Issue expansion and selective scrutiny – how opposition parties used parliamentary questions about the European Union in the national arena from 1973 to 2013

Abstract

The scrutiny exerted by opposition parties in European Union (EU) affairs crucially contributes to the functioning of democratic accountability in the EU. While our knowledge about the extent of control activities has increased, we know little about their content. This article investigates the policy issues that opposition parties address in parliamentary questions about the EU. I argue that Eurosceptic opposition parties follow vote-maximizing strategies and selectively scrutinize very general aspects of the EU. My expectation is tested using a unique data collection of written questions in the Danish Folketing (1973–2013). I find evidence that the content of questions has been broadened over time and that Eurosceptic parties strongly emphasize general EU matters. The results point to the importance of political parties in making democratic accountability in the EU work.

Keywords

EU Affairs issue emphasis parliamentary questions parliamentary scrutiny

Introduction

European integration has brought radical changes to domestic legislatures. Most striking is their loss of power towards EU institutions because of transfers of legislative authority from national parliaments to the European Commission, the European Council, the Council of the European Union and the European Parliament (Katz and Wessels, 1999; Brouard et al., 2012; Hix and Høyland, 2013). These shifts in authority have important implications for the way delegation and accountability function in the domestic legislative arena (Bergman, 2000). Among other things, legislatures are remote from EU decision-making. For a long time, national parliaments have suffered from ‘information asymmetries’ vis-a-vis members of the executive who vote and negotiate in the Council and European Council (Auel and Benz, 2007). While national parliaments' overall access to EU documents has improved over time, there are still large differences between government and opposition parties' access to information about EU politics. Government parties regularly receive relevant and timely information at informal party meetings. Opposition parties, however, tend to struggle to find information about what the national executive is doing and deciding at the European level (Holzhacker, 2002; Proksch and Slapin, 2011). This influences the democratic quality of the EU, but also party responsiveness to European multi-level governance. It seems that in order to make democratic accountability work, opposition parties would need to actively address EU policy making in their parliamentary oversight operations.

However, not all opposition parties have incentives to acknowledge the transfer of legislative authority to the EU. While the overall politicization of European integration in domestic politics has increased in recent decades, parties pursue very different strategies regarding attitudes towards the EU, issue emphasis and issue framing (Hooghe and Marks, 2009; Hoeglinger, 2016). I suggest that these differences will also be reflected in the scrutiny parties exert in EU affairs. By linking insights from EU politicization theory with the literature on parliamentary scrutiny in EU affairs, I argue that Eurosceptic opposition parties selectively scrutinize general aspects of the EU, such as treaties, institutions and enlargement. It is in these parties' interests that broad issues about the future of the EU are raised in public debate (Hertner, 2015). In their parliamentary scrutiny activities, they will therefore focus on policies about the EU (such as positions towards the purpose and future of European integration) and de-emphasize policies through the EU (such as votes and bargaining positions on EU legislation). To test my expectation, I analyse the content of written parliamentary questions about the EU in the Danish Folketing from 1973 to 2013. Denmark is a suitable case for studying a framework on partisan questioning dynamics. Most importantly, its party system is broadly representative of Europe in general, with a mix of pro-European and anti-European parties from both sides of the ideological spectrum. Before I analyse the differences between opposition parties, it is however necessary to investigate the expansion of policy issues as a whole. If parties do not ask EU-related questions about a broad range of different issues, it would be redundant to explain variation with respect to general aspects of the EU across parties. The first part of the analysis therefore provides evidence from count response regression models and policy issue diversity measurements. It shows that opposition parties ask questions about a wide range of different policy issues, especially since the ratification of the treaty of Maastricht. While this can be seen as a sign that parties wish to be involved in EU policy making, I find evidence for differences between parties. Results from fractional logit models show that Eurosceptic competitors tend to be more likely to scrutinize general aspects of the EU. This finding suggests that party competition at the national level has important implications for the functioning of democratic accountability in EU affairs. If Eurosceptic competitors, who tend to be the most active parties in scrutinizing the government in EU affairs, mainly focus on issues that are favourable to them, in other words, polity-related issues, accountability with regard to EU policy-making processes suffers. In sum, the results provide empirical evidence for the link between party strategic factors at the domestic level and democratic accountability in the EU.

Political parties and parliamentary scrutiny activities

Scholars with an interest in parliamentary involvement in EU affairs have typically analysed formal rules for national parliaments – treated as unitary actors – to control EU decision-making (Norton, 1996; Maurer and Wessels, 2001). Several studies take formal agreements on legislatures' rights in EU affairs (e.g. access to information, information processing, mandating rights) as their point of departure, and investigate the variation in committee organization and overall strength in EU affairs between member-state parliaments (Bergman, 1997; Raunio, 2005; Karlas, 2011; Winzen, 2012). In the process, more and more scholars have stressed the relevance of political parties to the realization of active parliamentary control in EU affairs (Holzhacker, 2002; Saalfeld, 2005; Raunio, 2009; Winzen, 2013; Finke and Herbel, 2015). If parties and their MPs do not make use of scrutiny instruments, the rights of parliaments to control the executive in EU affairs become ineffective. As a result, recent studies increasingly take parties' active use of EU-related parliamentary control instruments into account. They show that higher levels of conflict between parties about the EU increase the number of reasoned opinions submitted within the Early Warning Mechanism (Gattermann and Hefftler, 2015). With regard to parliamentary debates, we know that internal party cohesion in combination with public support for European integration tend to be a crucial driver of EU politicization (Auel and Raunio, 2014; Rauh, 2015). The only study that looks into EU-related questioning activities at the domestic level – investigating the case of France – does not find evidence that party-based indicators matter, and instead points to the nature of institutional procedures to explain the Europeanization of parliamentary questions (Navarro and Brouard, 2014). Quite surprisingly, the authors show that there are no differences in the number of questions tabled between government and opposition parties as well as Europhiles and Eurosceptics. Furthermore, scholars have investigated the use of parliamentary questions at the European level. Here, findings reveal that MEPs from national opposition parties and Eurosceptic parties are particularly active in asking questions in the European Parliament – also in efforts to alert the Commission to cases of non-compliance (Proksch and Slapin, 2011; Jensen et al., 2013).

In sum, a large body of research suggests that party political contestation over European integration plays an important role in the number of EU scrutiny activities. However, we are clearly missing information about the policy issues that are addressed in parties' control activities about the EU. Practically all studies treat the EU as a single issue.¹ This means that, for example, scholars do not differentiate between control activities that address EU-wide rules on genetically engineered food or the latest treaty adaptation, but approach them as belonging to one and the same policy area. Hence, our knowledge about parliamentary accountability in EU affairs is limited to the salience of control activities. We know how many debates, resolutions and questions have been submitted, but we lack information about the actual content of these activities. However, it is important to know which aspects of the EU are raised in parties' scrutiny activities. Findings show that issue emphasis strategies of political parties have implications for vote choice (Bélanger and Meguid, 2008; Green and Hobolt, 2008) and the quality of political debate more generally, since parties that selectively emphasize issues end up ‘talking past each other’ (Budge and Farlie, 1983: 24). Hence, knowledge about the policy issues that parties address in their EU control activities will contribute to our understanding of whether and how EU scrutiny contributes to a strengthening of democratic processes in the EU.

Selective issue emphasis in EU scrutiny activities

Many different aspects of the EU can be raised in parliamentary control activities. Parties can address very general issues such as EU treaties and institutions, or they can talk about very specific issues such as agricultural subsidies, recycling and social benefits for low-income families. Recent research shows that parties do not necessarily pay attention to the same aspects when they talk about the EU. Parties tend to choose the EU issues they wish to address selectively to fit their general strategies and ideological outlook (Helbling et al., 2010; Hutter and Grande, 2014; Senninger and Wagner, 2015). In addition, parties tend to frame EU issues selectively as being either economic or cultural matters (McLaren, 2006; Kriesi, 2007). Similar patterns are likely to occur with respect to issue emphasis in opposition parties' parliamentary questions about the EU. In particular, I expect mainstream and more extreme Eurosceptic opposition parties to act differently regarding the attention they devote to general (polity-related) and policy-specific EU issues. The differentiation between general and policy-specific issues takes up a recent conceptualization of European policy that distinguishes between policies about the EU and policies through the EU (Hertner, 2015). The former refers to general positions towards European integration: that is, whether a party is in favour or against the deepening and widening of the EU. This includes topics such as treaties, the main institutions of the EU and enlargement. The latter refers to specific pieces of legislation that are decided in the Council of the EU and the European Parliament: that is, party positions towards legislative proposals.

Mainstream parties are constrained by past and future government participation (Green-Pedersen and Mortensen, 2010). This implies that they are expected to explain and react to important decisions at the European level and thus regularly talk about EU policy making – Hertner's policies through the EU. In addition, most mainstream parties have benefited from the dynamics of European integration in terms of access to power and funding (Topaloff, 2012). As a result, mainstream parties control the central bodies of the EU and decide at the European level. I therefore expect that the goal of future government participation will give mainstream opposition parties incentives to deal with the details of EU policy making. Even if they are not able to wield influence because of their current status in opposition, they are not able to afford to neglect policy making at the EU level. They need to be up to date about current policies, are expected to have an opinion about them and might even present policy alternatives. Hence, mainstream opposition parties will ask questions about all sorts of policy issues through the EU. In addition, mainstream parties are not keen on discussing general questions about the EU. Most mainstream parties are much more in favour of European integration than the median voter, which leads them to keep silent about general topics such as the future of the EU, enlargement or treaties (Steenbergen and Scott, 2004). They will therefore limit the number of questions about the EU.

In contrast, extreme Eurosceptic parties lack access to the EU's centre of power, which gives them a leeway for being more critical towards the EU. I expect that they will broadly avoid becoming involved in technical and complicated discussions about policies through the EU, especially, since they are not held accountable for EU policy output and not expected to react to it. They will therefore ask fewer questions about specific policy issues but talk more about general aspects of European integration. In addition to the fact that extreme Eurosceptic parties are neither responsible nor accountable for EU policy outputs, they also have strategic incentives to focus on general debates about the EU. This is especially the case in Eurosceptic societies in which the median voter increasingly opposes European integration (De Vries, 2007; Van der Eijk and Franklin, 2007). Parties with critical attitudes towards the EU thus have a greater potential to mobilize the Eurosceptic electorate by making general discussions about the EU salient on the political agenda. These general debates about European integration are also more likely to be covered by news media (Hertner, 2015).

Apart from their mobilization potential, general EU aspects are also issues suited to use in attacking mainstream competitors. European integration is considered a wedge issue, which means that it cannot easily be subsumed in the left-right dimension and is likely to divide mainstream coalition governments and mainstream parties (Tzelgov, 2014; Van de Wardt et al., 2014). Extreme parties face no constraints of present or future government participation and are therefore likely to exploit the potential of general discussions about the EU to confront their mainstream competitors with questions about EU aspects that they would prefer to avoid talking about. Parliamentary questions in particular seem to be suitable for the emphasis of issues that divide mainstream competitors, as the instrument allows for the government to be addressed in a very direct way. In sum, general debates about the EU help extreme competitors to advertise their Eurosceptic outlook and offer them the opportunity to harm their mainstream competitors. My expectation concerning parties' EU issue emphasis is thus as follows:

H1: Eurosceptic opposition parties selectively emphasize general EU aspects in their parliamentary questions about the EU (selective issue emphasis).

To test this conjecture, I need to have an idea about policy issue variation in EU-related parliamentary questions. If all parties wished to address the same aspects of the EU, the investigation of the reasons that explain variation across political parties with respect to general matters would be redundant. Hence, I expect that the policy issues addressed in parliamentary questions have been expanded over time:

H2: The range of policy issues in parties' parliamentary questions about the EU has increased over time (issue expansion).

It is evident that the EU has evolved tremendously. The number of member-states has grown, new institutions have arisen and policy issues covered by EU decision-making have expanded from trade, initially, to almost all policy areas (Hooghe and Marks, 2001; Bartolini, 2005). Hence, the deepening of the political, economic and cultural integration of the EU links ever more policy issues – which have traditionally been decided at the domestic level – to decision-making at the European level. The literature pictures the task expansion of the EU quite comprehensively (e.g. Lindberg and Scheingold, 1970; Pollack, 1994; Börzel, 2005). Several studies investigate the expansion of EU authority to new policy areas and compare the importance of EU decision-making with the domestic level.² Research findings provide evidence that legislative authority has increasingly been transferred from the domestic to the European level (König et al., 2012). This implies that national legislatures deal with ever more pieces of legislation from a broadened set of policy areas that have their origin at the European level. In addition, there is ample evidence that political parties increasingly deal with European issues and apply European frameworks, for example, during European election campaigns (Wüst, 2009). This increased engagement has repercussions for public debate about Europe and attitudes towards the EU at the domestic level (Van der Brug and De Vreese, 2016).

EU scrutiny in Denmark

As this study investigates the case of Denmark, some features of the Danish political system have to be taken into account. I therefore discuss the Danish EU scrutiny system and underline the relevance of investigating parliamentary questions in the light of the dominant role of the European Affairs Committee (EAC). In addition, I provide information about the Danish minority parliamentarism, parliamentary questions and party landscape.

Parliamentary questions: As already mentioned, scholars in the early days of the EU scrutiny research focussed on the institutional adaptation of national parliaments in the form of EACs. This is also true for the Danish case, in which the EAC is considered as particularly powerful and acts at the heart of the EU coordination system (Maurer and Wessels, 2001; O'Brennan and Raunio, 2007). Apart from comprehensive access to information about the EU policy-making process, the Danish EAC stands out with regard to its mandating system (Jensen, 2003; Laursen, 2005). Committee members regularly meet with ministers to discuss the legislative agenda of the following meeting of the Council of the EU and (orally) decide on the government's proposed negotiation position. If a majority of committee members votes against the government, a new proposal has to be introduced. Hence, the Danish mandating system ensures that the government does not agree to EU proposals that have no majority support in the Danish Folketing (for a more detailed discussion see Jensen, 2003). While the strong involvement of the Danish parliament in the form of its EAC is generally seen as a good thing, more recent research has identified several problems caused by the central role of the EAC that might hinder EU scrutiny (Sousa, 2008; Christensen, 2015).

For example, the EAC targets issues discussed and decided in the upcoming meeting of the Council of the EU and other EU institutions (e.g. the European Parliament or the European Court of Justice)³. This implies that committee members are limited by the EU's legislative agenda and are not able to raise issues that they wish to discuss. Another point of criticism is the conduct of the mandating system, which follows a consensual logic and makes it very difficult for either political parties or individual MPs to impact on the government's opinion. The procedure mirrors the functioning of Danish parliamentarism, according to which minority governments, which are the rule rather than the exception, secure broad cooperation either through bloc majorities along ideological lines or, in the case of EU decision-making, with the support of pro-EU parties (Green-Pedersen and Thomsen, 2005). One problem with the mandating system is that it is seen as a restriction of parliamentary involvement, especially since only a very small group of 17 MPs (out of 179) is regularly involved in EU affairs committee meetings. Recent efforts to involve sectoral committees more in EU coordination have not shown the desired effects as yet (Christensen, 2015). In sum, the consensual style and tight agenda of committee meetings prevent parties from adding issues to the agenda and raising conflicting positions in committee meetings. Party representatives in the EAC who would wish to address other issues than those on the Council's agenda feel that there is no time to do so (Sousa, 2008).

Hence, while EU scrutiny in EACs is certainly a very important part of democratic accountability in the EU, it does not allow for an analysis of the EU-related issues that parties wish to address. I suggest that written parliamentary questions are better suited to indicate the content of parties' control activities. Questions have no limitations regarding the issues parties wish to address and the point in time they wish to raise them. Moreover, there are no differences between questions that are related to the EU and questions that address domestic issues (Folketinget, 2011; Christensen, 2015). Parliamentary questions, therefore, provide an insight into the degree to which political parties in parliament are willing to integrate the EU in their day-to-day scrutiny work, and complement research findings based on EU control activities in parliamentary committees.

In sum, the Danish Folketing has three questioning procedures: interpellations, questions to ministers and questions to ministers from committees. Written questions are part of questions to ministers and are generally known as §20 questions, as §20 of the standing orders determines the rules and requirements (Folketinget, 2011). Most importantly, any individual MP is entitled to ask a question, and there are (almost) no restrictions regarding the number of questions.⁴ A written question has to be briefly and concisely worded. The questioner is allowed to include a brief explanatory statement and may ask for a written or oral answer. Before 2007, there were no strict rules regarding the content of a written question. MPs were entitled to ask for information on any public matter of interest given that the information necessary to answer the question was immediately accessible to the minister and her staff. In 2007, a reform resulted in a large decrease in the number of questions tabled. Since then, questions have to concern ministers' opinions, whereas questions concerning factual information have to be addressed through committees. Whether a question complies with the rules or not is decided by the chairperson of parliament.

Party landscape: The time period under study begins with the accession of Denmark to the European Community in 1973, after a majority of Danish citizens had voted ‘yes’ in a referendum the year before. A general election in the year the Scandinavian country joined the EU fundamentally changed the Danish party system, doubling the number of parties represented in parliament from five to 10 (Green-Pedersen, 2014). Most strikingly, the radical right-wing Progress Party secured 16 percent of the vote share. Before the election, four parties had dominated Danish politics: the Conservative People's Party, the Liberals, the Social Democrats and the Social Liberal Party. The increase of parties represented in parliament destabilized Danish politics, with frequently changing minority governments (Elklit and Pedersen, 2003). The second electoral consequence with crucial implications for the Danish party system was the success of the right-wing Danish People's Party, a split-off of the Progress Party, from the mid-1990s onwards. The party stands out on its Eurosceptic and anti-immigration position and has been very successful in the two most recent elections in Denmark. However, Euroscepticism at the party level existed long before the Danish People's Party entered into parliament. Ever since Danish citizens voted for accession to the European Community in 1972, there have been political parties in parliament and political movements outside of parliament with negative and sceptical orientations towards European integration.⁵

While most parties that have been represented in parliament during the time period under study are (at least to some extent) critical towards European integration, today, four parties (the Conservative People's Party, the Liberals, the Social Democrats and the Social Liberals) are considered as fully supporting the EU. However, two of those parties, the Social Liberal Party and the Social Democratic Party, have had intensive and controversial internal debates about the EU and have adjusted the official party line over time (Knudsen, 2008). The most Eurosceptic competitors in the post-Maastricht era, besides the far-right Danish People's Party, are rather small. The far-leftist Unity List/Red-Green Alliance is not only in favour of a Danish withdrawal from the EU but also opposes the EU as such. The classical liberal New/Liberal Alliance appreciates the EU's free market but rejects most other regulations coming from the EU. Before the 1990s, communist parties such as the Left Socialists and the Socialist People's Party as well as the special interest competitor Justice Party of Denmark were the most critical towards European integration.

Table 1 presents an overview of all parties represented in parliament during the period under study and provides information about party family, duration of parliamentary representation and average vote and seat shares.

Table 1.

Party family and average vote/seat shares of parties represented in the Danish Folketing, 1973–2013.

Party name	Party family	Represented in parliament	Vote share (%) (avg).	Seats (avg.)
Centre Democrats	Social Democracy	1973–2001	4.6%	8
Christian Democrats/ People's Party	Christian democracy	1973–94; 1998–2005	2.5%	4
Conservative People's Party	Conservative	✓	12.6%	23
Common Course	Communist/Socialist	1987–88	2.0%	1
Communist Party of Denmark	Communist/Socialist	1973–79	2.9%	4
Danish People's Party	Right-wing	1995–2013	11.8%	21
Danish Social-Liberal Party	Liberal	✓	6.0%	11
Justice Party of Denmark	Special Issue	1973–75; 1977–81	2.3%	3
Left Socialists	Communist/Socialist	1975–87	2.4%	4
New/Liberal Alliance	Liberal	2007–2013	3.9%	7
Liberal Party of Denmark	Liberal	✓	18.8%	34
Progress Party	Liberal	1973–2001	8.1%	14
Social Democrats	Social Democracy	✓	31.2%	59
Socialist People's Party	Communist/Socialist	✓	8.6%	16
Unity List/Red-Green Alliance	Communist/Socialist	1994–2013	3.2%	5

Source: ParlGov database; ✓ = entire period (1973–2013).

In sum, the Danish party landscape presents an interesting case for a longitudinal study of EU scrutiny activities of political parties with mainstream and extreme Eurosceptic competitors representing different party families and ideological orientations.

Data and methods

To test my expectations about selective issue emphasis and the expansion of policy issues, I need information that allows me to identify the policy issues that are addressed in parties' parliamentary questions about the EU. The Danish Policy Agendas Project, which is part of the Comparative Agendas Project (CAP), has collected several data sets that capture the issue agendas of different political actors across several decades, including parliamentary questions tabled by political parties.⁶ Most importantly, the manually-coded CAP data set contains a dummy variable that states whether the unit of observation (the individual parliamentary question) is related to the EU. This variable is coded ‘1’ if the EU is explicitly mentioned in the text of the parliamentary question (e.g. in actors, institutions, legislation). In total, 5249 out of 95,910 questions tabled by opposition parties in the period between 1973 and 2013 are related to the EU. These questions build the basis of my analysis. In addition, two variables (main topic and subtopic levels) indicate the policy issue category assigned to the question. The main level differentiates between 21 broad categories, including policy areas such as macroeconomics, health or education. The main topic categories are divided into 236 subtopic categories that account for more specific policy issues. Examples of subtopic categories for the main topic ‘macroeconomics’ are ‘tax policy and tax reform’, ‘price control’ and ‘monetary policy’.⁷ The detailed coding on the subtopic level also allows me to identify parliamentary questions that address general aspects of the EU. The subtopic ‘EU’, which is part of the main topic category ‘foreign policy’, is assigned if a parliamentary question is related to general questions of the EU including its institutions, treaties, referendums or enlargement. Moreover, the data cover information about the party that has asked the question as well as the date when the question was tabled. In sum, the Danish CAP data set is well-suited to analysis of the expansion of issues over time, especially since the logic of the coding approach perfectly suits my differentiation between general and policy-specific issues related to the EU.

Issue expansion

The second hypothesis regarding the expansion of policy issues in opposition parties' parliamentary questions over time is tested in two different ways. First, I created a variable that indicates how many different subtopic categories have been addressed in parliamentary questions about the EU from all opposition parties during the year. Thus, the unit of analysis is the annual number of subtopics related to the EU (during 1973–2013). This leads to a total of 41 units of observation. The count response is then used as the dependent variable in the following analysis of issue expansion over time. Theoretically, the variable runs from 0 (no EU-related questions have been tabled) to 236 (EU-related questions with regard to all possible subtopic categories have been tabled). However, empirically, the variable never reaches its minimum and maximum value. Because of the dispersion of the variable, I estimate negative binomial regression models. Unlike Poisson models with a single parameter μ (the mean), the negative binomial regression adds a dispersion parameter α that allows for a wide variety of shapes.⁸ The following independent variables are included in the models. I test for time dependence using a linear predictor (years) and restricted cubic time splines that estimate the relationship between the independent and dependent variables as a piecewise function. The function is linear before the first and after the last knot and cubic polynomial in between. Because of the small sample size, I decided to include three knots instead of the five determined by Harrel's default percentiles (Harrel, 2015). Moreover, I include a dummy variable that takes into account the reform of questioning regulations in the Danish parliament in 2007. I also add the share of EU-related questions tabled in parliament. I expect a positive relationship between the relative number of questions about the EU and the number of issues addressed in these questions. Furthermore, addressing a wide range of policy issues in parliamentary questions about the EU might be a function of the number of interests represented in parliament. If more parties are holding seats in parliament, we might see more diversity with regard to the policy issues addressed. I therefore control for the number of parties represented in parliament.

In addition to count response models, I test the issue expansion hypothesis by calculating diversity scores. In contrast to the simple count of policy issues, diversity scores also consider the distribution across policy issues. I use Shannon's H, a well-established diversity index that has recently been used to investigate issue attention of political actors (Bischof, 2015; John and Jennings, 2010). It is a probabilistic measure of the relative concentration of observations across a defined number of categories. In this study, it considers the probability of parliamentary questions to fall within a specific subtopic category. A higher score implies a higher level of uncertainty in determining the category in which a parliamentary question is about to fall. Shannon's H can be expressed as follows

H' = (- 1) \sum_{i = 1}^{236} p (x_{i}) \ln (p (x_{i}))

(1)

Scores (H′) are estimated as the negative sum for all subtopic categories of the likelihood, p, that an object (x) falls within a particular subtopic category i, multiplied by the natural log of that likelihood.⁹ I follow Greene (2015) and transform Shannon's H into the effective number of issues (ENI) by exponentiating the diversity score. Hence, larger values of ENI represent more issue expansion.

Selective issue emphasis

The second dependent variable refers to the selective issue emphasis hypothesis, which states that Eurosceptic opposition parties are more likely to address general aspects of the EU than mainstream opposition parties. I use a fractional response variable that captures the annual proportion of EU-related questions assigned to the subtopic category ‘EU’ for each opposition party represented in parliament. This results in an unbalanced panel with 241 units of analysis. Fractional data raise several methodological issues. Most importantly, one must take the bounded nature of the data (0 ≤ y ≤ 1) into account. Linear models would generate predictors outside the unit interval and are therefore not suitable. Moreover, fractions of 0 and 1 are not only theoretical values but actually appear in my data. I decided to make use of the flexibility of generalized linear models. In particular, I follow Papke and Wooldridge (1996) and apply fractional logit models. The fraction y can be expressed as a function of covariates x

g (E (y)) = x β, y ~ F

(2)

where g(.) represents the link function and F is the distributional family. For the fractional logit model, this becomes

E [\log (y / (1 - y))] = x β, y ~ Bernoulli

(3)

My data have an unbalanced panel structure in which observations for the same units (parties) are observed over multiple points in time (years). This means that I have to account for correlation of error terms for parties over time. I decided to estimate population-averaged estimators with cluster-robust standard errors, which simply regresses my dependent variable on an intercept and my independent variables. Time-specific effects are assumed to already be captured in the regressors. Alternatively, one could estimate fixed or random effects models to deal with the correlation of error terms. However, fixed effects models limit the explanatory power of regressors with little within-variation. Thus, many of my regressors would be imprecisely estimated. Random effects models seem to be inappropriate as well. My models represent important indicators for cross-party variation (e.g. measurements based on party ideology), which makes it likely that unobserved individual-effects are correlated with the regressors.

As I assume Eurosceptic opposition parties to be more likely to address general aspects of the EU, my key independent variable captures party-based Euroscepticism. I draw on data from the Comparative Manifesto Project (MRG/CMP/MARPOR), which is widely used to identify party positions towards European integration (Williams and Spoon, 2015). More specifically, I use the difference between positive and negative quasi-sentences about European integration, both measured as a proportion of the total quasi-sentences in the manifesto. To ease interpretation, the variable is rescaled so that the values for party-based Euroscepticism are allowed to fall within 0 and 10 (5 marks a neutral position towards the EU). Empirically, the range goes from 3.7 to 5.7 with values above five representing negative positions towards the EU. Furthermore, I account for party extremism using the distance to the vote share-weighted mean of the left-right score from the CMP. The variable is rescaled as well and ranges from 5.01 to 9.05 with higher values representing more extreme party positions (5 marks the mean for all parties weighted by the share of votes).

I control for party size using the absolute number of seats in parliament. Moreover, I take opposition parties' distance from the government into account. As mentioned before, Denmark usually has minority governments that rely on the support of opposition parties from the same ideological bloc. A dummy variable is included that indicates whether an opposition party belongs to the same bloc as the parties in government. This allows me to control whether specifics of the Danish party system impact on questioning behaviour in EU affairs. In addition, I include a variable that accounts for the degree of the EU's legislative authority. The variable is an additive function of the scope, delegation and pooling of EU competences (see Biesenbender, 2011; Rauh, 2015). It includes the number of policy areas covered by EU primary law, the share of EU decision-making powers with a Commission right of initiative and the share of EU decisions subject to simple or qualified majorities in the Council. I expect to find a negative relationship, as higher levels of EU legislative authority should lead parties to focus less on general aspects of the EU. As in the test of the issue expansion hypothesis, I include the reform dummy. In addition, I control for potential changes in the behaviour of Eurosceptic parties after the reform by including an interaction term. Eurosceptic parties might focus even more on general aspects when they ask for ministers' opinions than when they ask factual questions. Finally, a dummy variable indicates whether a referendum about the EU has taken place in the given year. Referendums should increase the fraction of general aspects of the EU in parties' parliamentary questions. An overview of all relevant variables can be found in the online appendix.

Results

I begin my discussion with the issue expansion hypothesis and present results for the development of policy issues addressed in opposition parties questions about the EU. Figure 1(a) shows the number of subtopic issue categories addressed per year and suggests that there has been an increase over time. While the number of policy issues was relatively stable from Denmark's accession to the European Community in 1973 until the 1990s (around 10 to 20 issues per year), we see a steep rise thereafter. Many different subtopic categories have been addressed, especially following the ratification of the treaty of Maastricht (more than 60 different issue categories) and around the discussion of the treaty establishing a constitution for Europe. However, every increase in the number of policy issues is followed by a downturn. This is particularly visible during the years after the policy issue count reaches its maximum in 2006, but also the increases in the late-1970s, mid-1980s and mid-1990s are accompanied by decreases in the number of policy issues soon after. To account for differences in emphasis across policy issues, Figure 1(b) plots the mean ENI in parties' parliamentary questions about the EU based on Shannon's H.

Again, the results suggest that the content of EU-related parliamentary questions has become more diverse over time. While parties on average asked questions on less than five different policy issues before the ratification of the Single European Act, the ENI has clearly increased thereafter. The level of diversity reaches its maximum in 2005 with an average of 17 different policy issues per party. Subsequently, the number goes down again, which might be related to the reform of questioning regulations in the Danish parliament. The whiskers in Figure 1(b) represent the standard deviation of the ENI and suggest that not only has the mean ENI increased over time, but so too has variation across opposition parties. The online appendix provides more detailed information about the differences between parties. It shows that before the 1990s, it was mainly the Socialist People's Party, the Social Democratic Party and the Progress Party that were interested in asking questions on a broad set of policy issues. Subsequently, the two new Eurosceptic competitors, the Danish People's Party and the Unity List, were particularly active in addressing different aspects of the EU. After the reform in 2007, the ENI tends to go down again, especially for the two Eurosceptic competitors.

Table 2 takes up the descriptive information in Figure 1(a) and reports results from my regression models. All three are negative binomial models with the count of subtopic policy issue categories addressed in EU-related parliamentary questions per year as the dependent variable. I report exponentiated coefficients in order to have a change in the predictor value reflect a change in the number of policy issues addressed. The variable time1 represents the time period or years considered in this study. Model 1 shows that every year, opposition parties address 8.7 percent more policy issues. This linear effect of time is statistically significant in all three model specifications, however is smaller in model 3. The natural time spline time2 is significant in model 1. The marginal effect (not shown here) is positive until the year 1995 and turns negative after the turn of the millennium. However, adding control variables makes the coefficient become insignificant. The reform in 2007 has a negative but non-significant effect. While the relative number of EU-related questions asked per year has a significant positive effect on the dependent variable with a 9.1 percent increase for each one-unit change, the number of parties represented in parliament does not significantly contribute to the explanation of variation in the number of subtopic categories addressed. In sum, the results support my expectation that the number of policy issues have been expanded over time, and endorse further analysis about selective issue emphasis and the differences between political parties.

Table 2.

Results – Negative binomial models.

	Model 1	Model 2	Model 3
	EU issues (count)	EU issues (count)	EU issues (count)
time1	1.087***	1.084***	1.067***
	(0.009)	(0.009)	(0.008)
time2	0.945***	0.969	0.980
	(0.009)	(0.017)	(0.016)
reform 07'		0.672	0.714
		(0.141)	(0.142)
EU questions (%)			1.091***
			(0.023)
no. of parties			1.006
			(0.036)
lnalpha	−3.267	−3.475	−4.130
log-likelihood	−147.5	−145.8	−138.5
chi2	144.1	161.0	306.9
N	41	41	41

Exponentiated coefficients; Robust standard errors in parentheses.

*p < 0.05; **p < 0.01; ***p < 0.001.

I now turn to the results for the selective issue emphasis hypothesis. First, Figure 1(c) shows the overall share of general aspects of the EU in parliamentary questions over time. It also shows the share of pro-EU and anti-EU parties.¹⁰ The figure suggests that Eurosceptic parties constantly have higher shares of general questions about the EU than pro-European mainstream competitors. Even after the treaty of Maastricht, which lead to a clear increase in the policy issues addressed in EU-related questions, around 25 percent of Eurosceptic parties' questions deal with EU polity-related aspects. While the divide between Eurosceptics and Europhiles was dramatic during the 1970s and early 1980s, they seem to have become more similar over time. In sum, the figure suggests that the overall fraction of general EU aspects has decreased slightly over time.

Figure 2 reports coefficients from the fractional logit regression based on Model 3 in Table A2 in the online appendix, which includes all relevant regressors. It shows that party-based Euroscepticism has a significantly positive effect on the proportion of general aspects addressed in opposition parties' parliamentary questions about the EU. Thus, I find support that parties with more negative attitudes towards the EU are more likely to emphasize general EU matters than mainstream opposition parties. The coefficient for party extremism is positive but not statistically significant in any of the models. The direction of the coefficient, however, points in the expected direction, suggesting that parties further away from the mean ideological party position tend to ask more questions about general EU aspects than moderate competitors. Party size, being a member of the government bloc, the reform in 2007 as well as its interaction term with Euroscepticism and the referendum variable also have no significant effects. The positive coefficient of the bloc party variable suggests that opposition parties ask more questions on general aspects of the EU if they belong to the same bloc. This is a sign that parties are distinguishable by their position towards European integration even if they have a similar ideological outlook. The reform in 2007 does not show any significant changes in parties' questioning behaviour with respect to general EU aspects. Thus, even though the reform has reduced the number of questions asked, this has no effect on the share of general questions about the EU. This can also be seen in the fraction of general EU questions for individual parties reported in the online appendix, which shows no clear trend after the reform in 2007. While the fraction of general aspects for the Danish People's Party and the Unity List tends to move slightly upwards, the non-significant interaction term between the reform and Euroscepticism in the online appendix does not support the conclusion that the reform has biased the questioning behaviour of Eurosceptic parties towards more general EU matters. This leads me to conclude that the reform has reduced the ENI for some parties but has had no effect on the emphasis of general aspects of the EU. The result for the EU authority variable shows a significant negative effect, which implies that the fraction of parties' general EU questions decreases if the EU's legislative authority rises.

To more fully understand how party Euroscepticism acts on selective issue emphasis in parties' EU-related questions, Figure 3 plots the marginal effect of Euroscepticism on the fraction of general EU aspects. The value of 5 on the Euroscepticism scale (x-axis) can be interpreted as a neutral position towards the EU. It clearly shows that parties that are more sceptical towards the EU have a higher probability to focus on general aspects of the EU than Europhile mainstream parties.

Conclusion

The question of whether and how political parties scrutinize their government in EU affairs features prominently in the literature about parliamentary control in the EU. Especially opposition parties are considered as key actors able to contribute to the realization of democratic accountability in EU affairs. While many studies ask how active opposition parties actually are in controlling their government in EU affairs, we still know relatively little about the policy issues they address. In this study, I therefore analyse the content of opposition parties' EU-related parliamentary questions. In particular, I investigate differences between mainstream and Eurosceptic competitors with regard to the emphasis of general and policy-specific aspects of the EU. Moreover, I argue that parties have been responsive to the increase in the EU's legislative authority and expect that parliamentary questions about the EU have become more diverse over time. To test my theoretical expectations, I rely on a unique collection of written parliamentary questions tabled in the Danish Folketing. Results based on count and fractional logit regression models as well as diversity measures show that parties have indeed broadened the range of issues addressed in EU-related questions. Especially since the ratification of the treaty of Maastricht, parties have been increasingly responsive to the transfers of legislative authority from the domestic to the European level. While most opposition parties reach adequate levels of EU issue diversity, parties with Eurosceptic outlooks are indeed more likely to emphasize general aspects of the European Union in their parliamentary questions.

I believe that the results contribute to our understanding of democratic accountability in the EU. The upward trend in the level of issue diversity signals that opposition parties acknowledge the increased authority of the EU in the policy-making process. The result allows for a positive assessment of the functioning of democratic accountability in EU politics. The more policy issues that parties address in their parliamentary questions, the more information they will receive about decision-making in EU affairs, and the more pressure they will be able to put on the executive. In sum, the development of issue expansion in parliamentary questions suggests that domestic parliamentary actors wish to play an active role in EU decision-making. However, the fact that Eurosceptic parties are strongly emphasizing general aspects of the EU darkens the overall picture. The finding implies that the most active parties make strategic use of the scrutiny instruments at hand. Emphasizing issues about the EU supports extreme competitors in advertising their Eurosceptic outlook and helps them become more visible in public debates about the future of the EU. However, this behaviour does not actively contribute to holding the government's policy-decisions in EU affairs to account.

The wider implication of my finding is that political parties hold the reins in parliamentary control in EU affairs. Formal rules provide domestic parliamentary actors with opportunities; however, the extent to which these opportunities are actively used as well as the purpose of EU scrutiny depend on the incentive structures of political parties. This should be taken into account in the discussion about the future role of national parliaments in the policy-making process of the EU.

Finally, what do we learn from the Danish case? I have argued that Denmark is a suitable case for testing a framework of partisan questioning dynamics. This is certainly true. The Danish party landscape is informative because of the wide range of parties represented in the period under study. However, two specifics about the Danish case are important to keep in mind. First, the frequency of minority governments distinguishes Denmark from most other EU member-states. The results show that opposition parties' support to the government does not influence their EU-related questioning behaviour. However, one cannot rule out that the feature of Danish minority parliamentarism does not matter at all. Second, the Danish opt-outs from certain portions of the body of European Community Law might scale down the range of issues that parties address in their questions. I, therefore, suggest that future research investigates issue expansion and selective issue emphasis with respect to the EU from a comparative perspective.

Footnotes

Acknowledgements

The author would like to thank Daniel Bischof, Jens Blom-Hansen and Christoffer Green-Pedersen for valuable comments on earlier drafts of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Auel

Benz

(2007) Expanding national parliamentary control: Does it further European democracy? In: Kohler-Koch

Rittberger

(eds) Debating the Democratic Legitimacy of the European Union, Lanham: Rowman & Littlefield, pp. 57–74.

Auel

Höing

(2014) Parliaments in the euro crisis: Can the losers of integration still fight back? Journal of Common Market Studies 52(6): 1184–1193.

Auel

Raunio

(2014) Debating the state of the union? Comparing parliamentary debates on EU issues in Finland, France, Germany and the United Kingdom. Journal of Legislative Studies 20(1): 13–28.

Bartolini

(2005) Restructuring Europe: Centre Formation, System Building and Political Structuring Between the Nation-State and the European Union, Oxford: Oxford University Press.

Bélanger

Meguid

(2008) Issue salience, issue ownership, and issue-based vote choice. Electoral Studies 27(3): 477–491.

Bergman

(1997) National parliaments and EU affairs committees: Notes on empirical variation and competing explanations. Journal of European Public Policy 4(3): 373–387.

Bergman

(2000) The European Union as the next step of delegation and accountability. European Journal of Political Research 37(3): 415–429.

Biesenbender

(2011) The dynamics of treaty change – measuring the distribution of power in the European union. European Integration Online Papers 15(5): 1–24.

Bischof D (2015) Towards a renewal of the niche party concept: Parties, market shares and condensed offers. Party Politics. 23(3): 220–235.

10.

Blom-Hansen

Olsen

(2015) National parliamentary control of EU policy: The challenge of supranational institutional development. Journal of Legislative Studies 21(2): 125–143.

11.

Börzel

(2005) Mind the gap! European integration between level and scope. Journal of European Public Policy 12(2): 217–236.

12.

Brouard

Costa

König

(2012) The Europeanization of Domestic Legislators, New York: Springer.

13.

Budge

Farlie

(1983) Explaining and Predicting Elections - Issue Effects and Party Strategies in Twenty-Three Democracies, London: George Allen & Unwin.

14.

Christensen

(2015) The danish folketing and EU affairs: Is the Danish model of parliamentary scrutiny still best practice? In: Hefftler

Neuhold

Rozenberg

(eds) The Palgrave Handbook of National Parliaments and the European Union, Basingstoke: Palgrave Macmillan, pp. 275–289.

15.

De Vries

(2007) Sleeping giant: Fact or fairytale?: How European integration affects national elections. European Union Politics 8(3): 363–385.

16.

Elklit

Pedersen

(2003) Decembervalget 1973: 30 år efter. Politica 35(4): 365–376.

17.

Finke

Dannwolf

(2013) Domestic scrutiny of European Union politics: Between whistle blowing and opposition control. European Journal of Political Research 52(6): 715–746.

18.

Finke

Herbel

(2015) Beyond rules and resources: Parliamentary scrutiny of EU policy proposals. European Union Politics 16(4): 490–513.

19.

Folketinget (2011) Håndbog i Folketingsarbejdet, Copenhagen: Folketinget.

20.

Gattermann

Hefftler

(2015) Beyond institutional capacity: Political motivation and parliamentary behaviour in the early warning system. West European Politics 38(2): 305–334.

21.

Green

Hobolt

(2008) Owning the issue agenda: Party strategies and vote choices in British elections. Electoral Studies 27(3): 460–476.

22.

Green-Pedersen

(2014) Party-system development in Denmark: Agenda-setting dynamics and political Change. In: Green-Pedersen

Waalgrave

(eds) Agenda Setting, Policies, and Political Systems, Chicago: University of Chicago Press, pp. 69–85.

23.

Green-Pedersen

Mortensen

(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.

24.

Green-Pedersen

Thomsen

(2005) Bloc politics vs. broad cooperation? The functioning of Danish minority parliamentarism. Journal of Legislative Studies 11(2): 153–169.

25.

Greene Z (2015) Competing on the issues: How experience in government and economic conditions influence the scope of parties' policy messages. Party Politics. 22(6): 809–822.

26.

Harrel

FEJ

(2015) Regression Modeling Strategies – With Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis, New York: Springer.

27.

Helbling

Hoeglinger

Wüest

(2010) How political parties frame European integration. European Journal of Political Research 49(4): 495–521.

28.

Hertner

(2015) Is it always up to the leadership? European policy-making in the Labour Party, Parti Socialiste (PS) and Sozialdemokratische Partei Deutschlands (SPD). Party Politics 21(3): 470–480.

29.

Hix

Høyland

(2013) Empowerment of the European parliament. Annual Review of Political Science 16(1): 171–189.

30.

Hoeglinger

(2016) Politicizing European Integration – Struggling with the Awakening Giant, Basingstoke: Palgrave Macmillan.

31.

Holzhacker

(2002) National parliamentary scrutiny over EU issues: Comparing the goals and methods of governing and opposition parties. European Union Politics 3(4): 459–479.

32.

Hooghe

Marks

(2001) Multi-Level Governance and European Integration, Lanham: Rowman & Littlefield.

33.

Hooghe

Marks

(2009) A postfunctionalist theory of European integration: From permissive consensus to constraining dissensus. British Journal of Political Science 39(1): 1–23.

34.

Hutter

Grande

(2014) Politicizing Europe in the national electoral arena: A comparative analysis of five west European countries, 1970–2010. Journal of Common Market Studies 52(5): 1002–1018.

35.

Jensen

Proksch

Slapin

(2013) Parliamentary questions, oversight and national oppositions status in the European parliament. Legislative Studies Quaterly 37(2): 259–282.

36.

Jensen

(2003) Europaudvalget - Et Udvalg i Folketinget, Aarhus: Aarhus Universitetsforlag.

37.

John

Jennings

(2010) Punctuations and turning points in British politics: The policy agenda of the Queen's speech, 1940–2005. British Journal of Political Science 40(3): 561–586.

38.

Karlas

(2011) Parliamentary control of EU affairs in central and eastern Europe: Explaining the variation. Journal of European Public Policy 18(2): 258–273.

39.

Katz

Wessels

(1999) The European Parliament, the National Parliaments, and European Integration, Oxford: Oxford University Press.

40.

Knudsen

ACL

(2008) Euroscepticism in Denmark. In: Taggart

Szczerbiak

(eds) Opposing Europe? The Comparative Party Politics of Euroscepticism, Oxford: Oxford University Press, pp. 152–167.

41.

König

Dannwolf

Luetgert

(2012) EU legislative activities and domestic politics. In: Brouard

Costa

König

(eds) The Europeanization of Domestic Legislators, New York: Springer, pp. 21–38.

42.

Kriesi

(2007) The role of European integration in national election campaigns. European Union Politics 8(1): 83–108.

43.

Laursen

(2005) The role of national parliamentary committees in European scrutiny: Reflections based on the Danish case. Journal of Legislative Studies 11(3-4): 412–427.

44.

Lindberg

Scheingold

(1970) Europe's Would-Be Polity: Patterns of Change in the European Community, Englewood Cliffs, NJ: Prentice-Hall.

45.

Maurer

Wessels

(2001) National Parliaments After Amsterdam: Losers or Latecomers?, Baden-Baden: Nomos.

46.

McLaren

(2006) Identity, Interests and Attitudes to European Integration, Basingstoke: Palgrave Macmillan.

47.

Navarro

Brouard

(2014) Who cares about the EU? French MPs and the Europeanisation of parliamentary questions. Journal of Legislative Studies 20(1): 93–108.

48.

Norton

(1996) National Parliaments and the European Union, London: Frank Cass.

49.

O'Brennan

Raunio

(2007) National Parliaments within the enlarged European Union. From Victims of Integration to Competitive Actors?, Milton Park: Routledge.

50.

Papke

Wooldridge

(1996) Econometric methods for fractional response variables with an application to 401(k) plan participation rates. Journal of Applied Econometrics 11: 619–632.

51.

Pollack

(1994) Creeping competence: The expanding agenda of the European community. Journal of Public Policy 14(2): 95–145.

52.

Proksch

Slapin

(2011) Parliamentary questions and oversight in the European Union. European Journal of Political Research 50(1): 53–79.

53.

Rauh

(2015) Communicating supranational governance? the Salience of EU Affairs in the German Bundestag, 1991–2013. European Union Politics 16(1): 116–138.

54.

Raunio

(2005) Holding governments accountable in European affairs: Explaining cross-national variation. Journal of Legislative Studies 11(3-4): 319–342.

55.

Raunio

(2009) National parliaments and European integration: What we know and agenda for future research. Journal of Legislative Studies 15(4): 317–334.

56.

Saalfeld

(2005) Deliberate delegation or abdication? Government backbenchers, ministers and European union legislation. Journal of Legislative Studies 11(3-4): 343–371.

57.

Senninger

Wagner

(2015) Political parties and the EU in national election campaigns: Who talks about Europe, and how? Journal of Common Market Studies 53(6): 1336–1351.

58.

Sousa

(2008) Learning in Denmark? The case of Danish parliamentary control over European union policy. Scandinavian Political Studies 31(4): 428–447.

59.

Steenbergen

Scott

(2004) Contesting Europe? the salience of European integration as a party issue. In: Marks

Steenbergen

(eds) European Integration and Political Conflict, Cambridge: Cambridge University Press.

60.

Topaloff

(2012) Political Parties and Euroscepticism, New York: Palgrave Macmillan.

61.

Tzelgov

(2014) Cross-cutting issues, intraparty dissent and party strategy: The issue of European integration in the House of Commons. European Union Politics 15(1): 3–23.

62.

Van de Wardt

De Vries

Hobolt

(2014) Exploiting the cracks: Wedge issues in multiparty competition. The Journal of Politics 76(4): 986–999.

63.

Van der Brug

De Vreese

(2016) (Un)intended Consequences of European Parliamentary Elections, Oxford: Oxford University Press.

64.

Van der Eijk

Franklin

(2007) The sleeping giant: Potential for political mobilization of dissatisfaction in Europe. In: Van der Brug

Van der Eijk

(eds) European Elections and Domestic Politics: Lessons from the Past and Scenarios for the Future, Notre Dame, IN: University of Notre Dame Press, pp. 189–208.

65.

Williams

Spoon

(2015) Differentiated party response: The effect of Euroskeptic public opinion on party positions. European Union Politics 16(2): 176–193.

66.

Winzen

(2012) National parliamentary control of European Union affairs: A cross-national and longitudinal comparison. West European Politics 35(3): 657–672.

67.

Winzen

(2013) European integration and national parliamentary oversight institutions. European Union Politics 14(2): 297–323.

68.

Wüst

(2009) Parties in European parliament elections: Issues, framing, the EU, and the question of supply and demand. German Politics 18(3): 426–440.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.10 MB

0.02 MB

1.80 MB

1.42 MB