You could also have withdrawn the debate, because it is no longer relevant. Or am I mistaken? (Handelingen van de Tweede Kamer der Staten Generaal II 2013/14, no.9, item 16, p. 1).

• 1 Introduction

  One of the least noteworthy moments in the career of a member of parliament (MP) is perhaps the announcement that a 30-member debate they have requested has been withdrawn. This decision is announced pro forma by the Speaker during the agenda-setting meeting. The insignificance of this moment is further emphasised by the fact that the MP does not need to be present at that moment, does not need to justify their choice and that a withdrawal rarely leads to discussion.
  For political scientists, however, these moments offer unique insight into agenda setting – a crucial but often unobserved element of politics. The Dutch parliament is the only European parliament where decisions about the agenda are made in public. In other parliaments these decisions are made behind the closed doors of the Speaker’s office. This Dutch exception allows us to directly see politicians exercising their agenda-setting power (Döring, 1995). Furthermore, withdrawals provide insight into the decision to ultimately keep something of the parliamentary agenda: a kind of non-decision-making that is often even less visible (Lukes, 2004). Yet this is a crucial expression of power. If an issue is not discussed it is impossible to consider or discuss alternatives to the status quo (Otjes, 2019).
  Because in most European parliaments the agenda is decided behind closed doors (Döring, 1995), studies on parliamentary agenda setting are often based on oral or written questions from MPs to the executive (Green-Pedersen & Mortensen, 2010; Green-Pedersen & Stubager, 2010; Meijers & Van der Veer, 2019; Thesen, 2013; Van de Wardt, 2015; Vliegenthart & Walgrave, 2011). There are two key limitations to this kind of study. First, they do not examine how the plenary agenda is actually determined. In essence, they focus on a sideshow: MPs ask oral and written questions on issues that do not make it into the plenary. If an issue makes it onto the plenary floor, there is no need to ask written questions. Written questions rarely receive direct attention from actors outside of parliament (Green-Pedersen, 2010), and they do not have tangible consequences (Walgrave et al., 2007). Secondly, one cannot see how political parties interact with and respond to each other in the agenda-setting process (Otjes, 2019). Parliamentary questions allow us to see what issues parties focus on without any formal constraints. But it is precisely dealing with these constraints that makes agenda-setting research challenging: you can only see a party’s true priorities when they are making decisions under constraints, in particular when time is a scarce resource (Döring, 1995).
  The agenda-setting meeting of the Tweede Kamer therefore offers a unique possibility to study parliamentary agenda setting. However, this opportunity has been used remarkably rarely. Otjes (2019) is the only study that has analysed the agenda-setting meetings in detail. This study has two key limitations. First, it ignores the fact that agenda setting is a two-stage process, in which debates are requested and actually held. While Otjes (2019) studies the former in detail, it does not study which debates are actually held. It therefore cannot make definitive statements about control over the agenda, since 77% of the successfully requested 30-member debates are withdrawn before they are held. Understanding the conditions under which MPs cancel debates is crucial to understanding who effectively controls the plenary agenda. Secondly, that study approached agenda setting from the perspective of issue competition (e.g. Green-Pedersen, 2007). It neglects other complementary insights from the broader field of political science, in particular the relation between anti-elitism and the use of parliamentary instruments (Louwerse & Otjes, 2019; Otjes & Louwerse, 2021a). It is therefore relevant to inquire into the link between political style and the use of 30-member debates. Therefore, our central research question is, when do parliamentary groups cancel thirty-member debates?
  The rest of this article has the following structure: first, we discuss the existing literature on agenda setting focusing on issue competition and look into lessons we can learn from anti-elitism. We then look at the Dutch case and discuss the possibilities of and limitations to generalisation from this case. On the basis of the theoretical and case-specific discussion, we formulate two hypotheses. Next, we discuss our research design. We then look at the result of the analysis. In the conclusion, we discuss the broader theoretical relevance of our results.

• 2 Theory

  This study examines whether parties actually hold the debates that they request. This is part of a two-stage agenda-setting process. The first stage of this process, requesting (30-member) debates, has been studied in depth in Otjes (2019). The present study examines the second stage: actually holding these debates. We assume that some of the mechanisms behind requesting and holding debates are the same, specifically, our two key explanations: issue ownership and anti-elitism.

  2.1 Issue Ownership

  Most research studying parliamentary agenda setting focuses on issue competition. Issue competition understands politics as a struggle between political actors on the question of which policy issues should dominate the political agenda (Budge, 2015; Green-Pedersen, 2007; Inglehart & Klingemann, 1976; Petrocik, 1996; Robertson, 1976). In this view, political competition concerns the question of which issue is emphasised, rather than the direct confrontation on those issues. Issues on the political agenda are hierarchically ordered, with some issues receiving more attention than others. Once policy issues are on the political agenda, political actors are constrained, since political parties and MPs sense that they have to address the issues that are on the agenda (Green-Pedersen & Mortensen, 2010).
  Political parties compete to place particular issues higher on the political agenda. This motivation can be intrinsic or strategic: a radical right-wing populist party may truly believe that immigration is the greatest crisis facing a country and therefore focus all its parliamentary activity on this issue. It may also be persuaded that this issue benefits them strategically. Politicians have an almost intuitive understanding of which issues benefit them and their party and which do not (Carmines, 1991). Political parties want to draw attention to issues that show themselves as most competent to handle them (Petrocik, 1996) and prefer to make other parties speak on issues on which those parties are not competent (Walgrave et al., 2015).
  Political scientists mostly understand this from an electoral perspective (but see Green-Pedersen & Otjes, 2019). The crucial concept here is issue ownership (Walgrave et al., 2012): issue ownership is often understood as the association voters make between political parties and political issues. For example, most voters associate the environment with green parties and immigration with radical right-wing populist parties. This association partially entails the expectation of voters that specific parties offer the best solutions to specific issues or are most competent to deal with them (Walgrave et al., 2015). Therefore, if a policy issue is particularly salient during an election (e.g. climate), voters are more inclined to vote for the owner of this issue (e.g. a green party). Election results throughout recent decades can be explained increasingly by this process (Green-Pedersen, 2007, 2019).
  Political parties work to obtain and maintain issue ownership in their parliamentary work (Otjes & Louwerse, 2018). Parties reinforce the association between themselves and ‘their’ issues by taking visible initiatives on the issues they own (Green-Pedersen, 2010). They thereby signal to other political parties that those issues are ‘theirs’, marking their territory. They can also be used by parties to signal to other actors such as interest groups, party activists, journalists and voters that their MPs are ‘working’ on those issues (Otjes & Louwerse, 2018). A party risks losing issue ownership if it temporarily neglects an issue while another party takes initiatives on it (Holian, 2004). In this sense, the work in parliament is part of a ‘permanent election campaign between parties’ (Otjes & Louwerse, 2018). Parties are more likely to request parliamentary debates on issues that they own (Otjes, 2019).

  2.2 Anti-elitist Parliamentary Style

  In their analysis of parliamentary behaviour, Louwerse and Otjes (2019) distinguish between two styles of opposition parties. On the one hand, they see a constructive style where MPs use tools to influence policy: MPs submit amendments to legislation and write private members’ bills. On the other hand, they see a critical style where MPs use oversight tools to criticise the government and its policies. This is focused on the assessment of the appropriateness of government action (Auel, 2007, p. 500). This can be done through written and oral questions but also by requesting debates. The extent to which parties use these oversight tools is, in part, a function of the parties’ anti-elitism (Louwerse & Otjes, 2019; Otjes & Louwerse, 2021a). We follow Otjes and Louwerse (2021a) in focusing on anti-elitism rather than populism.¹x A brief note about populism is therefore warranted. We understand populism from the perspective of Mudde’s (2007) ideational approach. It is a thin ideology based on three claims: (1) That this people are virtuous and homogeneous. (2) That the current elite is corrupt and acts as one. And (3) that when populist politicians gain power the will of the people will become the basis of government policy. The thin nature of populism means that it can be combined with different ideologies that can fill in the terms ‘people’ and ‘elite’ differently. In our view, it is the anti-elitism of populist parties (i.e. point 3) that drives them to request parliamentary debates; that is, they use parliament to expose the mistakes of the government. Opposition parties that are anti-elitist but not necessarily populist (such as communist parties that are critical of the current ‘bourgeois’ elites but that, owing to their Marxist societal analysis, does not see the people as homogeneous).
  To understand the link between anti-elitism and the use of scrutiny tools, we can build on the difference between responsive and responsible politics developed by Mair (2011). Parties can act as responsible actors focused on changing policies through compromise and cooperation, cognisant of the ‘small margins’ of democratic politics. Parties can also focus on responsive politics. Central to responsive politics is the link between citizens and politicians. Louwerse and Otjes (2019) have applied this distinction to parliamentary politics. Politicians can use parliament as a platform to express public discontent with policies, and scrutiny tools lend themselves particularly well to this responsive style: parliamentary debates can put a spotlight on mistakes made by government actors. Anti-elitism and responsive politics go hand in hand: anti-elitist opposition parties see it as their role to express public discontent with government policies and therefore use their scrutiny tools more often: to voice their opposition to the parties in power, to direct attention to issues that the current government ignores or to expose incompetence and corruption of governing elites. This makes anti-elitist parties more likely to request parliamentary debates as it allows them to use the plenary floor as their bully pulpit.

• 3 Case Selection and Description

  This article studies 30-member debates in the lower house of the Dutch parliament. In the following section, we explain why we study the Netherlands and what exactly these 30-member debates entail.

  3.1 Tweede Kamer

  This study examines the Tweede Kamer der Staten-Generaal, the only lower house in Western Europe that has full control over its own agenda (Döring, 1995). In most parliaments the agenda is set by the presidium or the Speaker (Döring, 1995; Yamamoto, 2007, p. 63), in only rare cases does the plenary majority or even a plenary minority decide the agenda. The Tweede Kamer is therefore an exceptional case that allows us to gain insight into the agenda-setting process, which usually occurs behind closed doors. The fact that in the Tweede Kamer agenda setting occurs in the open makes it likely that the process is different from other countries. For one, requesting a debate is a public activity that in itself signals to other MPs, journalists and the wider public that the party ‘owns’ that issue. Still, the factors shaping the agenda are likely to transfer to other systems where MPs have to work through more opaque parliamentary decision-making processes to have their debate planned. The incentives of the parties (e.g. to schedule debates on issues that they own) are the same, although the procedures may be more likely to benefit majorities over minorities. Patterns are likely to be similar in the Finnish, Swedish, Danish and Norwegian parliaments and the German and Belgian lower houses because in these multiparty systems the parliamentary majority can overrule the Speaker or presidium (Döring, 1995).

  3.2 Thirty-member Debates

  We focus on 30-member debates. These are debates that can be put on the agenda with the approval of only 30 of 150 MPs. There are many different types of parliamentary debates, the most prominent ones being legislative debates, majority debates, reports on committee meetings and 30-member debates (Otjes & Louwerse, 2021b). Legislative debate concerns legislation, while the other three kinds of debate concern policy in more general terms. The first three debates require a majority to plan, the last a minority.²x Reports on committee meetings require a single MP to request it, but they also require a committee meeting, which requires a majority in the committee. Plenary debates are important not only because they allow parties to voice their opinion, but also for parties to propose motions that can then be voted on. Reports on committee meetings are short follow-ups to committee meetings meant specifically to allow for the introduction of motions.
  These 30-member debates are riddled with contradictions. They were originally called ‘urgent debates’ (spoeddebatten) as they would allow MPs and ministers to discuss urgent issues that sprang up in society. The low threshold set for requesting these debates brought forth many requests for them. Yet because of the large number of requests, it was often long before these debates were held.³x It takes, on average, 204 days (nearly 7 months) before a debate is held (data between 2012 and 2021). The urgency of the request often disappeared by the time the issue came up on the agenda, and opportunities to discuss the matter in other plenary or committee debates may have come up. Therefore, many debates were cancelled. As we will see in greater detail later, about 23% of these debates are held and 77% cancelled. On average, a debate is cancelled 180 days (almost 6 months) after requesting it. Since 2011, these debates have been called 30-member debates. This reflects the new consensus that these debates rank low in importance when compared with majority debates and therefore are not scheduled soon. Plenary debates are requested during the agenda-setting meeting, which is held every day at the start of the plenary meeting. A major agenda-setting meeting is held on Tuesday at the beginning of the parliamentary week. The Speaker also uses these planning meetings to announce which debates are cancelled by the MPs who requested them. The debate request we study here may share some similarity with urgent questions and interpellations that many parliaments have (Yamamoto, 2007, pp. 52, 59-61): here a minority of the parliament can ask a minister for information or clarification on government policy.⁴x The Tweede Kamer still has a separate procedure for interpellations, which also requires 30 members but is placed on the agenda with high urgency, and here the requesting MP has a special role. In 2012-2017 three interpellations were held, and in 2017-2021 two were held.
  We focus on two parliamentary terms to look at the withdrawal of these requests: 2012-2017 and 2017-2021. Thirty-member debates were introduced in May 2004, and since then only these two parliamentary terms have been completed. If the parliamentary term ends prematurely, debates scheduled to be held later in the parliamentary terms are likely to be cancelled. During these terms the Netherlands had a centrist Liberal-Labor coalition and a centre-right cabinet of the Liberal Party, the social-liberal Democrats ’66, the Christian Union and Christian-Democratic Appeal. The parties in parliament are listed in Table 2.

• 4 Hypotheses

  We base our hypotheses on both our preceding theoretical discussion and the specific characteristics of the Dutch systems and the 30-member debates. First, given the strong link between issue ownership and debate requests (Otjes, 2019), we can conclude that MPs request debates in order to signal to other political players (other parties, interest groups, media, citizens) that they are working on a specific issue. They are a way to build and maintain issue ownership (Otjes & Louwerse, 2018). Requesting a debate is a way of attracting attention,⁵x In the recent trend towards filming snippets of debates and spreading them through social media, even a debate request that is not fulfilled can be used to show voters that established parties are trying to keep specific issues off the plenary agenda. and once they are requested, parties are also more likely to actually hold debates on issues that they own. These are more important to the parties both intrinsically and strategically. Therefore, we expect that:

  1. Issue-ownership hypothesis: the more salient an issue is to a political party, the more likely the party is to maintain a minority debate on that topic.

  Second, as we saw previously, anti-elitist parties have an incentive to use scrutiny tools. Thirty-member debates, in particular, tend to revolve around democratic scrutiny rather than policymaking. The key question in these debates often is ‘is the position of the minister tenable?’ (Van der Heiden, 2006). This makes them important tools for opposition parties that focus on scrutiny. As Socialist Party MP Paul Ulenbelt said:

  It is important to discuss laws, but opposing the government by means of [thirty-member debates] is as important … It is our job to increase pressure. If we don’t poke and prod the coalition, they will become complacent.⁶x Peeperkorn, M. & Sitalsing, S. (11 October 2007) Spoeddebat omdat het op tv komt. Volkskrant. Translation by the authors.

  This also makes these ‘responsive’ opposition parties less likely to cancel debates; they are less likely to abandon the tools that allow them to voice discontent. Moreover, given the pressure on the plenary agenda, unfulfilled debate requests serve as currency when bargaining with the Speaker. In the plenary we can see some of this wheeling and dealing; removing a 30-member debate from the ever-growing list may be a reason for the Speaker to extend speaking times in another debate or to schedule that debate sooner. A large part of this wheeling and dealing, however, occurs outside of the plenary floor and cannot be studied systematically. If an earlier opportunity to propose the motion comes along, constructive parties may be more likely to withdraw their debate request. Therefore, we expect that:

  2. Anti-elitism hypothesis: the more anti-elitist a political party is, the more likely it is to maintain a minority debate.

• 5 Methods

  We coded all requests made for 30-member debates between the start of the parliamentary term in 2012 (20 September 2012) and the start of the election recess in 2021 (12 February 2021). Each 30-member debate was requested during an agenda-setting meeting. For each one we traced whether the debate was held or not: we checked whether a debate with the name that was entered in the long-term agenda after the request was part of the floor proceedings using the search engine of officielebekendmakingen.nl. At the end of both parliamentary terms all debate requests for 30-members were cancelled. This binary variable is our dependent variable.⁷x At the end of the 2017-2021 term, 56 debates were still on the roll. We ignore these debates in the following discussion as it is unclear whether these were held or not. Table 1 lists the descriptives of variables used in the analyses.
  We employ a number of independent variables: our first hypothesis is that the more a party prioritises the issue for debate, the less likely it is to retract the debate. To measure issue ownership, we used the election manifestos of parties. The share of a party’s manifesto that concerns a particular issue is a good predictor of the extent to which voters consider a party an issue owner (Walgrave & de Swert, 2007). The subjects of the debate requests and party election manifestos were all coded in the same scheme (the Comparative Agenda Project scheme, see Appendix 1). For the manifestos we relied on Green-Pedersen and Otjes (2019), and for the debate requests we coded these ourselves.⁸x To check the intercoder reliability we coded 60 requests (10%) twice. The Krippendorf’s alpha was 0.67, indicating ‘substantial’ agreement between the coders and low but acceptable reliability. This coding allowed us to link each debate request to the extent to which the requesting party prioritised the issue. We used the election manifestos from the start of the parliamentary term (2012 for 2012-2017 and 2017 for 2017-2021).
  Our second hypothesis is that parties that use more anti-elite rhetoric are more likely to maintain their debate requests. We measure this using the Chapel Hill Expert Survey (CHES) (Polk et al., 2017), which includes a variable that measures the extent to which anti-elite rhetoric is salient for the parties. This goes from 0 (not important at all) to 10 (extremely important). We assigned parties the level of anti-elite rhetoric from the CHES closest to the start of the parliamentary terms (2014 for the 2012-2017 and 2019 for 2017-2021). We checked the reliability of these results by using two alternative measures: the first is whether the party was classified as populist by the PopuList team (Rooduijn et al., 2019), and the second was whether the party used anti-elite rhetoric in their election manifestos using the list of Pauwels (2011).⁹x bedrieg*, *bedrog*, *verraad*, *verrad*, absurd*, arrogant*, belof*, beloof*, belov*, capitul*, kapitul*, consensus*, corrupt*, direct, elite*, establishm*, heersend*, kaste, klasse, leugen*, lieg*, maffia, meningsuit*, ondemocratisch*, ondemokratisch*, oneerlijk*, particrat*, politic*, propaganda*, regime*, schaam*, schand*, toegeven, traditio*, volk, waarheid*.

  Table 1 Descriptive Table

  Variable	Mean	Median	S.D.	Min.	Max.	Low	High
  Debate held	0.23	–	–	0	1	Not held	Held
  Seats	12.42	14.00	5.64	2	41	2 Seats	41 Seats
  Coalition	0.05	–	–	0	1	Opposition	Coalition
  Left-right distance	3.21	3.39	1.65	0.21	5.26	Close to coalition mean	Far from coalition mean
  CHES anti-elitism	4.91	5.92	2.89	1.00	9.91	Anti-elitism not important at all	Anti-elitism extremely important
  PopuList populism	0.43	–	–	0.00	1.00	Not populist	Populist
  Pauwels anti-elitism	0.16	0.12	0.12	0.04	0.48	No anti-elitist rhetoric	High anti-elitist rhetoric
  Share of manifesto	0.08	0.07	0.05	0.00	0.40	Low priority	High priority
  Number of debates on the list	56.04	48	31.58	0	120	No debates	Many debates
  Years into term	1.80	1.65	0.94	0.17	4.24	Early in term	Late in term
  Period = 2017	0.54	–	–	0.00	1.00	2012-2017	2017-2021

  N = 687.

  In addition to issue competition and anti-elitism, other factors also affect parliamentary behaviour. Here we present six control variables. First, government participation. Opposition parties have to use their parliamentary tools to reach their strategic and policy goals, while coalition parties have access to the tools and benefits of the executive to attain their goals. Their ministers can take executive actions to enact policy change and have easier access to the media that allows them to direct attention to an issue. Moreover, coalition parties have an interest in maintaining cohesion within the coalition internally (while fighting out policy disagreements behind closed doors). This ensures the stability of the coalition and the image of cohesion and competence in the eyes of voters (Boston & Bullock, 2012). Parliamentary debates highlight and politicise disagreements within the coalition. Opposition parties are therefore more likely to request debates than coalition parties (Otjes, 2019). We also expect that opposition parties are more likely to maintain minority debate requests than coalition parties. Therefore, we include whether or not parties are in the coalition (as a binary variable).
  Second, the ideology of parties may also play a role. Policy-driven opposition parties are less likely to propose policy change when the coalition is ideologically close to them. The greater the ideological difference between a party and the coalition, the more likely that they will disagree with actions of the coalition and therefore request debates to challenge their decisions or point out the consequences of their policies. This explains why parties that are further from the coalition request more debates (Otjes, 2019). We also expect that parties whose preferences are further from the coalition mean are more likely to maintain minority debate requests than parties whose preferences are closer to the coalition mean. We include distance from the government in the model. To this end we look at the absolute distance between the party and the coalition in terms of the left-right distance variable from the CHES (Polk et al., 2017). We compare the party position with the seat-weighted coalition mean (at the beginning of the period).
  The capacity of parliamentary party groups also differs. A parliamentary party group with more MPs will have more time to prepare and participate in debates than parties with fewer MPs. A smaller parliamentary party group may therefore choose to participate in fewer debates than a larger parliamentary party group. Given the agenda pressure that members of these smaller groups experience, smaller parliamentary party groups are less likely to maintain minority debate requests. Therefore, we include party size in the models (in terms of the number of seats they have in parliament when the debate was requested),
  We also include three controls related to the timing of the debate request. These are mechanisms specific to the second stage of agenda setting, because a key condition of holding a debate is the availability of time in the plenary. We expect that minority debates requested when there are fewer minority debates on the list of minority debates are more likely to be held than minority debates requested when there are more minority debates on the list of minority debates. In that case the parliamentary calendar simply offers more possibilities to hold debates. We therefore include the number of days in between the debate request and the start of the parliamentary term, the number of debates that were requested before the debate was requested and that were not cancelled or held yet. We can expect that minority debates requested earlier in the parliamentary term are more likely to be held than minority debates requested later in the parliamentary term. It simply is the case that there are opportunities to actually hold the debate if it is requested earlier. We also include a dummy for the parliamentary period in which the debate was requested. We expect that minority debates requested in the 2017-2021 parliamentary term are less likely to be held than minority debates requested in the 2012-2017 term. The last year of the 2017-2021 term saw the coronavirus crisis. In response to this, the parliamentary agenda was cleared, and many parties cancelled their debate requests.
  Since the dependent variable is binary, the hypotheses are tested by means of a logistic regression. The observations are not truly independent of each other but vary by party in different terms: for example, D66 in the 2017-2021 term has the same coalition status, distance to the government and anti-elitism scores for each of their debate requests. Therefore, we ran the analysis using standard errors clustered at the party-period level.

  Table 2 Debate Requests by Party by Period

  Period		2012-2017			2017-2021
  Abb.	Full Name (English)	Seats	Requests	Share Held (%)	Seats	Requests	Share Held (%)
  50PLUS		2	6	50.0	4	14	28.6
  CDA	Christian-Democratic Appeal	13	65	20.0	19	9	11.1
  CU	ChristianUnion	5	11	9.1	5	2	0.0
  D66	Democrats 66	12	42	7.1	19	8	12.5
  DENK	Think/Equal	–	–	–	3	11	63.6
  FvD	Forum for Democracy	–	–	–	2	4	25.0
  GL	GreenLeft	4	16	0.0	14	62	9.7
  PVV	Freedom Party	15	62	58.1	20	27	18.5
  PvdA	Labor Party	38	4	25.0	9	45	20.0
  PvdD	Party for the Animals	2	12	41.7	5	42	23.8
  SGP	Political Reformed Party	3	2	0.0	3	0	–
  SP	Socialist Party	15	95	26.3	14	86	18.6
  VVD	Liberal Party	41	4	0.0	33	1	0.0
  Total		150	319	27.3	150	311	19.3

  Seats at the start of the term; government parties have their seats in bold; 2017-2021 period excludes debates that were not yet held at the end of the term;
  ^a ‘Denk’ means ‘think’ in Dutch and ‘equal’ in Turkish.

  Table 3 CAP Themes and Debate Requests

  #	Theme	2012-2017		2017-2021
  		Share of Proposals (%)	Share of Debates Held (%)	Share of Proposals (%)	Share of Debates Held (%)
  1	Macroeconomics	4.4	35.7	2.9	22.2
  2	Civil Rights & Migration	9.1	13.8	6.1	26.3
  3	Health	19.7	44.4	18.3	10.5
  4	Agriculture	4.4	42.9	3.5	27.3
  5	Labour	4.7	20.0	5.5	23.5
  6	Education	4.4	7.1	4.8	20.0
  7	Environment	5.0	12.5	13.8	27.9
  8	Energy	2.5	25.0	4.8	20.0
  9	Transportation	2.5	25.0	3.5	9.1
  10	Law & Crime	10.7	20.6	7.4	17.4
  11	Social Welfare	11.0	31.4	4.5	21.4
  12	Housing	2.2	28.6	5.5	17.6
  13	Domestic Commerce	1.9	16.7	4.5	14.3
  14	Defence	2.2	28.6	2.6	12.5
  15	Technology	0.0	–	0.0	–
  16	Foreign Trade	0.9	33.3	1.0	0.0
  17	International Affairs	4.4	35.7	3.9	33.3
  18	Government Operations	8.8	14.3	6.8	14.3
  19	Nature	1.3	25.0	0.6	50.0
  	Total	100.0	27.3	100.0	19.3

  The 2017-2021 period excludes debates that were not yet held at the end of the term.

• 6 Results

  Before we turn to our regression results, it may be useful to look at some more descriptive patterns to understand which parties keep and which cancel their debate requests. Table 2 shows the pattern per party, and Table 3 shows the patterns per issue. Table 2 clearly shows that debate requests scale with party size and government status: coalition parties request only a handful of debates (28% or 5% of the total). Smaller parties also request fewer debates than larger parties. We can see that most parties abandon more than 75% of their debate requests: the only exceptions are the pensioners’ party, 50PLUS; the party of, by and for bicultural Dutch people, DENK; the radical right-wing populist PVV; and the animal advocacy party, PvdD. The parties that maintain their debates score relatively high on anti-elite rhetoric (mean of 6.6 on the CHES anti-elitism score compared with 3.0 of those who maintain less than 25%). Many other parties (GL, SGP, VVD, CU) cancel all their debate requests in one or both periods. This makes it likely that anti-elitism plays a major role here, although the regression analysis provided further on will allow us to get a better grasp of the underlying patterns.
  Table 3 shows the pattern per issue. It shows that most debates are requested on health, the environment, law and crime and social welfare. The share of debates held fluctuates strongly per issue and period. The highest share of debates held can be found for nature during 2017-2021. Here, one of the two requested debates was held. We also see a high share of debates held for healthcare (in 2012-2017), where four out of nine debates are actually held, but this falls to less than one in nine during 2017-2021. The lowest score of debates held is for foreign trade in 2017-2021, where none of the three debates actually requested was held.
  Table 4 reports the logistic regressions. We run three models to test the robustness of the anti-elitism variable. We look at three operationalisations of this variable: the CHES anti-elitism scale, the binary PopuList populism scale and the percentage of manifesto devoted to anti-elite words (Pauwels anti-elitism). In the Appendix 1 and 2, we test the robustness further by looking at some analyses without outliers and some interactions.

  Table 4 Logistic Regressions

  Variable	Model 1	Model 2	Model 3
  Seats	−0.06** (0.02)	−0.09** (0.03)	−0.07** (0.03)
  Coalition	1.09* (0.55)	0.89 (0.59)	1.07* (0.57)
  Left-Right Distance	−0.01 (0.08)	−0.04 (0.09)	0.14** (0.07)
  CHES Anti-elitism	0.19*** (0.04)
  PopuList Populism		1.09*** (0.37)
  Pauwels Anti-elitism			3.51*** (0.82)
  Share of manifesto	2.68** (1.37)	4.27*** (1.35)	3.46** (1.51)
  Years into term	−0.69*** (0.16)	−0.76*** (0.17)	−0.71*** (0.18)
  Number of debates on the list	−0.02*** (0.01)	−0.02*** (0.01)	−0.02*** (0.00)
  Period = 2017-2021	−0.01 (0.05)	−0.01 (0.07)	−0.07 (0.05)
  Constant	30.35 (109.81)	12.48 (138.46)	139.24 (96.18)
  Log pseudolikelihood	−279.83	−284.23	−282.85
  Pseudo R-squared	0.18	0.17	0.17
  N	630	630	630

  Standard errors clustered by party * period; 0.1 < * < 0.05 < ** < 0.01 < ***.

  Share of manifesto and share of debates held
  Based on Model 1 with 95% confidence interval and bar reflecting the share of manifesto.

  The issue-ownership hypothesis proposed that the choice to hold a requested debate reflects issue ownership. Parties are more likely to maintain debates if they ‘own’ the issue that the debate concerns. All three models point in the same direction, but the effect in Model 2 is stronger than in Models 1 and 3. Figure 1 visualises the results from Model 1. It shows that as parties prioritise an issue more, they are more likely to keep it on the parliamentary agenda. The percentage increases from just below 20% for issues parties do not mention at all in their manifesto to 40% for issues that are half a party’s manifesto. The uncertainty also increases strongly in the second half of Figure 1. This means that parties are more likely to maintain a debate on issues that comprise quarter of their manifesto compared with an issue that they do not mention in their manifesto at all. But beyond that there are too few cases to say anything with certainty.

  The question is why the results are stronger in Model 2 than in the other models. The reason for this is that we use a binary measure of populism that puts the PvdD in the non-populist category. The PvdD, however, maintained quite a high share of debates, in particular on agriculture, when this was almost a third of their manifesto. In Model 2 this is captured by the share of manifesto variable, while in Models 1 and 3 this is captured by the anti-elitism measures. This does show, however, that an important part of the result for the share of manifesto variable is driven by a limited number of cases. The cases that drive this effect appear to be PvdD and PVV, which devote a larger share of their manifesto to specific issues (agriculture and civil rights and migration). If we drop all the debates requested on issues that encompass more than 20% of the requesting party’s election manifesto (26 debates all requested by the PVV and PvdD), the coefficient for party priority is no longer significant in two out of three models (see Appendix 1). This clearly shows that the effect we see in Table 2 and Figure 1 is driven mainly by these outliers. All in all, the issue-ownership hypothesis finds only limited support.

  CHES anti-elitism and share of debates held
  Based on Model 1 with 95% confidence interval and bar reflecting the distribution of CHES anti-elitism.

  The anti-elitism hypothesis proposed that anti-elitist parties are more likely to keep debates on the books and that this reflects their political style, which emphasises scrutiny of the government instead of making policy. Figure 2 shows the relationship between anti-elitism and the share of debates. This increases from 10% of the debates for the least anti-elitist parties to 40% for the most anti-elitist parties. The robustness tests reflect the same patterns. If we use the PopuList binary populism measure, we see that the share of debates that were actually held increases from 17% for the non-populist parties to 33% for the populist parties. If we use the Pauwels dictionary-based measure of populism, we see an increase in the share of debates that were actually held, from 16% for the parties that do not use any anti-elitist words to 45% for the parties that use the most anti-elitist words (0.5% of the manifesto devoted to anti-elitist words). This provides strong and consistent evidence for the anti-elitism hypothesis. The evidence clearly shows that anti-elitist parties are more likely to keep their debates on the agenda.¹⁰x In the Appendix 2, we look at an interaction relationship between the anti-elitism of the proposing party and the share of the manifesto devoted to an issue. This shows that anti-elitist parties will actually hold a debate on an issue independent of how important it is for them, for parties that score low on anti-elitism, the relationship between the importance of an issue and the likelihood of holding the debate is present. But do note that among these parties the uncertainty is substantial among high levels of priority.
  We included a number of control variables in the model. A number of them concerned the party that made the request. Here we find only one persistent pattern: smaller parties tend to keep their debates on the agenda, while larger parties are more likely to cancel them. This runs counter to our expectation. Holding a debate that actually got on the agenda might be more important for smaller parties that are less visible, in general, and have less access to traditional media than for larger parties. For these parties getting 30-member debates on the books might also be more difficult because they need the support of other parties. The smallest parties hold 33% of the requested debates, while the largest parties hold only 6%. For coalition participation the results are not consistently significant: government parties maintain more debates than opposition parties, although this result is significant only in Models 1 and 3 and only at the 0.1 level. This suggests that if coalition parties propose a debate, they intend to hold it. For left-right distance the results are not even consistent in terms of the direction or significance. The poor results for ideological distance may be illustrated by the fact that the two parties that were furthest away from the government (SP and GL) behave in opposite ways. The SP maintains a fair share of their debates (more than 23%), while GL cancels most of the debates (92%).
  The two of three timing-related controls show more persistent patterns. The more debates that are already on the roll when a debate is requested, the more likely a debate is cancelled: a debate that is requested when the roll is clean has a 43% chance of being held; a debate that is requested when there are 120 debates on the list has a 7% chance of being held. Likewise, a debate that was requested on the first day of the new term has a 41% chance of being held compared with a debate that was requested 4 years into the term, which has only a 5% chance of being held. We find no difference for the two periods.

• 7 Conclusion

  We examined the conditions under which parties maintain or withdraw 30-member debate requests. We proposed two possible explanations: issue competition and an anti-elitist political style. We find that the choice to maintain a 30-member debate reflects the style of a political party. Anti-elitist parties tend to maintain these debates. These are the same parties that are more likely to vote against legislation and ask parliamentary questions. For them a 30-member debate has the same purpose: to direct attention to the failures of the sitting government. The results for issue competition are much weaker: we find that this pattern is driven mainly by two opposition parties (Party for the Animals and Freedom Party) with a very clear issue focus on requesting and holding debates on their ‘own’ issues.
  These results allow us to draw three conclusions: about the parliamentary agenda in the Netherlands, about the political styles of opposition parties and about the process of agenda setting. Previous research showed that between 1998 and 2017 opposition parties became more successful in making proposals for the agenda (Otjes, 2019). This study nuances those results in two ways: first, it shows that a large majority of successfully proposed debates are never held, namely four out of five 30-member debates. Debate requests show only part of reality. Yet the parties that actually hold 30-member debates are among the most vocal, anti-elitist opposition parties. This still means that 30-member debates give the opposition, and even relatively small, anti-elitist opposition parties, considerable agenda-setting power. This power is greater than that which they held before the introduction of 30-member debates and greater than that held in other countries. The keys to the plenary are held by the opposition. However, their enthusiasm has caused a lag between proposed and actual debate for as long as a year, thereby undermining the actual agenda control they give.
  Our results show that anti-elitist opposition parties are more likely to hold the debates they request. This adds to earlier research that shows a difference in political style between anti-elitist and other parties (Otjes & Louwerse, 2019, 2021a). These parties tend to use their parliamentary votes and their right to ask questions as means to criticise the government. Proposing to hold and actually holding 30-member debates fits with this strategy, which sees parliament as a bully pulpit to voice opposition rather than a marketplace to create majorities for policy change. If anything, the conundrum of this study is why ‘responsible’ opposition parties retract so many of their debates. Why do some parties ask for 30-member debates but never actually hold them? While we believe that anti-elitism clearly plays a role in the incentives, that opposition parties have to poke, prod and politicise issues, the behaviour of opposition parties willing to play ball is still obscured. It seems likely that some parties cancel 30-member debates in backroom negotiations with the Speaker and committee chairs about other plenary debates: in exchange for giving up a 30-member debate, a committee debate on the same issue may be planned earlier. Future research may consider delving deeper into the negotiations between parliamentary party groups and the Speaker by means of a qualitative study.
  The extent to which the results of this study can be extended to other countries is limited. The openness of the Dutch agenda-setting process, in terms of both its public nature and the rights that minorities have to set the agenda is unparalleled. Still, they may suggest patterns of agenda setting that are likely to occur in other countries. Anti-elitism is associated with the use of oversight tools in other parliaments, in particular with voting against government bills (Otjes & Louwerse, 2021a). These results make it likely that when more anti-elitist opposition parties in other countries have a chance to set the parliamentary agenda, they will be more likely to actually exploit that opportunity than less anti-elitist opposition parties. In addition to taking into account government-opposition dynamics and issue competition in agenda setting, future research into this subject may consider taking into account the oversight-oriented political style of ‘responsive’ anti-elitist parties.

References

• Auel, K. (2007). Democratic Accountability and National Parliaments: Redefining the Impact of Parliamentary Scrutiny in EU Affairs. European Law Journal, 13(4), 487-504. doi: 10.1111/j.1468-0386.2007.00380.x.

• Boston, J. & Bullock, D. (2012). Multi-Party Governance: Managing the Unity-Distinctiveness Dilemma in Executive Coalitions. Party Politics, 18(3), 349-368. doi: 10.1177/1354068810382937.

• Budge, I. (2015). Issue Emphases, Saliency Theory and Issue Ownership: A Historical and Conceptual Analysis. West European Politics, 38(4), 761-777.

• Carmines, E. G. (1991). The Logic of Party Alignments. Journal of Theoretical Politics, 3(1), 65-80. doi: 10.1177/0951692891003001005.

• Döring, H. (1995). Time as a Scarce Resource: Government Control of the Agenda. In H. Döring (Ed.), Parliaments and Majority Rule in Western Europe. Frankfurt: Campus Verlag.

• Green-Pedersen, C. (2007). The Growing Importance of Issue Competition: The Changing Nature of Party Competition in Western Europe. Political Studies, 55(3), 607-628. doi: 10.1111/j.1467-9248.2007.00686.x.

• Green-Pedersen, C. (2010). Bringing Parties into Parliament: The Development of Parliamentary Activities in Western Europe. Party Politics, 16(3), 347-369. doi: 10.1177/1354068809341057.

• Green-Pedersen, C. (2019). The Reshaping of West European Party Politics: Agenda-Setting and Party Competition in Comparative Perspective. Oxford: Oxford University Press.

• Green-Pedersen, C. & Mortensen, P. B. (2010). Who Sets the Agenda and Who Responds to It in the Danish Parliament? A New Model of Issue Competition and Agenda-Setting. European Journal of Political Research, 49(2), 257-281. doi: 10.1111/j.1475-6765.2009.01897.x.

• Green-Pedersen, C. & Otjes, S. (2019). A Hot Topic? Immigration on the Agenda in Western Europe. Party Politics, 25(3), 424-434. doi: 10.1177/1354068817728211.

• Green-Pedersen, C. & Stubager, R. (2010). The Political Conditionality of Mass Media Influence: When Do Parties Follow Mass Media Attention? British Journal of Political Science, 40(3), 663-677. doi: 10.1017/S0007123410000037.

• Holian, D. B. (2004). He’s Stealing My Issues! Clinton’s Crime Rhetoric and the Dynamics of Issue Ownership. Political Behavior, 26(2), 95-124. doi: 10.1023/B:POBE.0000035959.35567.16.

• Inglehart, R., & Klingemann, H.-D. (1976). Party Identification, Ideological Preference and the Left-Right Dimension among Western Mass Publics. In I. Budge, I. Crewe, & D. Farlie (Eds.), Party Identification and Beyond. Representations of Voting and Party Competition (pp. 243-273). London: Wiley & Sons.

• Louwerse, T. & Otjes, S. (2019). How Populists Wage Opposition: Parliamentary Opposition Behaviour and Populism in Netherlands. Political Studies, 67(2), 479-495. doi: 10.1177/0032321718774717.

• Lukes, S. (2004). Power: A Radical View. London: Palgrave Macmillan.

• Mair, P. (2011). Bini Smaghi vs. The Parties: Representative Government and Institutional Constraints. Robert Schuman Centre for Advanced Studies / EUDO – European Union Democracy Observatory No. RSCAS 2011/22.

• Meijers, M. J. & van der Veer, H. (2019). Issue Competition Without Electoral Incentives? A Study of Issue Emphasis in the European Parliament. The Journal of Politics, 81(4), 1240-1253. doi: 10.1086/704224.

• Mudde, C. (2007). Populist Radical Right Parties in Europe. Cambridge: Cambridge University Press.

• Otjes, S. (2019). ‘No Politics in the Agenda-Setting Meeting’: Plenary Agenda Setting in the Netherlands. West European Politics, 42(4), 728-754. doi: 10.1080/01402382.2019.1565735.

• Otjes, S. & Louwerse, T. (2018). Parliamentary Questions as Strategic Party Tools. West European Politics, 41(2), 496-516. doi: 10.1080/01402382.2017.1358936.

• Otjes, S. & Louwerse, T. (2021a). Do Anti-elitist Parties Use Their Parliamentary Tools Differently? Parliamentary Affairs, 74(3), 704-721. doi: 10.1093/pa/gsab027.

• Otjes, S. & Louwerse, T. (2021b). The Netherlands. Party Specialists Taking the Floor. In J. Fernandes, H. Bäck & M. Debus (Eds.), The Politics of Legislative Debate. Oxford: Oxford University Press.

• Pauwels, T. (2011). Measuring Populism: A Quantitative Text Analysis of Party Literature in Belgium. Journal of Elections, Public Opinion and Parties, 21(1), 97-119. doi: 10.1080/17457289.2011.539483.

• Petrocik, J. R. (1996). Issue Ownership in Presidential Elections, with a 1980 Case Study. American Journal of Political Science, 40(3), 825-850. doi: 10.2307/2111797.

• Polk, J., Rovny, J., Bakker, R., Edwards, E., Hooghe, L., Jolly, S., … Zilovic, M. (2017). Explaining the Salience of Anti-elitism and Reducing Political Corruption for Political Parties in Europe with the 2014 Chapel Hill Expert Survey Data. Research & Politics. doi: 10.1177/2053168016686915.

• Rooduijn, M., Van Kessel, S., Froio, C., Pirro, A., De Lange, S., Halikiopoulou, D., … Taggart, P. (2019). The PopuList: An Overview of Populist, Far Right, Far Left and Eurosceptic Parties in Europe. https://popu-list.org.

• Thesen, G. (2013). When Good News Is Scarce and Bad News is Good: Government Responsibilities and Opposition Possibilities in Political Agenda-Setting. European Journal of Political Research, 52(3), 364-389. doi: 10.1111/j.1475-6765.2012.02075.x.

• Van de Wardt, M. (2015). Desperate Needs, Desperate Deeds: Why Mainstream Parties Respond to the Issues of Niche Parties. West European Politics, 38(1), 93-122. doi: 10.1080/01402382.2014.945247.

• Van der Heiden, P. (2006). Leve het spoeddebat! In C. van Baalen (Ed.), Jaarboek Parlementaire Geschiedenis 2005. Den Haag: Sdu Uitgevers.

• Vliegenthart, R. & Walgrave, S. (2011). Content Matters: The Dynamics of Parliamentary Questioning in Belgium and Denmark. Comparative Political Studies, 44(8), 1031-1059. doi: 10.1177/0010414011405168.

• Walgrave, S. & De Swert, K. (2007). Where Does Issue Ownership Come from? From the Party or from the Media? Issue-Party Identifications in Belgium, 1991-2005. Harvard International Journal of Press/Politics, 12(1), 37-67. doi: 10.1177/1081180X06297572.

• Walgrave, S., Lefevere, J. & Tresch, A. (2012). The Associative Dimension of Issue Ownership. Public Opinion Quarterly, 76(4), 771-782.

• Walgrave, S., Soroka, S. and Nuytemans, M. (2007). The Mass Media’s Agenda-Setting Power: A Longitudinal Analysis of Media, Parliament and Government in Belgium 1993 to 2000. Comparative Political Studies, 41(6), 814-836.

• Walgrave, S., Tresch, A. & Lefevere, J. (2015). The Conceptualisation and Measurement of Issue Ownership. West European Politics, 38(4), 778-796. doi: 10.1080/01402382.2015.1039381.

• Walgrave, S., Van Camp, K., Lefevere, J. & Tresch, A. (2016). Measuring Issue Ownership with Survey Questions. A Question Wording Experiment. Electoral Studies, 42, 290-299. doi: 10.1016/j.electstud.2015.12.001.

• Yamamoto, H. (2007). Tools for Parliamentary Oversight. A Comparative Study of 88 National Parliaments. Geneva: Interparliamentary Union.

Appendix 1 Logistic Regressions Without Outliers

Variable	Model A1	Model A2	Model A3
Seats	−0.06** (0.03)	−0.08** (0.04)	−0.07** (0.03)
Coalition	1.03* (0.56)	0.83 (0.60)	1.01* (0.58)
Left-Right Distance	−0.02 (0.08)	−0.05 (0.09)	0.14** (0.07)
CHES Anti-elitism	0.19*** (0.04)
PopuList Populism		1.10*** (0.36)
Pauwels Anti-elitism			3.54*** (0.80)
Share of manifesto	3.15 (3.42)	4.45 (3.35)	4.01 (3.49)
Years into term	−0.68*** (0.17)	−0.75*** (0.18)	−0.71*** (0.18)
Number of debates on the list	−0.02*** (0.01)	−0.02*** (0.01)	−0.02*** (0.01)
Period = 2017-2021	−0.00 (0.06)	0.01 (0.07)	−0.06 (0.05)
Constant	10.05 (124.04)	−9.25 (148.44)	119.89 (107.65)
Log pseudolikelihood	−266.77	−271.13	−269.83
Pseudo R-squared	0.18	0.17	0.17
N	607	607	607

Without debate requests on issues that encompass more than 20% of the requesting party’s election manifesto. Standard errors clustered by party * period; 0.1 < * < 0.05 < ** < 0.01 < ***.

Appendix 2 Logistic Regressions With Interactions

Variable	Model A4	Model A5	Model A6
Seats	−0.06*** (0.02)	−0.08** (0.03)	−0.07** (0.03)
Coalition	1.03* (0.56)	0.83 (0.59)	1.02* (0.58)
Left-Right Distance	−0.02 (0.08)	−0.06 (0.10)	0.14** (0.07)
CHES Anti-elitism	0.24*** (0.07)
PopuList Populism		1.43*** (0.52)
Pauwels Anti-elitism			4.17*** (1.34)
Share of manifesto	6.17* (3.16)	5.82*** (1.67)	5.10** (2.11)
CHES Anti-elitism * Share of manifesto	−0.55 (0.53)
PopuList Populism * Share of manifesto		−3.91 (3.21)
Pauwels Anti-elitism * Share of manifesto			−8.00 (9.24)
Years into term	−0.69***	−0.77***	−0.71***
	(0.16)	(0.17)	(0.17)
Number of debates on the list	−0.02*** (0.01)	−0.02*** (0.00)	−0.02*** (0.01)
Period = 2017-2021	−0.01 (0.06)	−0.00 (0.07)	−0.07 (0.05)
Constant	23.58 (113.31)	−3.63 (139.29)	133.00 (98.76)
Log pseudolikelihood	−279.54	−283.77	−282.60
Pseudo R-squared	0.18	0.17	0.17
N	630	630	630

Without debate requests on issues that encompass more than 20% of the requesting party’s election manifesto. Standard errors clustered by party * period; 0.1 < * < 0.05 < ** < 0.01 < ***.

Share of manifesto, CHES anti-elitism and share of debates held. Black line is minimal anti-elitism; grey line is maximal anti-elitism.

Share of manifesto, PopuList score and share of debates held. Black line is not populist; grey line is populist.

Share of manifesto, Pauwels’ anti-elitism and share of debates held. Black line is minimal populism; grey line is maximal populism.

Noten

• * The authors would like to thank Cynthia van Vonno and the anonymous reviewers for their insightful comments and suggestions.

• 1 A brief note about populism is therefore warranted. We understand populism from the perspective of Mudde’s (2007) ideational approach. It is a thin ideology based on three claims: (1) That this people are virtuous and homogeneous. (2) That the current elite is corrupt and acts as one. And (3) that when populist politicians gain power the will of the people will become the basis of government policy. The thin nature of populism means that it can be combined with different ideologies that can fill in the terms ‘people’ and ‘elite’ differently. In our view, it is the anti-elitism of populist parties (i.e. point 3) that drives them to request parliamentary debates; that is, they use parliament to expose the mistakes of the government. Opposition parties that are anti-elitist but not necessarily populist (such as communist parties that are critical of the current ‘bourgeois’ elites but that, owing to their Marxist societal analysis, does not see the people as homogeneous).

• 2 Reports on committee meetings require a single MP to request it, but they also require a committee meeting, which requires a majority in the committee.

• 3 It takes, on average, 204 days (nearly 7 months) before a debate is held (data between 2012 and 2021).

• 4 The Tweede Kamer still has a separate procedure for interpellations, which also requires 30 members but is placed on the agenda with high urgency, and here the requesting MP has a special role. In 2012-2017 three interpellations were held, and in 2017-2021 two were held.

• 5 In the recent trend towards filming snippets of debates and spreading them through social media, even a debate request that is not fulfilled can be used to show voters that established parties are trying to keep specific issues off the plenary agenda.

• 6 Peeperkorn, M. & Sitalsing, S. (11 October 2007) Spoeddebat omdat het op tv komt. Volkskrant. Translation by the authors.

• 7 At the end of the 2017-2021 term, 56 debates were still on the roll. We ignore these debates in the following discussion as it is unclear whether these were held or not.

• 8 To check the intercoder reliability we coded 60 requests (10%) twice. The Krippendorf’s alpha was 0.67, indicating ‘substantial’ agreement between the coders and low but acceptable reliability.

• 9 bedrieg*, *bedrog*, *verraad*, *verrad*, absurd*, arrogant*, belof*, beloof*, belov*, capitul*, kapitul*, consensus*, corrupt*, direct, elite*, establishm*, heersend*, kaste, klasse, leugen*, lieg*, maffia, meningsuit*, ondemocratisch*, ondemokratisch*, oneerlijk*, particrat*, politic*, propaganda*, regime*, schaam*, schand*, toegeven, traditio*, volk, waarheid*.

• 10 In the Appendix 2, we look at an interaction relationship between the anti-elitism of the proposing party and the share of the manifesto devoted to an issue. This shows that anti-elitist parties will actually hold a debate on an issue independent of how important it is for them, for parties that score low on anti-elitism, the relationship between the importance of an issue and the likelihood of holding the debate is present. But do note that among these parties the uncertainty is substantial among high levels of priority.