The impact of EU decision-making on national parties’ attitudes towards European integration

Abstract

This study examines the implications of EU decision-making for national parties’ support for European integration. We argue, first, that EU decision-making promotes the support of parties that were closely located to EU outcomes. Second, we expect higher support of parties with governmental experience due to their access to key offices in EU decision-making. For the period from 1979 to 2012, we measure and interact both variables with the political, parliamentary and decentralizing nature of EU acts. In addition to the expected effects of outcome distance and government participation, we find that non-bureaucratic legislation generally improves national parties’ attitudes towards European integration. The involvement of the European Parliament increases the support of more distant parties, but discourages parties with high governmental experience. Finally, peripherally located (government) parties appreciate the decentral discretion of directives.

Keywords

EU decision-making European integration governmental experience legislative outcomes partisan support

Introduction

Do national parties’ attitudes towards European integration respond to the distributional and informational implications of European Union (EU) decision-making? And do bureaucratic, executive and central acts promote the support of government parties, while political, parliamentary and decentral acts rather meet the approval of opposition parties? The rise of Euroscepticism has drawn considerable scholarly attention to the sources of national parties’ attitudes, which are explained by the relationship to the voters from both demand and supply side (Golder, 2016). Demand-side arguments refer to economic and societal problems in Western and Eastern Europe, in particular the voters’ fears for the future (Lucassen and Lubbers, 2012) and immigration concerns (Ivarsflaten, 2008). In contrast, supply-side explanations focus on electoral rules (Norris, 2005), party organization (Tavits, 2012) and programmatic attractiveness (Rovny, 2013). However, scholars know remarkably little about the relation between EU decision-making and national parties’ attitudes.

This study introduces an alternative view of European integration, according to which EU decision-making plays a central role for the support of national parties as the EU’s key actors. In accordance with the literatures on cartel parties (Katz and Mair, 2009) and the ‘transformation of national political spaces’ (Kriesi et al., 2008: 3–9), we motivate this rationale with two developments. First, EU decision-making covers a growing number of policy fields (Hix and Høyland, 2011) and increasingly affects the legislatures of the Member States, in which electoral party competition is a dominant characteristic of democratic governance. Second, national parties are key actors for selecting representatives of EU institutions, send important messages about EU activities to the voters (Gabel et al., 2002) and play a vital role in referendums on treaty amendments.

Following the distributive theory of legislative organization (e.g. Weingast and Marshall, 1988), we conceive EU decision-making as a collection of party policy positions trying to divide a pie in an (bicameral) institutional framework that provides a way to enforce prior agreements (Groseclose and King, 2001). As a consequence, we expect that national parties respond positively to European integration when they have experienced the inclusion of their policy positions in the outcomes of EU decision-making. Due to the high hurdles of EU decision-making, such outcomes are more likely to reflect the interests of national parties which pursue centrally located policy positions on an electorally important dimension, such as left versus right (Crombez and Hix, 2015).

With respect to the informational theory of legislative organization (Gilligan and Krehbiel, 1987), we can also argue that EU decision-making provides access for political parties to acquire special expertise in offices. This may explain the strengthening of the responsibility orientation of parties at the expense of their responsiveness to the voters’ interests (Caramani, 2017). We accordingly expect that national parties are more in favour of European integration when they increased their informational advantages vis-à-vis their domestic competitors through holding offices in EU institutions (cf. Williams and Spoon, 2015: 180). By and large, we hypothesize that national parties express higher support for European integration when they profited from outcome closeness and informational access to EU institutions.

To complete our argument on the relationship between EU decision-making and national parties’ attitudes towards European integration, we go a step further and interact the distributional and informational variables with distinct characteristics of EU legislative acts. We thereby assume that national parties know the pros and cons of each type of act that may (dis)advantage (non)governmental parties with (non)central policy positions. In this vein, we distinguish between political acts in secondary legislation and bureaucratic acts in tertiary legislation, the latter conferring powers on the Commission (Junge et al., 2015). A second distinction refers to the participation of the European Parliament, which offers access to EU decision-making for non-governmental parties. Finally, we separate regulations and decisions from directives, the latter providing for discretion in the transposition into national law. To mitigate endogeneity concerns we evaluate national parties’ attitudes towards European integration in retrospect. Concretely, we cumulate the past experiences with EU decision-making and divide by the number of years.

Our empirical analysis includes 3570 observations of national parties in 27 Member States, which declared their programmatic statements before each election in the period from 1979 to 2012. Using the EULIS data (König and Luig, 2012) on party positions, legislative saliencies and governmental office-holding we generate our dependent and two main independent variables, which we interact with the characteristics of about 90,000 EU legislative acts. Multilevel analyses with country-specific groups and several control variables provide evidence for our argument on the implications of EU decision-making. National parties that were closer located to EU legislative outcomes and acquired more governmental experience significantly express a higher support for European integration.

In addition to the empirical evidence of our argument on the distributional and informational implications of EU decision-making, we reveal that secondary legislation generally improves national parties’ attitudes towards European integration. In contrast, we find compensating effects of parliamentary involvement and directives, which provide access and deliver information to otherwise excluded national parties. Furthermore, we observe a discrepancy between public and partisan support for European integration. When parties need to trade off their vote- and policy-seeking interests, this suggests that the advantages of EU decision-making provide an incentive for national parties to sacrifice their responsiveness to the voters for their responsibility role in European integration.

Theory and implications: Why EU decision-making matters

Compared to the growing literature on power distribution and policy-making in the EU, there is a remarkable lacuna on our knowledge about the relationship between EU decision-making and national parties’ attitudes towards European integration. This is surprising for at least two reasons. First, the parlamentarization of EU decision-making – whether this parlamentarization results from a strengthening of the European Parliament or an empowering of national parliaments – comes along with an increasing importance of national parties. Second, the heads of state and government – most of them are also the leaders of their parties – are responsible for treaty amendments and their ministers for general EU policies. If these parties are empowered at both the EU and national level, and if they send their representatives to bargain, sign and ratify EU treaties and policies, their attitudes towards European integration need to be explained (cf. Garrett and Tsebelis, 1996: 270).

A major reason for this gap is that European integration has been largely considered as a public non-issue after the decline of the permissive (elite) consensus in the beginning of the 1990s (Reif, 1993). It neither offered political parties electoral incentives for positioning nor were European issues related to the basic conflicts that dominated national electoral competition (Hooghe and Marks, 2008: 6–7). For this period, the literature on national party competition also provided a mixed picture on the relevance of European integration with scholars pointing to no effect (Mair, 2000), while others find an emerging second dimension that cross-cuts the traditional left/right dimension in Europe (Marks and Steenbergen, 2002).

With the anew expansion of policy competencies and the accession of new countries, scholars found that public opinion is becoming more structured in the EU (Van der Eijk and Franklin, 2004), affects national elections (De Vries, 2007) and aligns to the national policy spaces (Hix, 1999). However, most of the public protests happened on the streets (Imig, 2004) and in referendums (Hobolt, 2009). Against this background, Marks and Wilson (2000: 433) argue that European integration is ‘assimilated into preexisting ideologies’. Party families, which traditionally span over the left/right-spectrum, are therefore supposed to absorb most of the variation among national parties (Hellström, 2008).

According to the more recent literature on European integration, national parties are playing an increasingly important role in the structuring of electoral competition over European issues (Lindberg et al., 2008). This is consistent with a perspective on party systems that focuses on societal challenges for the transformation of national policy spaces (Kriesi et al., 2008) and strategic considerations about party positioning (De Vries and Hobolt, 2012). However, the conventional view on parties’ positioning is that they form their attitudes towards European integration through historical cleavages (Marks and Wilson, 2000), socioeconomic classes (Kitschelt, 2013) and interest groups (Moravcsik, 1998).

Alternatively, we follow Prince and Overby (2005: 69–71) and propose to apply a party-dominant rationale to examine the implications of EU decision-making. From this perspective, we analyse the distributive and informational implications of EU decision-making for national parties’ attitudes towards European integration. These attitudes are currently receiving more attention with the rise of Euroscepticism (e.g. Topaloff, 2012). Compared to the focus of Euroscepticism research, however, our analysis addresses more generally the motivation of political parties to succeed in national party competition.

We argue that national parties play a key role in EU decision-making, in particular when they hold government offices and send their representatives to the Commission, Council and European Parliament. The closer a national party was located to EU outcomes, and the more it participated in EU decision-making, the higher are the distributional and informational advantages of this party. In summary, we derive the following two conjectures from theories on legislative organization (Gilligan and Krehbiel, 1987; Weingast and Marshall, 1988):

The first conjecture concerns the distributional effects of European integration. In general, political parties profit differently from the outcomes of EU decision-making. We therefore expect that closely located parties support European integration, while more distant parties tend to prefer less integration.

The second conjecture concerns informational effects. Government parties enjoy informational advantages from sending representatives to the Council and nominating Commissioners. In contrast, opposition parties can only participate in EU decision-making through the European Parliament. We expect that parties with governmental experience support European integration, while parties without office-holding experience prefer less integration.

We additionally consider the specific characteristics of EU legislation, which are admittedly complex. Compared to voters, however, we assume that political parties are more likely to understand the game of EU decision-making that offers various procedures and types of acts. In all procedures, the Commission has a monopoly to propose legislative acts (Crombez, 1996) and also decides about adopting tertiary legislation alone. Tertiary legislation consists of bureaucratic acts (Junge et al., 2015) that refer to secondary legislation, which is approved by the Council (and eventually the European Parliament). Because the number of bureaucratic acts is by far higher than the number of political acts under secondary legislation, the nomination of the Commissioners by the Member States is important.

Another characteristic of EU decision-making is the involvement of the European Parliament. Several authors argue that parliamentary involvement reduces the so-called ‘democratic deficit’ (Follesdal and Hix, 2006). Over time, the European Parliament has been empowered by strengthening the parliamentary veto right (Crombez and Vangerven, 2014) and expanding the application of the co-decision procedure. Compared to the Commission and the Council, non-governmental parties have direct access to EU decision-making through the European Parliament. They can possibly move outcomes towards their policy positions, especially when the Council does not share parliamentary interests.

Finally, legislative acts are directly binding or need transposition into national law in case of directives. Directives normally leave Member States with a certain amount of leeway to achieve the adopted goals. This decentral nature of directives offers more discretion, in particular for the governmental actors in charge that can also decide about starting the transposition process. König and Luig (2014) find that governmental office-holders even refrain from starting this transposition process when they fear disadvantages in national party competition.

To provide more support for our party-dominant rationale, we take a closer look at the implications of these specific types of EU legislative acts for the attitudes of national parties, which can generate information and influence outcomes when their representatives are key players in EU decision-making.

Data and measures: Positions, offices and activities

To examine our argument, we would ideally like to measure the implications of EU decision-making on partisan support for European integration directly. For measuring directly the distributional implication, we would need to estimate the outcome location of each of about 90,000 EU acts (in the period from 1979 to 2012) and the corresponding policy positions of each party. We would also need information about whether and how much each party ‘values’ the distance between its own policy position and the outcome as important. A similar hurdle exists for assessing directly the informational implication, because we would need to measure the informational level of each party before and after its involvement in EU decision-making.

Unfortunately, such direct and precise measures are currently out of reach, primarily because of the difficulty of estimating the distance and information level for each legislative act in a quantitative framework. We therefore employ an indirect approach that utilizes the EULIS data (König and Luig, 2012) on party positions, legislative saliencies and governmental office-holding to approximate the annual distances and information levels.¹

At this annual level of aggregation, we conduct a retrospective analysis of the implications of EU decision-making to emphasize the unidirectional causality of national parties’ learning from experiences. For each year, we cumulate the parties’ past experiences from EU decision-making, which we divide by the number of years. In other words, we expect that national parties learn over time the distributional and informational implications of EU decision-making. This procedure may also avoid leapfrogging of parties’ attitudes towards European integration – that is the extent to which political parties drastically shift their attitude relative to rival parties as a phenomenon that has raised controversy in the literature on estimating parties’ policy positions (Adams, 2001).

Estimates for parties’ attitudes towards European integration

We approximate the dependent variable on parties’ attitudes towards European integration by the EULIS estimate for the EU issue. This issue is derived from the Manifesto Project coding (MRG/CMP/MARPOR; Volkens et al., 2013) of the quasi sentences in the parties’ electoral declarations. Compared to other sources, such as roll call data on voting in the Council and European Parliament as well as expert survey data, a major advantage of the Manifesto Project is the coding of parties of all EU Member States since World War II. Starting with the first (direct) election to the European Parliament, our analysis covers the period between 1979 and 2012, in which several treaty amendments were ratified and 18 countries joined the EU at different points in time.

With the EULIS estimate for the EU issue we establish a simple measure of our dependent variable. It is based on the pair of Manifesto categories that account for positive and negative references to European integration. Because the original procedure of the Manifesto Project has been criticized for concentrating on salience, we estimate parties’ attitudes towards European integration by subtracting the logarithm of the negative references from the logarithm of the positive references.²

Classified by Member State, Table 1 shows the number of observations as well as summary statistics of the dependent variable.³ We observe the highest standard deviation of national parties’ attitudes towards European integration in crisis-ridden Greece. Within the ‘old’ EU15 countries, Belgian parties appear to be most enthusiastic about European integration, whereas parties in Denmark and Finland belong to the most Eurosceptic. Regarding the Eastern accession countries, the most anti-EU parties exist in Poland.

Table 1.

Partisan support for European integration by states.

State	Obs.	Mean	SD	Min.	Max.	Med.
Belgium	362	2.84	1.35	−0.32	5.53	3.04
France	183	1.39	1.66	−2.20	4.20	1.95
Germany	153	2.38	1.25	0.00	4.83	2.56
Italy	330	2.02	1.87	−2.56	5.94	1.95
Luxembourg	182	1.19	2.55	−3.66	5.07	1.67
Netherlands	224	2.13	1.89	−4.29	5.03	2.36
Denmark	290	0.01	1.72	−4.11	4.53	0.00
Ireland	185	0.82	1.80	−2.66	4.23	0.97
United Kingdom	184	1.36	1.68	−4.20	4.81	1.10
Greece	121	0.80	2.71	−4.47	4.72	1.10
Portugal	153	1.03	2.21	−4.08	5.62	0.43
Spain	268	2.60	2.10	−3.56	5.96	2.94
Austria	83	1.81	2.10	−2.04	5.37	2.23
Finland	149	0.50	1.63	−4.86	3.43	0.00
Sweden	128	0.92	2.51	−3.14	4.17	1.95
Cyprus	36	3.52	0.71	2.55	4.47	3.54
Czech Republic	43	2.13	1.61	−0.76	4.37	1.95
Estonia	52	2.91	0.68	1.95	4.14	2.64
Hungary	42	2.82	1.44	−1.01	4.99	2.94
Latvia	57	1.53	0.58	1.10	3.04	1.61
Lithuania	83	2.06	1.52	−0.90	5.44	1.95
Malta	18	1.34	2.28	−0.88	3.56	1.34
Poland	53	1.15	2.28	−3.22	4.78	1.10
Slovakia	59	3.13	1.57	−2.51	4.93	3.71
Slovenia	64	2.35	1.87	−1.95	4.65	2.71
Bulgaria	39	1.31	1.51	−1.61	3.22	1.10
Romania	29	2.34	0.92	1.10	4.34	2.71
Total	3570	1.68	2.04	−4.86	5.96	1.95

To illustrate the time trend of the dependent variable, the online appendix also presents national parties’ attitudes towards European integration by box plots for each observation year. Accordingly, the support for European integration declined in the beginning of the 1980s, that was characterized by economic stagnation and ‘Eurosclerosis’ (Thorn Commission). Under the three Delors Commissions (1985–1994) and the completion of the internal market followed a period of increasing euphoria, while disillusionment set in during the Santer/Marin Commission that was disgraced by corruption scandals and mismanagement. After a strong upward trend at the beginning of the new millennium (Prodi Commission) there was a period of stabilization under the Barroso I Commission. However, the sovereign debt crisis negatively affected partisan support for European integration in the last observation years (Barroso II Commission).

Estimates for distributional implications

For measuring the distributional implications, we follow previous studies and assume that the left/right scale is the most important dimension for national parties’ electoral competition (Crombez and Hix, 2015). As the legislative activities in EU politics widely differ over time, we calculate the absolute distances between the EULIS estimates of party policy positions and legislative outcome positions on the left/right scale. By weighting the issue prominence in the texts of EU legislative acts, EULIS aggregates data from 13 issue scales, which are derived from 32 of the 56 Manifesto Project categories (see the online appendix), into the more general profile of the left/right dimension.

EULIS applies a weighting procedure by a keyword search derived from the coding instructions of the Manifesto Project categories. This keyword search identifies how often each category is mentioned in the documents of EU legislative acts within each Commission term. Based on this aggregation, we estimate the legislative outcomes and their distances to party positions. Finally, we cumulate the outcome distances over the years and then divide them by the number of years to account for an incremental learning process of political parties from EU decision-making. Formulas, descriptive statistics and examples listed in the online appendix provide detailed information about the calculation of the parties’ left/right-distances to the outcomes in retrospect.

In the online appendix, we illustrate the distributional implications using histograms for each party family, which refers to the membership of national party delegations in the political groups of the European Parliament. Left and Green parties more often experience larger distances from the outcomes of EU decision-making, while we find that parties from the centre-right of the political spectrum are generally closer located to those outcomes. However, nationalist and special issue parties are also closely located to the legislative outcomes. This result may be due to the logarithmic transformation of their typically low number of programmatic sentences or the more general problem to locate these parties on the left/right dimension that comprises both economic and societal issues. As a consequence, we will control for party families in the multivariate analyses.

Estimates for informational implications

The informational implications of EU decision-making are derived from the office advantages, which national parties receive from their access to EU decision-making. These advantages are considered to shift parties’ centre of gravity from their vote-seeking responsiveness to their policy-seeking responsibility role (cf. Caramani, 2017). Except for the European Parliament, national parties particularly benefit from governmental office-holding through selecting and sending delegates to the Council that negotiates and adopts Commission proposals. Council working groups exchange governmental viewpoints on Commission proposals (together with the European Parliament) before they send their findings to the Committee of Permanent Representatives (COREPER). This committee searches for further solutions of open points from a cross-area perspective. At the ministerial level, Commission proposals are adopted, but ministers are in charge of the approval and the transposition of directives into national law.

Through the European Council, the governmental parties also nominate the Commissioners. This access is particularly exclusive in case of bureaucratic activities, such as tertiary acts, that allow the Commission to set legal standards autonomously. Depending on holding governmental offices, political parties may profit from their informational advantages in national party competition and are thus more likely to support European integration.

To account again for a retrospective evaluation as an incremental learning process of political parties from EU decision-making, we use data on the time each party spent in government, which we cumulate before dividing by the number of years. In the online appendix, we illustrate governmental experience by party family. In our period of study, Left, Nationalist, Regional/Ethnic and Special issue parties rarely participated in the formation of governments in the Member States. In contrast, Green Parties sometimes and Social Democratic, Liberal as well as Conservative parties are often members of (coalition) governments. Christian Democratic and (few observations of) Agrarian parties most often hold ministerial offices in national governments.

EU decision-making activities

To shed more light on the impact of the type of legislative act, we extracted information from the EUR-Lex database containing characteristics of more than 90,000 acts adopted in the period between 1979 and 2012. This allows us to distinguish between secondary and tertiary legislation, whereby the latter promotes a bureaucratization of EU decision-making. Our data also allows to identify whether the European Parliament is involved in EU legislation via the co-decision procedure. This may offer access to a broader spectrum of political parties. Finally, we can differentiate between directives, regulations, (ordinary) decisions and other acts to control for instrumental effects. The contrast between decentralized and centralized legislation (directives vs. regulations, decisions and other acts) is especially important because directives provide for discretion in their transposition into domestic law. Figure 1 presents all EU activities over time. The labelling of the time axis provides information about the start of Commission terms (assignment based on full-year ratio) and EU enlargement rounds.

Figure 1.

EU legislation 1979–2012.

According to Figure 1, there was a sharp increase of tertiary (bureaucratic) acts in the mid-1990s, which also increased the total number of EU activities. In spite of a decreasing trend afterwards, there remains a dominance of tertiary over secondary (political) legislation. Over time, the European Parliament gets increasingly involved in EU decision-making since the 1990s, while decentral decision-making by directives tends to stagnate. To account again for a retrospective evaluation of the implications of EU decision-making, we cumulate the respective shares over the years and then divide by the number of years. We also create interaction terms between the distributive and informational variables of EU decision-making on the one hand and the three legislative characteristics on the other.

Control variables

In addition to the features of EU decision-making, we control for several party- and country-specific characteristics. At the individual level, we firstly use the size in the national parliament. The effect of party size on the attitudes towards European integration may be positive because large parties often appoint (highly informed) heads of government or negative since large (government) parties are in general losing in the elections to the European Parliament (Hix and Marsh, 2007). Further party characteristics concern the salience that a party attributes to European integration (in relation to the left/right dimension), the divisiveness on the positional issue measured by the standard error of the EU references⁴ and the party family. We expect that a higher salience results in more support (in accordance with the salience theory) and that a higher divisiveness negatively affects parties’ attitudes towards European integration.

At the country level, we firstly include the public support for EU membership. This is measured by the difference between those appreciating country membership as a ‘good thing’ from those seeing it as a ‘bad thing’ in the Standard Eurobarometer surveys.⁵ To account for a cosmopolitan trend (Grande and Kriesi, 2015), we also control for broader participation and openness of a society by the respective index from the Worldbank’s Worldwide Governance Indicators Project.⁶ According to the utilitarian perspective, attitudes towards European integration may also be influenced by national benefits from the EU budget (Perkins and Neumayer, 2007) and the common market in an enlarging Europe (Gabel and Palmer, 1995). We therefore include the operating budgetary balance⁷ in percent of gross national income (GNI)⁸ and the value of intra-EU trade divided by the value of intra- and extra-EU trade.⁹

Briefly summarized, the share of seats in the national parliament ranges between 0.15% for the Ulster Unionist Party (2006–2010) and 63.58% for the Labour Party in the United Kingdom (1998–2001). Regarding the salience of the EU issue, there are some parties without any related quasi sentences, while the Liberal Party of Denmark stands out with a prominence of 44.66% (1991–1994). Similarly, there are some parties without divisiveness on the EU issue, while the Left Alliance in Finland appears to be extremely divided in 2000.

At the country level, the minimum value of public support comes from the United Kingdom in 1980, while the maximum value is reported for the Netherlands in 1991. All values above 70% are observed for Ireland, Italy, Luxembourg, Netherlands, Portugal and Spain. However, compared to the development of partisan support for European integration, public support reached its peak in the 1990s. Regarding broader participation and openness of a society, ‘civil society’ seems to be least developed in Romania in 2012 and strongest in Denmark in 2004. More generally, we find low values for Bulgaria, Romania and Greece, while Scandinavian countries and the Netherlands show the highest values.

The highest net contribution in percent of GNI was paid by Luxembourg in 1991 and the largest net recipient was Greece in 1983. Overall, Germany, the Netherlands, Luxembourg and Sweden provided considerable contributions to countries like Ireland (in early years), Lithuania and Greece. The minimum share of intra-EU exports is reported for Malta in 2012, while Luxembourg holds the maximum share in 2004. To sum up, the data of the control variables indicate high party- and country-specific variation over time (see the online appendix for further details).

Multilevel analyses

National parties declare their attitudes within each Member State to compete in national elections. We therefore acknowledge two levels of analysis: parties and countries. To account for this data structure, we apply multilevel statistics with a random intercept on each Member State. We also include a random slope on observation years at the country level and allow for (unstructured) covariance between both random effects. This requires multilevel mixed-effects linear regression models as follows:

y_{j} = X_{j} β + Z_{j} u_{j} + ε_{j}, with Σ = Var [u_{0 j} u_{1 j}] = [σ_{u 0}^{2} σ_{01} σ_{01} σ_{u 1}^{2}]

where y_i is the vector of responses (partisan support for European integration),

j = 1, \dots,, M

is the Member State level of grouping, X is the covariate matrix for the fixed effects β, Z_j is the covariate matrix for the random effects u_j (random intercept

u_{0 j}

and random slope

u_{1 j}

) and ε_j is the error term. Technically, we use the mixed command in Stata 14 to fit the models via maximum likelihood (StataCorp, 2013: 292–298).

Table 2 reports the results of three multilevel models. In Model 1, only the two main independent variables on the distributional and informational implications as well as party size, EU salience and divisiveness are specified as predictors. We successively include the (interaction terms with the) characteristics of EU legislative acts in Model 2 and all other control variables in Model 3. Overall, our multilevel modelling is corroborated in the sense that the random slope on years, the random intercept on Member States and the covariance between random slope and intercept (in Model 2 and Model 3) are statistically significant. Hence, the time effect varies across countries.

Table 2.

Results of multilevel analyses.

	Model 1		Model 2		Model 3
	Coeff.	SE	Coeff.	SE	Coeff.	SE
Distance to the outcome in retrospect	−0.646***	0.050	−0.989***	0.209	−1.508***	0.261
Government party in retrospect	1.110***	0.089	2.667***	0.382	1.206***	0.379
Share of secondary legislation in retrospect			0.073***	0.019	0.052***	0.018
Share of parliamentary involvement in retrospect			0.078	0.068	0.015	0.064
Share of directives in retrospect			−0.137	0.135	−0.461***	0.139
Distance to the outcome × Secondary			−0.024*	0.014	−0.023*	0.014
Distance to the outcome × Parliamentary			0.027	0.048	0.096**	0.046
Distance to the outcome × Directives			0.251**	0.108	0.443***	0.105
Government party × Secondary			−0.085***	0.025	−0.066***	0.024
Government party × Parliamentary			−0.379***	0.084	−0.270***	0.079
Government party × Directives			0.270	0.197	0.351*	0.186
EU party family: Green					0.129	0.143
EU party family: Left					−0.926***	0.144
EU party family: Social democratic					0.606***	0.132
EU party family: Liberal					0.639***	0.125
EU party family: Christian democratic					0.737***	0.130
EU party family: Conservative					−0.429***	0.160
EU party family: Nationalist					−1.427***	0.168
EU party family: Agrarian					1.428***	0.533
EU party family: Special issue					−2.282***	0.283
Public support for EU membership					−0.005*	0.003
Civil society					−0.003	0.272
Net recipient					0.033	0.036
Share of intra-EU exports					0.033***	0.005
Party size	0.027***	0.002	0.024***	0.002	0.016***	0.002
EU salience	0.057***	0.006	0.059***	0.006	0.050***	0.006
Divisiveness	−0.823***	0.107	−0.803***	0.107	−0.871***	0.101
Random slope on years at the Member State level	−3.154***	0.221	−3.077***	0.217	−3.083***	0.242
Random intercept on Member States	0.520***	0.179	0.383*	0.214	0.496**	0.220
Covariance between random slope and intercept	−0.380	0.257	−1.175***	0.302	−1.566***	0.288
Log-likelihood	−6835.683		−6796.195		−6519.169
Akaike information criterion	13689.365		13628.390		13100.338
Bayesian information criterion	13744.988		13739.636		13291.928

Note: Number of observations: 3570. Number of groups: 27.

p < 0.1, **p < 0.05, ***p < 0.01.

In all models, the distance to the outcome in retrospect negatively affects national parties’ attitudes towards European integration, while governmental experience increases their euphoria about the EU. This clearly supports our general arguments on the distributional and informational implications of EU decision-making. As shown in the online appendix, this result is robust to other model specifications, that is when we include fixed effects for years or Commission terms and when we exclude (the interaction with) the share of secondary legislation. At the party level, the effect of size is positive in all models. This corroborates the relevance of the informational implications of EU decision-making compared to the (potential) loss of some parliamentary seats. Also as expected, EU salience positively and divisiveness negatively affects the dependent variable in all models.

In Model 2, the distributional and informational effects of outcome distance and governmental experience gain in importance under control of and in interaction with the characteristics of EU legislative acts. The overall picture does not change when controlling for party families and country characteristics in Model 3. With regard to party families, our findings reveal the well-known inverted U-shape configuration. Radical parties are generally more against, while moderate parties are more in favour of European integration.

We also find a negative relationship between public support for EU membership and partisan support for European integration. This suggests that national parties sacrifice their responsiveness to the voters’ concerns for their responsibility on European integration. As expected, ‘civil society’ increases national parties’ support for European integration, but this effect is not significant when we include economic variables in Model 3. In contrast to the operating budgetary balance, the share of intra-EU exports positively affects parties’ attitudes towards European integration. The model with the highest explanatory power is Model 3. It is best performing according to the log-likelihood, Akaike’s information criterion (AIC) and the Bayesian information criterion (BIC) that penalizes the number of parameters more strongly.

For a substantive interpretation of the interaction terms in the best performing Model 3, Figure 2 illustrates the marginal effects. Following the recommendations of Brambor et al. (2006: 76) the solid line shows the marginal effect and the dashed lines represent the 95% confidence intervals. In addition, we overlay a histogram and a rug plot for the distributional and informational variables. On the left-hand side of Figure 2, the impact of legislative characteristics on partisan support for European integration is shown at different levels of outcome distance. On the right-hand side this impact is presented at different levels of governmental experience.

Figure 2.

The impact of legislative characteristics on partisan support for European integration at different levels of the main independent variables.

The marginal effect of the share of secondary legislation is presented in the top row (Figure 2(a) and (b)). Regarding the distributional and informational implications of EU decision-making, the marginal effect of secondary legislation is significantly positive when national parties are either closely located to the outcomes on the left/right dimension or lack governmental experience. However, these effects are of trivial magnitude, while the main effect of secondary legislation (as shown in Table 2) is both significant and positive. Accordingly, secondary legislation generally increases partisan support for European integration.

The marginal effect of the share of parliamentary involvement is illustrated in the middle row. According to Figure 2(c), the involvement of the European Parliament significantly increases the support of less centrally located parties, which hardly benefit from EU legislative outcomes. Once their distance is less than 0.8 to these outcomes, this effect disappears. There are two (non-exclusive) interpretations for this finding. First, the involvement of the European Parliament legitimizes EU outcomes, to which national parties have larger policy distances. Second, the European Parliament can – as a veto player in EU decision-making – shift outcomes closer to more distant parties. According to Figure 2(d), parliamentary involvement decreases the support of government parties. This effect is statistically significant and substantially meaningful for national parties with certain governmental experience, covering more than one third of observations.

Finally, the marginal effect of the share of directives is shown in the bottom row (Figure 2(e) and 2(f)). The more distant national parties are from the outcomes, the higher is the influence of (decentralizing) directives on parties’ support for European integration. This impact is statistically significant below 0.7 and above 1.5 – a substantially meaningful range of outcome distances comprising many observations. The effect of decentralization is negative for national parties with low governmental experience (two-thirds of observations). Obviously, more distant parties appreciate the decentralizing instrument of directives while non-governmental actors suffer from being excluded in the national transposition.

Conclusion

This study introduced an alternative perspective on national parties’ attitudes towards European integration by drawing the attention to the implications of EU decision-making. National parties are key players in (the transposition of) EU legislation. We therefore argued that their attitudes reflect experiences with distances to EU legislative outcomes and from government offices, which provide access to and knowledge about EU institutions. In support of our argument, we further investigated three major characteristics of EU acts, which include their political, parliamentary and decentral nature. Because national parties vary in their access to these types of acts, we considered interaction effects with outcome distances in retrospect and governmental experience.

Our empirical analysis comprised more than 90,000 EU acts adopted in the period from 1979 to 2012. We combined this information with the EULIS data on party positions, legislative saliencies and government participation. This allowed us to estimate our dependent variable as well as outcome distances and office-holding advantages for exploring the distributional and informational implications of EU decision-making. In addition to these main explanatory variables, we included the (interaction with the) share of secondary legislation, co-decision cases and directives into our analysis. We also controlled for several party characteristics, political and economic features of the Member States.

The results, which control for countries by multilevel modelling, provide empirical evidence for our hypotheses. The closer a party was located to legislative outcomes and the more governmental experience a party acquired, the more this party supports European integration. We also find that secondary legislation significantly promotes the support of national parties. This is somehow different for parliamentary acts. The co-decision procedure with the involvement of the European Parliament increases the support of national parties, which are more distant to EU outcomes. This may specify the role of the European Parliament for overcoming the criticism of the democratic deficit in the EU. However, parliamentary involvement decreases the support of parties with high governmental experience. With respect to the decentral nature of EU acts, directives increase the support of more distant (government) parties, which enjoy the discretion to transpose these acts into domestic law.

We also reveal that the key actors, in particular more centrally located parties with governmental experience, continue with their support for European integration while public support is already declining. With the ongoing transfer of policy competences the discrepancy between partisan and public support suggests that the resource provision through EU decision-making incentivizes parties to sacrifice electoral responsiveness for their responsibility role in European integration. This may also help to explain the increasing electoral success of Eurosceptic parties.

Our findings may contribute to the more general literature on European integration. The most prominent theories, such as liberal intergovernmentalism, supranationalism and constructivism claim that key actors pursue socio-economic, geopolitical or normative interests. In contrast to this concept, our party-dominated rationale provides an alternative view of European integration, according to which EU decision-making plays a central role for the support of key actors. Instead of governments, countries and supranational actors, this applies to the role of national parties, which compete with their rivals for information and material resources. This insight may have important ramifications for existing theories and the design of the institutional framework of the EU, which is currently confronted with criticism and scepticism.

Footnotes

Acknowledgements

Earlier versions of this article were presented at the University of Konstanz on 13 November 2014 and at Kiel University on 4 June 2015. We thank the participants for valuable comments. We also gratefully acknowledge three anonymous reviewers who provided very constructive and thoughtful suggestions to improve the manuscript.

Funding

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

Notes

References

Adams

(2001) A theory of spatial competition with biased voters: Party policies viewed temporally and comparatively. British Journal of Political Science 31(1): 121–158.

Brambor

Clark

Matt Golder

(2006) Understanding interaction models: Improving empirical analyses. Political Analysis 14(1): 63–82.

Caramani

(2017) Will vs. reason: The populist and technocratic forms of political representation and their critique to party government. American Political Science Review 111(1): 54–67.

Crombez

(1996) Legislative procedures in the European Community. British Journal of Political Science 26(2): 199–228.

Crombez

Hix

(2015) Legislative activity and gridlock in the European Union. British Journal of Political Science 45(3): 477–499.

Crombez

Vangerven

(2014) Procedural models of European Union politics: Contributions and suggestions for improvement. European Union Politics 15(2): 289–308.

De Vries

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

De Vries

Hobolt

(2012) When dimensions collide: The electoral success of issue entrepreneurs. European Union Politics 13(2): 246–268.

European Commission (2000) External and Intra-European Union Trade – Statistical Yearbook 1958–1999, Luxembourg: Office for Official Publications of the European Communities.

10.

Follesdal

Hix

(2006) Why there is a democratic deficit in the EU: A response to Majone and Moravcsik. Journal of Common Market Studies 44(3): 533–562.

11.

Gabel

Hix

Schneider

(2002) Who is afraid of cumulative research? Improving data on EU politics. European Union Politics 3(4): 481–500.

12.

Gabel

Palmer

(1995) Understanding variation in public support for European integration. European Journal of Political Research 27(1): 3–19.

13.

Garrett

Tsebelis

(1996) An institutional critique of intergovernmentalism. International Organization 50(2): 269–299.

14.

Gilligan

Krehbiel

(1987) Collective decisionmaking and standing committees: An informational rationale for restrictive amendment procedures. Journal of Law, Economics, & Organization 3(2): 287–335.

15.

Golder

(2016) Far right parties in Europe. Annual Review of Political Science 19(1): 477–497.

16.

Grande

Kriesi

(2015) The restructuring of political conflict in Europe and the politicization of European integration. In: Risse

(ed.) European Public Spheres: Politics is Back, Cambridge: Cambridge University Press, pp. 190–224.

17.

Groseclose

King

(2001) Committee theories reconsidered. In: Dodd

Oppenheimer

(eds) Congress Reconsidered, 7th ed. Washington, DC: CQ Press, pp. 191–216.

18.

Hellström

(2008) Partisan responses to Europe: The role of ideology for national political parties’ positions on European integration. Journal of European Public Policy 15(2): 189–207.

19.

Hix

(1999) Dimensions and alignments in European Union politics: Cognitive constraints and partisan responses. European Journal of Political Research 35(1): 69–106.

20.

Hix

Høyland

(2011) The Political System of the European Union, Basingstoke: Palgrave Macmillan.

21.

Hix

Marsh

(2007) Punishment or Protest? Understanding European Parliament Elections. Journal of Politics 69(2): 495–510.

22.

Hobolt

(2009) Europe in Question: Referendums on European Integration, Oxford: Oxford University Press.

23.

Hooghe

Marks

(2008) Postscript to ’making of a polity’. EUSA Review 21(2): 5–8.

24.

Imig

(2004) Contestation in the streets: European protest and the emerging Euro-polity. In: Marks

Steenbergen

(eds) European Integration and Political Conflict, Cambridge: Cambridge University Press, pp. 216–234.

25.

Ivarsflaten

(2008) What unites right-wing populists in Western Europe? Re-examining grievance mobilization models in seven successful cases. Comparative Political Studies 41(1): 3–23.

26.

Junge

König

Luig

(2015) Legislative gridlock and bureaucratic politics in the European Union. British Journal of Political Science 45(4): 777–797.

27.

Katz

Mair

(2009) The cartel party thesis: A restatement. Perspectives on Politics 7(4): 753–766.

28.

Kitschelt

(2013) Social class and radical right: Conceptualizing political preference formation and partisan choice. In: Rydgren

(ed.) Class Politics and the Radical Right, London: Routledge, pp. 224–251.

29.

König

Luig

(2012) Party ideology and legislative agendas: Estimating contextual policy positions for the study of EU decision-making. European Union Politics 13(4): 604–625.

30.

König

Luig

(2014) Ministerial gatekeeping and parliamentary involvement in the implementation process of EU directives. Public Choice 160(3): 501–519.

31.

Kriesi

Grande

Lachat

et al. (2008) West European Politics in the Age of Globalization, Cambridge: Cambridge University Press.

32.

Lindberg

Rasmussen

Warntjen

(2008) Party politics as usual? The role of political parties in EU legislative decision-making. Journal of European Public Policy 15(8): 1107–1126.

33.

Lowe

Benoit

Mikhaylov

et al. (2011) Scaling policy preferences from coded political texts. Legislative Studies Quarterly 36(1): 123–155.

34.

Lucassen

Lubbers

(2012) Who fears what? Explaining far-right-wing preference in Europe by distinguishing perceived cultural and economic ethnic threats. Comparative Political Studies 45(5): 547–574.

35.

Mair

(2000) The limited impact of Europe on national party systems. West European Politics 23(4): 27–51.

36.

Marks

Steenbergen

(2002) Understanding political contestation in the European Union. Comparative Political Studies 35(8): 879–892.

37.

Marks

Wilson

(2000) The past in the present: A cleavage theory of party response to European integration. British Journal of Political Science 30(3): 433–459.

38.

Moravcsik

(1998) The Choice for Europe: Social Purpose and State Power from Messina to Maastricht, Ithaca: Cornell University Press.

39.

Norris

(2005) Radical Right: Voters and Parties in the Electoral Market, Cambridge: Cambridge University Press.

40.

Perkins

Neumayer

(2007) Do membership benefits buy regulatory compliance? An empirical analysis of EU directives 1978–99. European Union Politics 8(2): 180–206.

41.

Prince

Overby

(2005) Legislative organization theory and committee preference outliers in state senates. State Politics & Policy Quarterly 5(1): 68–87.

42.

Reif

(1993) Ein Ende des ’permissive consensus’? Zum Wandel europapolitischer Einstellungen in der öffentlichen Meinung der EG-Mitgliedstaaten. In: Hrbek

(ed.) Der Vertrag von Maastricht in der wissenschaftlichen Kontroverse, Baden-Baden: Nomos, pp. 23–40.

43.

Rovny

(2013) Where do radical right parties stand? Position blurring in multidimensional competition. European Political Science Review 5(1): 1–26.

44.

StataCorp (2013) Stata Multilevel Mixed Effects – Reference Manual. Release 13, College Station: Stata Press.

45.

Tavits

(2012) Organizing for success: Party organizational strength and electoral performance in postcommunist Europe. Journal of Politics 74(1): 83–97.

46.

Topaloff

(2012) Political Parties and Euroscepticism, Basingstoke: Macmillan.

47.

Van der Eijk

Franklin

(2004) Potential for contestation on European matters at national elections in Europe. In: Marks

Steenbergen

(eds) European Integration and Political Conflict, Cambridge: Cambridge University Press, pp. 32–50.

48.

Volkens A, Bara J, Budge I, et al. (eds) (2013) Mapping Policy Preferences from Texts III: Statistical Solutions for Manifesto Analysts. Oxford: Oxford University Press.

49.

Weingast

Marshall

(1988) The industrial organization of Congress; or, why legislatures, like firms, are not organized as markets. Journal of Political Economy 96(1): 132–163.

50.

Williams

Spoon

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

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

0.01 MB

0.07 MB

0.11 MB

0.00 MB

0.03 MB

2.33 MB

1.16 MB

0.08 MB

0.01 MB