candidates and/or lists. During the period under investigation, parties joined and left electoral coalitions, which makes it difficult to separate the party effect. 9 The Dominant party variable also captures PPG-size effects. Legislators of larger PPGs can afford to work less, because the mandatory work distributes between a larger number of MPs. In small PPGs, however, a smaller number of MPs are responsible for the parliamentary work, which makes them end up with a larger workload. 10 As there is a rather substantive variation in the length of party lists, the absolute list position of the MPs on the lists is a misleading indicator of the MP’s intra-party position, and is difficult to compare across parties, or elections. Relative list position, on the other hand, takes into account the length of the party lists (i.e. the number of candidates on the party list), thus offers a more suitable measure to capture the candidate’s actual standing relative to other candidates of the party. The relative list position variable is calculated in the following way: Relative list position =List positionList length. Obviously, the higher the candidate climbs on the party list (i.e. the lower the rank), the smaller the value of the relative list position. 11 Here, I have to add that the MPs ideological distance from their parties is also considered a relevant factor in influencing re-selection chances. The rationale behind this is that the closer MPs are to their parties on the ideological spectrum, the better their positions at re-nomination. However, no suitable measures are available for the whole time period that could control for differences in ideology. Roll-call votes being strictly controlled by the party leadership throughout the whole period under investigation (Ilonszki and Jáger, 2008) are not good indicators of intra-party differences, while MP surveys are—apart from their sporadic availability—look at differences from the MPs’ viewpoints instead of how gatekeepers perceive these differences. Nevertheless, it can also be speculated that party list positions at the previous elections correlate to ideological differences between the party, in a sense that MPs higher on the party lists are closer to the ideological position of the party leadership. Furthermore, list positions incorporate other important factors that characterize intra-party relations making list positions a better indicator of these relations than ideological distance. 12 Apart from multicollinearity discussed earlier, the models have been checked for outliers. Standardized residuals were used to identify outliers. In the models, a few cases have been identified as having residuals larger than 2. Removing these observations do not change model results, thus they remain included. 13 Values in the text are the differences in predicted probabilities. References Anderson W. D. , Box-Steffensmeier J. M. , Sinclair-Chapman V. ( 2003 ) ‘The Keys to Legislative Success in the U.S. House of Representatives’ , Legislative Studies Quarterly , 28 , 357 – 386 . Google Scholar CrossRef Search ADS Atmor N. , Hazan R. Y. , Rahat G. ( 2011 ) ‘Candidate Selection’. In Colomer J. M. (ed.), Personal Representation. The Neglected Dimension of Electoral Systems , Colchester , ECPR Press , pp. 21 – 36 . Bowler S. ( 2010 ) ‘Private Members Bills in the UK Parliament: Is There an “Electoral Connection”?’ , The Journal of Legislative Studies , 16 , 476 – 494 . Google Scholar CrossRef Search ADS Bräuninger T. , Brunner M. , Däubler T. ( 2012 ) ‘Personal Vote-Seeking in Flexible List Systems: How Electoral Incentives Shape Belgian MPs’ Bill Initiation Behaviour’ , European Journal of Political Research , 51 , 607 – 645 . Google Scholar CrossRef Search ADS Cain B. , Ferejohn J. , Fiorina M. P. ( 1987 ) The Personal Vote: Constituency Service and Electoral Independence , Cambridge, MA , Harvard University Press . Google Scholar CrossRef Search ADS Carey J. M. ( 2007 ) ‘Competing Principals, Political Institutions, and Party Unity in Legislative Voting’ , American Journal of Political Science , 51 , 92 – 107 . Google Scholar CrossRef Search ADS Crisp B. , Ingall R. E. ( 2002 ) ‘Institutional Engineering and the Nature of Representation: Mapping the Effects of Electoral Reform in Colombia’ , American Journal of Political Science , 46 , 733 – 748 . Google Scholar CrossRef Search ADS Curtice J. , Shively P. ( 2009 ) ‘Who Represents Us Best? One Member or Many?’. In Klingemann H.-D. (ed.) The Comparative Study of Electoral Systems , Oxford , Oxford University Press , pp. 171 – 192 . Enyedi Z. , Bojan T. ( 2009 ) ‘Adversarial politics, civic virtues and party identification in Eastern and Western Europe’. In John B. , Paolo B. (eds) Political parties and partisanship: Social identities and individual attitudes , Oxon , Routledge , pp. 246 – 284 . Enyedi Zs. , Benoit K. ( 2011 ) ‘Kritikus Választás 2010. A Magyar Pártrendszer Átrendeződése a Bal-jobb Dimenzióban’. In Enyedi Z. , Szabó A. , Tardos R. (eds) Új Képlet. A 2010-es Választások Magyarországon , Budapest , DKMKA , pp. 17 – 42 . Enyedi Zs. , Tóka G. ( 2007 ) ‘The Only Game in Town: Party Politics in Hungary’. In Webb P. , White S. (eds) Party Politics in New Democracies , Oxford , Oxford University Press , pp. 147 – 178 . Google Scholar CrossRef Search ADS Fenno R. F. ( 1978 ) Home Style: House Members in Their Districts , Glenview, IL , Longman . Field B. N. ( 2013 ) ‘Resolute Leaders and “Cardboard Deputies”: Parliamentary Party Unity in the New Spanish Democracy’ , South European Society and Politics , 18 , 355 – 374 . Google Scholar CrossRef Search ADS Field B. N. , Siavelis P. M. ( 2008 ) ‘Candidate Selection Procedures in Transitional Polities: A Research Note’ , Party Politics , 14 , 620 – 639 . Google Scholar CrossRef Search ADS Gallagher M. ( 1980 ) ‘Candidate Selection in Ireland: The Impact of Localism and the Electoral System’ , British Journal of Political Science , 10 , 489 – 503 . Google Scholar CrossRef Search ADS Gallagher M. , Holliday I. ( 2003 ) ‘Electoral Systems, Representational Roles and Legislator Behaviour: Evidence from Hong Kong’ , New Zealand Journal of Asian Studies , 5 , 107 – 120 . Glazer A. , Wattenberg M. P. ( 1996 ) ‘How Will Term Limits Affect Legislative Work?’ [online]. In Grofman B. (ed.), Legislative Term Limits: Public Choice Perspectives , The Netherlands , Springer , pp. 37 – 46 . Retrieved from http://link.springer.com/chapter/10.1007/978-94-009-1812-2_3. Google Scholar CrossRef Search ADS Hazan R. Y. , Rahat G. ( 2010 ) Democracy within Parties: Candidate Selection Methods and their Political Consequences , Oxford , Oxford University Press . Google Scholar CrossRef Search ADS Heitshusen V. , Young G. , Wood D. M. ( 2005 ) ‘Electoral Context and MP Constituency Focus in Australia, Canada, Ireland, New Zealand, and the United Kingdom’ , American Journal of Political Science , 49 , 32 – 45 . Google Scholar CrossRef Search ADS Helmke G. , Levitsky S. ( 2004 ) ‘Informal Institutions and Comparative Politics: A Research Agenda’ , Perspectives on Politics , 2 , 725 – 740 . Google Scholar CrossRef Search ADS Hill K. Q. , Hurley P. A. ( 2002 ) ‘Symbolic Speeches in the U.S. Senate and Their Representational Implications’ , The Journal of Politics , 64 , 219 – 231 . Google Scholar CrossRef Search ADS Ilonszki G. ( 2012 ) ‘Role Stability in the Context of Institutional and Positional Change’. In Blomgren M. , Rozenberg O. (eds) Parliamentary Roles in Modern Legislatures , London , Routledge , pp. 162 – 183 . Ilonszki G. , Edinger M. ( 2007 ) ‘MPs in Post-Communist and Post-Soviet Nations: a Parliamentary Elite in the Making’ , The Journal of Legislative Studies , 13 , 142 – 163 . Google Scholar CrossRef Search ADS Ilonszki G. , Jáger K. ( 2008 ) ‘Erős Kormány—Gyenge Parlament? A Törvényhozási Kapacitás as Törvényhozási Teljesítmény Politikai Szociológiai Összefüggései’ , Századvég , 2008 , 119 – 142 . Katz R. S. ( 2001 ) ‘The Problem of Candidate Selection and Models of Party Democracy’ , Party Politics , 7 , 277 – 296 . Google Scholar CrossRef Search ADS Körösényi A. ( 2013 ) ‘Political Polarization and Its Consequences on Democratic Accountability’ , Corvinus Journal of Sociology and Social Policy , 4 , 3 – 30 . Krook M. L. ( 2010 ) Quotas for Women in Politics: Gender and Candidate Selection Reform Worldwide , Oxford , Oxford University Press . Lancaster T. D. , Patterson W. D. ( 1990 ) ‘Comparative Pork Barrel Politics’ , Comparative Political Studies , 22 , 458 – 477 . Google Scholar CrossRef Search ADS Lazardeux S. ( 2005 ) ‘“ Une Question Ecrite, Pour Quoi Faire?” The Causes of the Production of Written Questions in the French Assemblée Nationale’ , French Politics , 3 , 258 – 281 . Google Scholar CrossRef Search ADS Manow P. ( 2013 ) ‘Mixed Rules, Different Roles? An Analysis of the Typical Pathways into the Bundestag and of MPs’ Parliamentary Behaviour’ , The Journal of Legislative Studies , 19 , 287 – 308 . Google Scholar CrossRef Search ADS Marangoni F. , Tronconi F. ( 2011 ) ‘When Territory Matters: Parliamentary Profiles and Legislative Behaviour in Italy (1987–2008)’ , The Journal of Legislative Studies , 17 , 415 – 434 . Google Scholar CrossRef Search ADS Marjai E. ( 2012 ) ‘Candidate Selection Patterns in the 2010 Hungarian Parliamentary Elections’, MA Thesis, Budapest, Central European University. Martin S. ( 2011a ) ‘Parliamentary Questions, the Behaviour of Legislators, and the Function of Legislatures: an Introduction’ , The Journal of Legislative Studies , 17 , 259 – 270 . Google Scholar CrossRef Search ADS Martin S. ( 2011b ) ‘Electoral Institutions, the Personal Vote, and Legislative Organization’ , Legislative Studies Quarterly , 36 , 339 – 361 . Google Scholar CrossRef Search ADS Mayhew D. R. ( 1974 ) Congress: the Electoral Connection , 2nd edn , New Haven, CT , Yale University Press . Norris P. ( 1997 ) ‘The Puzzle of Constituency Service’ , The Journal of Legislative Studies , 3 , 29 – 49 . Google Scholar CrossRef Search ADS Norton P. ( 2002 ) ‘Introduction: Linking Parliaments and Citizens’. In Norton P. (ed.) Parliaments and Citizens in Western Europe , London , Frank Cass , pp. 1 – 18 . Norton P. , Wood D. ( 1990 ) ‘Constituency Service by Members of Parliament: Does It Contribute to a Personal Vote?’ , Parliamentary Affairs , 43 , 196 – 208 . Google Scholar CrossRef Search ADS Papp Zs. ( 2016 ) ‘Shadowing the Elected: Mixed-Member Incentives to Locally Oriented Parliamentary Questioning’ , The Journal of Legislative Studies , 22 , 216 – 236 . Google Scholar CrossRef Search ADS Rasch B. E. ( 2009 ) ‘Opposition Parties, Electoral Incentives and the Control of Government Ministers: Parliamentary Questioning in Norway’. In Ganghof S. , Hönnige C. , Stecker C. (eds), Parlamente, Agendasetzung und Vetospieler , VS Verlag für Sozialwissenschaften , pp. 199 – 214 . Google Scholar CrossRef Search ADS Russo F. ( 2011 ) ‘The Constituency as a Focus of Representation: Studying the Italian Case through the Analysis of Parliamentary Questions’ , The Journal of Legislative Studies , 17 , 290 – 301 . Google Scholar CrossRef Search ADS Scheiner E. ( 2005 ) Democracy without Competition in Japan: Opposition Failure in a One-Party Dominant State , Cambridge , Cambridge University Press . Google Scholar CrossRef Search ADS Shugart M. S. , Valdini M. E. , Suominen K. ( 2005 ) ‘Looking for Locals: Voter Information Demands and Personal Vote-Earning Attributes of Legislators under Proportional Representation’ , American Journal of Political Science , 49 , 437 – 449 . Google Scholar CrossRef Search ADS Siavelis P. ( 2002 ) ‘The Hidden Logic of Candidate Selection for Chilean Parliamentary Elections’ , Comparative Politics , 34 , 419 – 438 . Google Scholar CrossRef Search ADS Siavelis P. M. , Morgenstern S. ( 2008 ) ‘Political Recruitment and Candidate Selection in Latin America: A Framework for Analysis’. In Siavelis P. M. , Morgenstern S. (eds) Pathways to Power: Political Recruitment and Candidate Selection in Latin America , University Park, PA , Pennsylvania State University Press , pp. 3 – 37 . Stratmann T. , Baur M. ( 2002 ) ‘Plurality Rule, Proportional Representation, and the German Bundestag: How Incentives to Pork-Barrel Differ across Electoral Systems’ , American Journal of Political Science , 46 , 506 – 514 . Google Scholar CrossRef Search ADS Strøm K. ( 1997 ) ‘Rules, Reasons and Routines: Legislative Roles in Parliamentary Democracies’, Müller W. C. , Saalfeld T. (eds) Members of Parliament in Western Europe. Roles and Behaviour , London , Frank Cass , pp. 155 – 174 . Tavits M. ( 2009 ) ‘The Making of Mavericks Local Loyalties and Party Defection’ , Comparative Political Studies , 42 , 793 – 815 . Google Scholar CrossRef Search ADS Tavits M. ( 2010 ) ‘Effect of Local Ties on Electoral Success and Parliamentary Behaviour: the Case of Estonia’ , Party Politics , 16 , 215 – 235 . Google Scholar CrossRef Search ADS Tunkis P. J. ( 2017 ) ‘The Ties that Bind: Do Group Associations among Legislators Matter for Political Parties?’ , Problems of Post-Communism , 64 , 79 – 93 . Google Scholar CrossRef Search ADS Ward L. J. ( 1998 ) ‘Second-Class MPs? New-Zealand’s Adaptation to Mixed-Member Parliamentary Representation’ , Political Science , 1998 , 127 . © The Author(s) 2018. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For Permissions, please e-mail: journals.permissions@oup.com

Journal

Parliamentary AffairsOxford University Press

Published: Jan 31, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off