Abstract
This chapter examines the democratic credentials of the EU. Beginning with a discussion of the idea of democracy beyond the state, it explores academic debates about whether the EU suffers a ‘democratic deficit’. The chapter evaluates the EU along various dimensions, including how powers are separated and divided within the EU, the extent to which executive accountability is established, and the various mechanisms of representation in the EU. It explores the nature of European elections, the role of European political parties, the role of national parliaments in EU policy-making and recent innovations in the way that the president of the European Commission is chosen. The chapter concludes with a discussion of fundamental rights, values, and the rule of the law in the EU with a particular focus on recent developments in Hungary and Poland.
Summary
With so many decisions taken at EU level, what are the implications for democracy? All member states have to be democratic as a condition for membership. But when taking collective decisions through European institutions, the individual choices available to national democracies are naturally constrained. This chapter explores some of the literature on democracy in the EU. To what extent do democratic procedures at the European level compensate for this narrowing of choices at national level? Is there a ‘democratic deficit’? How can we understand democracy beyond the state? The chapter goes on to consider the importance of fundamental rights, values, and the rule of law in the EU and the extent to which developments in several member states—in particular, Hungary and Poland—represent a threat to the EU’s democratic credentials.
6.1p. 136 Introduction: Democracy beyond the State?
Economic, environmental, and other forms of interdependence mean that national authorities alone cannot adequately deal with a growing number of problems. Many require concerted international action at various levels. But traditional methods of international cooperation are slow, cumbersome, and frequently opaque. They involve negotiations among ministers (and, in practice, mainly officials) representing their state. In most cases, nothing can be agreed between states without consensus. Consensus can be a good thing in policy-making as it reduces tensions between political actors and lowers the risk of policy being forced upon reluctant states and citizens. However, consensus can lead to inertia and slower policy-making. It also creates a bias towards less ambitious outputs—policy solutions often become a function of what the most reluctant or anti-reforming state will accept (lowest common denominator agreements).
When an agreement is reached, it is (perhaps) submitted as a fait accompli to national parliaments on a take-it-or-leave-it basis. The quality of democracy on such issues is low and—by often failing to tackle policy problems because of the problem of the lowest common denominator phenomenon—this can raise broader questions of legitimacy and efficiency. Such are the working methods of the World Trade Organization (WTO), the International Monetary Fund (IMF), the World Bank, the UN (including on climate change), NATO, regular summits such as the Groups of 7 or 20 (G7 and G20), and countless other international organizations and intergovernmental fora.
The EU purports to be different. It is not (always) hostage to the lowest common denominator. The EU has an elected Parliament, directly representing its citizens and bringing into the process representatives of both governing and opposition parties in each country. Decisions on legislation are taken in public. It also has an independent executive, the European Commission, headed by commissioners who are politically accountable to the Parliament. It has a common Court to ensure uniform interpretation of what has been agreed. It has safeguards to ensure that it respects fundamental rights. And it has more developed mechanisms than any other international organization for informing, and sometimes involving, national parliaments. On the basis of its decision-making procedures, can we therefore say that ‘the EU not only forms a Union of sovereign democratic states, but also constitutes a democracy of its own’ (Hoeksma 2010)? Can democracy work at all on an international basis?
6.1.1 Demos, democracy, and the democratic deficit
Some argue that democracy can only work when there is a demos (Box 6.1), that is, a common feeling of belonging to the same community. A demos is usually based on speaking the same language, and having a shared past, as well as similar expectations about behaviour and common values. Others argue that this view of p. 137↵democracy is tribalist and point out that if speaking a common language is a requirement, then Switzerland, India, Canada, South Africa, and many others cannot be categorized as democratic. Some argue that the EU lacks a demos—although each member state could be said to have one—and that, absent a demos, there can be no democracy.
Box 6.1 Key concepts and terms
Accountability is a fundamental concept in representative democracy—elected politicians are accountable to citizens for their actions and decisions through periodic elections. In-between elections governments (executives) are accountable to parliament for the powers they use, the laws they pass, and their conduct in office. Courts and various agencies also play a role in holding governments to account, ensuring the actions of government are within the law.
Bicameralism is from Latin bi, two + camera, chamber. This is when a legislature comprises two chambers, usually chosen by different methods or electoral systems.
Charter of Fundamental Rights of the European Union: the EU has its own Charter of Rights that binds the EU institutions, and with which EU law must comply. Adopted initially as a political declaration by the EU institutions in 2000, it was given treaty status by the Treaty of Lisbon. It obliges EU institutions to respect the rights contained in the ECHR and others.
Democratic deficit is a term used in multiple ways. Its narrowest definition suggests that EU institutions and decision-making processes are insufficiently democratic. Broader definitions concern the relationship between an increasingly powerful and visible EU and what some see as its missing demos.
Demos (‘the people’) is a root word of democracy, the other being kratos (‘power’). In ancient Greece, demos referred to the ruling body of free citizens in Athens who felt a sense of loyalty and attachment to Athens as a political community.
European Convention on Human Rights (ECHR) is completely separate from the EU and its CJEU. It is an international treaty drafted in 1950 by the then newly formed Council of Europe (CoE). All (the now 47) CoE member states are party to the Convention. Any person who feels his or her rights, as defined in the Charter, have been violated by a state can appeal to the European Court of Human Rights. Judgments finding violations are binding on the state(s) concerned.
Others contend that thinking in binary terms about whether the EU has a demos or not overlooks the nature of the EU. Nicolaïdis (2003) developed the concept of demoi-cracy, ‘demoi’ being a plural and representing the idea that the EU is a ‘Union of multiple peoples’. The EU is seen as a polity that was created by states ‘to govern together, but not as one’ (Nicolaidis 2013: 351). The ethos of a demoi-cracy is rooted in two ideas—the non-domination and mutual recognition of the demoi. While this may sound abstract, many of the concrete ways in which the EU works in practice reflect demoi-cratic principles. For example, the preference for mutual recognition p. 138↵of standards rather than uniformity, the principle of subsidiarity (see section 5.2.1) in the EU treaties, and an institutional set-up that rejects majoritarianism in favour of cooperation, consensus, and shared/diffuse leadership.
Following the logic of demoi-cracy, the EU’s democratic legitimacy will be at risk when conflicts between member states lead to the breakdown of mutual recognition and the domination of some demoi by others. Conflicts between EU member states regarding how to respond to the Eurozone crisis or the refugee crisis are prominent examples of this threat. The attempt to forge greater political unity in response to a crisis—especially if that means some demoi imposing unity on others against their wishes—risks sacrificing the EU’s democratic legitimacy by eroding the mutual recognition between demoi (Bellamy 2013).
A third perspective suggests that, in fact, the EU does possess a demos. Eurobarometer data confirms that a majority of the EU’s citizens have developed a dual identity—their national identity co-exists with a sense of (admittedly weaker or secondary) European identity. Risse (2014: 1207) contends that this ‘Europeanization of national identities is sufficient to sustain carefully crafted (re)-distributive policies on the European level’. Writing shortly after the most heated phase of the Eurozone crisis, Risse saw the ‘politicization of national public spheres’ as an opportunity for—as opposed to a threat to—the EU. It was up to the EU’s leaders to ‘fight for their visions and try to persuade sceptical publics’, taking on the challenge of eurosceptic parties in order to build European identity and strengthen the EU’s nascent demos. Weiler (1995)—in a classic text on the p. 139↵subject—saw a European civic, value-driven demos existing side by side with a national organic-cultural one.
Debates around ‘demos’ and decision-making will continue, but most agree that, as some decisions are taken at European level, this should be done in as transparent, accountable, and democratic a way as possible. The loss of democratic accountability (Box 6.1) that occurs when national parliaments transfer their right to legislate to ministers meeting in the Council (of ministers) was one of the initial triggers of claims that the EU suffered a ‘democratic deficit’ (Box 6.1—see Weiler 1995; Zweifel 2002). The continual strengthening of the EP’s role in the EU legislative process was in large part designed to respond to charges of a ‘democratic deficit’.
Some of the earliest academic responses to the ‘democratic deficit’ thesis were to deny that it was in fact a problem. One early critic was Moravcsik (2002) who, from his intergovernmental perspective, argued that the EU was an organization of democratic states (see Box 6.2) and that it was therefore indirectly democratic. Talk of a European demos was, for Moravcsik, somewhat beside the point. Majone (1998) developed a second line of argument that also played down the idea of a ‘democratic deficit’, one grounded in the output of EU policy and its efficiency (see Crombez 2003). Majone argued that legitimacy (Box 6.3)—even in a democracy—can be attained through the effective performance of tasks and functions that raise efficiency, increase welfare, and that would be difficult (or even impossible) to accomplish without the relevant governance or institutional arrangements. In national p. 140↵democratic systems there are numerous powerful institutions that we accept as part of our governing arrangements—and that are not democratic in a narrowly defined way (i.e. they are not elected)—but rather possess a technocratic legitimacy.
Box 6.2 Spotlight on: Democracy, values, and the treaties
A commitment to liberal democracy is prominent in the treaties.
Preamble: the signatories confirm ‘their attachment to the principles of liberty, democracy and respect for human rights and fundamental freedoms and of the rule of law’;
Article 2 TEU: ‘The Union is founded on the values of respect for human dignity, freedom, democracy, equality, the rule of law and respect for human rights, including the rights of persons belonging to minorities’;
Article 7 TEU details the procedure for dealing with ‘a serious breach’ of the ‘values referred to in Article 2’ and empowers the European Council to ‘suspend certain of the rights deriving from the application of the Treaties to the Member State in question, including the voting rights of the representative of the government of that Member State in the Council’.
Article 10 TEU states that ‘the functioning of the Union shall be founded on representative democracy’.
The EU’s Charter of Fundamental Rights obliges EU institutions to respect the rights contained in the ECHR (see Box 6.1). Finally, the Copenhagen Criteria require candidate countries to meet various democratic standards including provisions about the rule of law, human rights, and respect for and protection of minorities (see Box 9.2).
Box 6.3 Spotlight on: Legitimacy
Legitimacy refers to the right to rule and make political decisions and the notion that ‘the existing political institutions are the most appropriate ones for society’ (Lipset 1963: 64). Legitimate political and governing systems are underpinned by shared moral principles and are accepted by citizens even when they disagree with, or dislike, specific outcomes or decisions. Legitimacy is important because the greater the legitimacy of a given system, the more resilient it is to short-term shifts of public opinion that might be driven by specific controversies or crises. Lord and Magnette (2004) identified four different ways that the EU’s institutions could be evaluated as legitimate:
Indirect legitimacy: the EU’s legitimacy derives from the legitimacy of the member states that created and control it—this resonates with an intergovernmental view of the EU;
Technocratic legitimacy: the EU’s legitimacy derives from its capacity to solve policy problems and improve the welfare of its citizens—this resonates with a neofunctionalist view of the EU;
Parliamentary (or representative) legitimacy: the EU’s legitimacy is rooted in an elected EP and elected member state governments in the Council working together to make law and policy;
Procedural legitimacy: the EU’s legitimacy is enhanced when it follows due process, works transparently, engages in consultation when making policy, acts proportionally, and protects mutually recognized rights and values.
The term democratic deficit began to take on a less precise meaning, often linked to the distance between EU institutions and voters and broader questions about the EU’s legitimacy. As EU institutions gained more powers, especially in areas that might be considered ‘core state powers’—and as the ‘permissive consensus’ gave way to a ‘constraining dissensus’ (Hooghe and Marks 2009)—Majone changed his view to one of concern about not merely a deficit but the potential for ‘democratic default’ (Majone 2014). This shift in opinion was driven by his belief that the EU’s competences had expanded to a point that was not appropriately matched by its democratic credentials, thus raising deeper legitimacy questions. The Eurozone crisis—and what Majone saw as the power of the EU to shape fundamental economic policies in controversial ways—contributed to his change of opinion (also see Scharpf 2015).
6.2 Separation of Powers and Executive Accountability
Most democracies operate on the basis of a separation of powers, although the actual separation of the legislative, executive, and judicial functions is not always clear. Indeed, many democracies point to ‘sharing’ of powers across institutions, with checks and balances, rather than a separation of powers (Neustadt 1991). In particular, the executive and the legislative functions have tended to merge in most European democratic systems (although they remain more distinct in presidential systems, notably in the Americas). Some European countries blur this distinction completely. The UK and Ireland, for instance, actually require executive members of the government to be simultaneously members of the legislature (even if this sometimes just means appointment to the UK House of Lords or the Irish Senate). By contrast, in France, a parliamentarian who becomes a minister must resign their seat for the duration of their ministerial mandate. Either way, it is the norm in most European countries for the executive to have a majority in Parliament, which, through the party system and other mechanisms, is usually compliant with the wishes of the executive. The separation between these two branches thus becomes less than clear. Only the judicial function remains clearly separate.
In the EU, however, the separation between the executive and the legislature is more distinct. Recall that the European Commission is the EU’s executive branch. Commissioners may not simultaneously be MEPs. In the EP, there is no compliant governing majority for the executive. Thus, in adopting any piece of legislation, a majority has to be built anew for each item through explanation, persuasion, and negotiation. The EP’s role is thus more proactive than that of most national parliaments in Europe.
The EP is the most ‘obviously’ legislative institution in the EU but it forms only half of a bicameral (Box 6.1) legislature along with the Council, a body that occupies p. 141↵a more complex position in the conventional separation of powers typology (see section 5.2.3). In addition to being a co-legislator with the EP, the Council is also empowered to act as an executive in specific cases. When acting on macro-economic policy or foreign affairs, it is fulfilling an executive rather than a legislative function, albeit one that consists largely of coordinating national executives rather than constituting a European one.
The European Council does not directly act as either an executive or as a legislature (indeed, it is precluded from exercising a formal legislative role by the treaty). Yet it influences both. Formally, the European Council is a strategic body, charged with defining ‘the general political directions and priorities’ of the Union. Informally, as an institutionalized forum of the most powerful political figures of the member states and the Union, it is often called upon to settle thorny political questions that could be executive, legislative, or constitutional in nature, with its president playing a representational role, in addition to the central task of establishing and ensuring consensus. The European Council also nominates or appoints a number of key posts in the Union. For these reasons, it could be considered as a sort of collective ‘head of state’ of the Union, in that its political role is similar to that of heads of state in national semi-presidential systems.
Tracing the lines of accountability is thus complicated in the EU. The EP is the most straightforward to appraise, as MEPs are accountable to their citizens through European elections. For the main executive body (the Commission), a system close to what is found in most national contexts—that is, accountability to a parliament—has been established. The Council is responsible, collectively, to no one—though most legislative decisions it takes require EP approval—but each of its members is a member of a national government and thus is accountable to a national parliament (the same can be said for the European Council). Given that the Council’s main ‘executive’ tasks are essentially about coordinating national policies, this member-state-based accountability may be considered appropriate. In any event, the Council must additionally justify itself in EP debates and answer parliamentary questions.
The CJEU is perhaps the institution that most easily slots into the conventional separation of powers framework. As the highest Court, the CJEU is composed of judges ‘chosen from persons whose independence is beyond doubt’ (Article 253 TFEU) who take an oath to ‘perform their duties impartially and conscientiously’. The deliberations of the Court are secret, so individual judges cannot be pressurized about judgments. We never know how any judge voted on any case unless they reveal how they voted in their memoirs. Where the EU is once again different, however, is that judges are appointed neither by the EU’s executive nor by its legislature, but rather are appointed for a (renewable) six-year term of office by the member states (one per member state).
This system differs from, for example, the US Supreme Court, whose justices are appointed for life by federal authorities (president and Senate), not by the states. The appointment of individual judges therefore depends more on the government of the member state from which they originate than on the composition of the EU p. 142↵executive or legislature. As a result, any political considerations in their appointment are rooted in domestic politics and are thus diffused across the 27 member states. The only common European element in the appointment procedure of the members of the ECJ was introduced in the Lisbon Treaty (Article 255 TFEU). Appointees are now scrutinized by a panel of six members chosen by the Council and one chosen by the EP. That seven-strong panel comprises former judges of the Court—or of national supreme courts—or eminent lawyers.
The Court has exercised an important independent function, ensuring that the EU’s institutions respect the law. It has both struck down acts of the Union’s political institutions and ruled against member states when they have failed to apply European law. Despite making controversial rulings, compliance with its rulings is high—a sign of the legitimacy of the institution (Saurugger and Terpan 2017). With the exception of a 2021 ruling by Poland’s highest court (EURACTIV, 2021b), no member state has explicitly refused to respect its judgments.
Overall, the Union system is characterized by a separation, or sharing, of powers, which is at least as distinct as is the case in most of its member states. The EU is sui generis and it is to be expected that conventional frameworks about separation of powers, which anyway vary among national political systems, will have distinctive features when applied to the EU. As important as the separation of powers is the balance of interests between a Union of 27 member states, comprising 27 political communities, and the extra scrutiny inherent in every proposal being examined by those 27 communities.
6.3 Elected Representation
Alongside separation of powers and accountability, elected representation is a core aspect of any system that purports to be democratic. This section considers several issues in turn. It first outlines the dual representation system at the heart of the EU. Then, it explores the nature of EP elections and European political parties. It goes on to consider efforts to increase the role played by national parliaments in the EU’s policy process before exploring efforts to link the appointment of the head of the executive (i.e. the president of the Commission) to the outcome of the EP elections.
6.3.1 Dual representation
The treaties identify the EU as a representative democracy (Article 10 TEU—see Box 6.2). When thinking about representation in the EU, it is important to note that it has two channels—what Lord and Magnette call ‘dual legitimation’ (2004: 185). The EU comprises both states and citizens (see section 1.1.1) and its institutional design incorporates both, the former in the Council and the latter in the EP. Such arrangements are not uncommon at the national level. Consider Germany (a federal state): the citizens of Germany directly elect the lower chamber of parliament (the Bundestag) and the upper chamber (the Bundesrat) is made up of delegations from the governments of the constituent states of the federation (the Länder).
p. 143↵The current system of QMV features a double majority based on one vote per state, alongside a second vote in which votes are weighted by population (see Box 3.3). Those votes must be cast as a bloc (they cannot be split), which is another feature that the Council shares with the German Bundesrat (see Table 6.1) Thus, the two traditional representational features found in bicameral federal systems—equality of states in one chamber and equality of citizens in the other—are both found in the Council.
Table 6.1 The Council and the German Bundesrat
Council |
Bundesrat |
|
---|---|---|
Composition |
Ministers from member state governments; preparatory meetings by permanent representatives (ambassadors) of member states in Brussels |
Ministers from state (Länder) governments; preparatory meetings by permanent representatives of states in Berlin |
Voting |
Each member state’s vote weighted by size and cast as a bloc |
Each state’s (Land) vote weighted by size and cast as a bloc |
Majorities required |
Usually, threshold higher than simple majority (qualified majority or unanimity) |
To disagree with Bundestag, threshold usually higher than simple majority (absolute majority) |
Reconciliation with elected chamber |
Conciliation Committee with EP |
Conciliation Committee with Bundestag |
Prior to 2014, QMV worked differently—a fixed number of votes were given to each member state, ranging from three for Malta to 29 for France, Germany, and Italy (and—then—the UK). It required roughly 74 per cent of the total votes allocated. However, successive EU enlargements, adding mostly small or-medium-sized member states, led to a situation where a qualified majority in the Council could represent only a minority of the EU’s population. Larger member states felt they were becoming underrepresented, which is primarily why the system was reformed. The shift reflects the power asymmetries between larger and smaller member states but also goes to the heart of democratic legitimacy in replacing a system that could, in theory, have entrenched minority rule over the majority.
In the EP, representation is degressively proportional, slightly overrepresenting smaller states. Between them, the Parliament and Council can affirm (or not) the acceptability of proposed EU policy from the point of view of both a majority of member states and the majority of the EU’s citizenry. The Parliament and Council perform these functions in the context of a consensual, rather than an adversarial, style of political system. High thresholds are needed, notably in the Council, to adopt any legislation, budget, or policy. Taken together, the EU system involves a greater number of representative channels than can be found anywhere else in the world above the level of the nation state.
p. 144↵As we saw in section 3.6, the EP does not merely rubber-stamp the executive’s legislation. Nor is it subservient to the executive via a party-political whipping (management) system, as is common in parliamentary democracies where the executive emerges from a majority coalition in the legislature. From its original role as merely a consultative body of seconded national parliamentarians—with final decisions taken behind closed doors in the Council—the EP has grown into the role of co-legislator (see section 5.2.3). It also seems to be more appreciated by the electorate. According to Eurobarometer opinion polling, out-dated perceptions of it as a ‘toothless tiger’ have faded. Around half of EU citizens said they ‘tend to trust’ the EP in late 2019 / early 2020, considerably higher than average reported levels of trust in national governments (around 35 per cent).
6.3.2 European elections, European parties
One of the earliest debates about representation through the EP was about whether EP elections were anything more than ‘second-order’ elections—that is, elections that were primarily a proxy for voters to express their feelings about domestic politics. Given that EP elections frequently take place mid-term in most national political cycles, they often result in a larger share of the seats going to opposition parties and smaller parties than would be the case in most national elections. EP elections thus perhaps can be dismissed as little more than a protest vote, but the results matter: they elect a wide variety of parties and thus have the effect of balancing the Council, whose members come exclusively from governing parties. The EP thus enhances pluralism and ensures that EU decisions are not left exclusively to ministers, diplomats, and bureaucrats.
Turnout is a good place to start when engaging with the debate about the nature of EP elections. Turnout in the 2019 EP elections was 51 per cent, up from around 43 per cent in the previous two elections (in 2009 and 2014), or about the same as for mid-term US Congressional elections—and higher than most local and regional elections in EU member states. From one perspective, it is normal that turnout is lower than for national parliamentary elections because less is at stake. However, turnout fell consistency in each EP election from a high of 62 per cent in the first elections held in 1979. The 2019 result broke the trend. Declining participation is a challenge for democracy at all levels, not peculiar to the European level, even if the latter does have distinctive challenges. The phenomenon has led many to consider additional participatory forms of democracy (see Box 6.4).
Box 6.4 Spotlight on: Participatory democracy in the EU
A more recent trend in democratic theory is the concept of participatory democracy (see Kohler-Koch 2012), meaning the creation of new mechanisms to facilitate the direct participation of citizens in democratic processes. The Lisbon Treaty added a new, direct source of policy proposals: one million EU citizens can now sign a European Citizens’ Initiative (ECI) to invite the Commission to bring forward a legislative proposal (Greenwood 2013; Boussaguet 2016). While the Commission is not legally obliged to act (with only six such initiatives out of over 70 launched having reached the one million threshold), the route has led to legislative proposals concerning drinking water quality. The jury is out on whether the ECI heralds a new dimension in EU democracy, or whether it will merely entrench the power of already prominent NGOs and civic society organizations that know how to mobilize publics and work the EU policy system—remember, policy is politics, and politics is power (see section 5.3). The EU’s most recent venture into the realm of participatory democracy is the Conference on the Future of Europe (see section 10.4.3).
p. 145↵Even on a low turnout, European elections do, at least, result in all the main strands of public opinion being represented in the EP. A functioning system of political parties is typically considered a necessary requirement in a representative parliamentary democracy. In the EU context, pan-European political parties have emerged as federations of national parties—they are thus looser groupings than national political parties (see Box 6.5).
Box 6.5 Spotlight on: European political parties
European political parties are cross-national and can reach beyond the EU. They are linked to but distinct from the political groups in the EP (see Table 3.1). Three existed before the first direct elections to the EP:
the PES, comprising parties from the democratic socialist/labour tradition;
the EPP, comprising Christian Democrats and other centre-right or conservative parties; and
the ALDE, previously the European Liberal, Democratic and Reformist Party (ELDR), comprising a variety of liberal and allied parties.
Two more emerged between 1979 and 2004 (and MEPs from these two parties currently sit in the same group in the Parliament):
the European Green Party;
the EFA, comprising regionalist and nationalist parties such as the Scottish, Flemish, Basque, Corsican, Sardinian, Catalan, and Welsh nationalists.
A number of other (usually smaller) European political parties were created following the adoption of a system for financing such parties in 2004. Some on the right or far right have been through several configurations, and regroupings, as alliances have shifted, and some folded. Currently, the other parties are:
the Party of the European Left (which includes a number of Communist or former Communist parties);
the ECR (notably including (pre-Brexit) the UK Conservatives);
ID (far right, eurosceptic);
EDP (centrist);
European Christian Political Movement (socially conservative)
p. 146↵The recognition and the development of these parties have been incremental. The Maastricht Treaty introduced an article referring to the importance of European political parties (Article 10 TEU). Later, parties themselves were granted legal personality and, crucially, access to funding provided certain conditions are met, such as being represented in a sufficient number of member states (at least one-quarter) and respecting the principles of the EU (such as liberty, human rights, and so on), although they do not have to support the existence of the EU itself. Parties also must publish their accounts and have them independently audited, as well as publish the names of any donors contributing more than €1,500. Parties may not accept donations of more than €18,000 from any single donor, nor may they accept anonymous donations. Money provided from the EU budget for European parties may also not be passed on to national parties.
The main parties are also active in a growing number of areas. They organize regular congresses, composed of delegates from national parties. Their leaders can hold ‘summits’ with leaders of their national parties (often prior to European Council meetings). Parties can also adopt common manifestos for European elections, though in practice their decision-making tends to be by consensus among the national member parties, which in turn means that the content of their policies tends towards the lowest common denominator. They are generally unknown to the public, except in the broadest sense that—for the Socialists, Liberals, Greens, and Christian Democrats—national voters may be aware that they are part of a larger European grouping and that they work together in the EP. Only a sophisticated minority of voters will actually be aware of the common manifestos on which they stand in European elections. However, even without such awareness, their activities can lead to a degree of convergence around common policy positions.
In the EP, the vast majority of MEPs organize in political groups that are affiliated to—but distinct from—European parties (see Table 3.1). This makes sense given that the policy choices at stake when dealing with legislation are typically political choices—for example, higher environmental standards at greater cost to those regulated, or not? On these subjects, there are nearly always different views within each member state, irrespective of the position taken by their minister in the Council. MEPs are more likely to come together cross-nationally because of a shared political or ideological outlook than they are to come together nationally in spite of differing political or ideological outlooks.
Group cohesion, as measured by roll call votes, reaches very high levels—between 85 and 96 per cent in 2020 for all groups except the eurosceptic ECR and ID groups (see Table 3.1)—far higher than the cohesion measured by nationality. Cohesion varies across policy areas depending on how divisive an issue is. For example, the Progressive Alliance of Socialists and Democrats (S&D) group exhibits greater cohesion on issues concerning migration than international trade (VoteWatch 2020). Interestingly, enlargement did not significantly impact group cohesion, in part because acting cohesively, as a group, is important if MEPs want to have an impact on the EU’s policy-making process. The EP lacks the strict whipping p. 147↵system found in national parliaments, but positions taken by the groups—and the negotiations between them—are what count in determining majorities. Although the principal cleavage in the EP is the classic left-wing/right-wing one, a secondary cleavage—between pro- and anti-integration MEPs—has grown in recent years (Otjes and van der Veer 2016). Anti-integration sentiment (euroscepticism—see Box 6.6) challenges the EU’s legitimacy because it represents (in its purest form, or hard euroscepticism) a questioning of the EU’s existence, or (in a diluted form, or soft euroscepticism) a desire to see the EU to do far less than it currently does.
Box 6.6 What drives euroscepticism?
The phrase euroscepticism is a tricky to one to define (see Box 4.2) and levels of euroscepticism vary across the 27 member states (see section 4.2.3). Research on what drives euroscepticism can be organized into three perspectives, each identifying a different key driver (see Hobolt and de Vries 2016). The first emphasizes self-interest. People support the EU when they feel that their country and/or they personally benefit from it materially. A second strand of research explores the importance of identity in driving euroscepticism. If people feel a strong sense of national identity, and perceive European integration to challenge or conflict with national identity, they are more likely to develop eurosceptic attitudes. A third perspective suggests that, because the EU is quite remote from citizens, attitudes towards it are often a proxy for broader attitudes about politics and the quality of government at the national level. Trust in, and support for, national government can thus be correlated with trust in, and support for, the EU. According to this research, people’s opinions of the EU are susceptible to being shaped by the national political actors—whether parties, leaders, or commentators—that they place the most trust in. Hooghe and Marks (2005) neatly capture the three perspectives with the heuristic calculation, community, and cues.
Despite the cohesion of the groups, national allegiances and political influences are not entirely absent. MEPs are elected in national or regional constituencies and are nominated as candidates by national political parties. Their actions are refracted through the lens of national media. There are moments—for example, in the run-up to national or European elections—when national cleavages are more likely to trump the ideological splits (Koop et al. 2018). This reminds us that, again, the EU is perpetually divided between the vertical divisions of national politics and the horizontal divisions of ideology, which represents an ongoing (creative) tension in the democratic fabric of the Union.
6.3.3 The role of national parliaments
Recent years have seen attempts to strengthen the role of national parliaments in scrutinizing the participation of their government in EU institutions. In certain cases, national parliaments may intervene directly, independently of their national p. 148↵government. The treaty lists a number of ways in which national parliaments ‘contribute actively to the good functioning of the Union’ (Article 12 TEU). One notable innovation in the Lisbon Treaty is a Protocol ‘on the Application of the Principles of Subsidiarity and Proportionality’. The Protocol contains a procedure whereby national parliaments, within eight weeks of receiving a legislative proposal, can send a reasoned opinion back to the EU institutions stating why they consider the draft proposal to not comply with the principle of subsidiarity.
If such reasoned opinions come from enough national parliaments (roughly one-third), then the Commission must review the draft, and subsequently justify its decision to maintain, amend, or withdraw its proposal. Employing a football analogy, this procedure is known as the yellow card. Alternatively, if such reasoned opinions come from more than half of the national parliaments in the EU, then a special vote must take place in the Council and in the EP—by a simple majority in the EP or by a majority of 55 per cent of the members of the Council—either of which can immediately kill off the proposal. This is known as the orange card procedure, as it was proposed by the Dutch (whose football team wears orange), and is not quite a red card.
These procedures are an important safeguard to prevent the overcentralization of powers, even if they are rarely needed: in the first ten years of operation of the procedures, only three proposals triggered a yellow card—a regulation about strike action (2012), the proposal to establish the European Public Prosecutor’s Office (2013), and a proposal to review a regulation governing workers who, for a limited period, work in an EU (or European Economic Area; EEA) state other than the one they normally work in (2016). Only in the first example did the Commission withdraw its proposal, and even then only because it was unlikely to gather the necessary political support to pass. None, to date, have triggered an orange card. Respect of subsidiarity appears not to be a frequent problem with Commission proposals.
Nevertheless, the very existence of these procedures means that more national parliaments are paying close attention to European legislation—and in practice to its substance—rather than just checking it against the benchmark of subsidiarity. More may start holding committee hearings of their country’s minister before Council meetings, as is already standard practice in the Nordic countries, or send comments on the substance of proposals directly to the Commission (there are currently some 200 such submissions per year). In addition, national parliaments increasingly confer among themselves, exchanging documents through an electronic exchange system (Interparliamentary Exchange; IPEX), as well as meeting together at committee level or in interparliamentary conferences (see Neuhold and Hogenauer 2016).
National parliaments are also involved in the process of future treaty change. Unless the EP decides otherwise, any IGC to revise the treaty must be preceded by a Convention composed of members of national parliaments, the EP, the Commission, and a representative of each government. And, of course, in most member states, national parliaments must ultimately ratify such treaty changes.
The treaty provisions involving national parliaments are thus quite numerous. In truth, most national parliaments have little time to actively scrutinize the p. 149↵fine-grained detail of EU issues (de Wilde and Raunio 2015). These new processes involve a deviation from their standard role (Sprungk 2013). Unlike the EP, most are in a classic government/opposition structure where governing majorities mean that there is, in practice, little scope to amend government texts or reverse their policies. National parliaments also have less time, expertise, and staff to devote to European matters than does the EP itself, which works full-time on EU affairs. National parliamentary procedures, practices, and timetables all diverge. Nonetheless, national parliaments can scrutinize and sometimes take part in EU decision-making to a degree that simply does not exist in other international organizations. In sum, the involvement in the adoption of legislation of both a dedicated EP and national parliaments means that the EU’s credentials measured against this particular yardstick of democracy are substantial.
Although some have suggested that national parliaments have collectively become a ‘virtual third chamber’ in EU policy-making (Cooper 2012), concerns have been raised. National parliaments are fundamentally national institutions and encouraging networks and interactions between political elites that cross national borders could, potentially, entrench interests and alliances that jar with the interests of national publics (Puntscher-Riekmann and Wydra 2013). The EU’s 27 national parliaments are also not equal in terms of their resource capacity and any process that seeks to bring national parliaments into the EU policy-making process risks exacerbating existing power asymmetries (Auel and Hoing 2014).
6.3.4 The rise (and fall?) of Spitzenkandidaten
The relationship between the outcome of European parliamentary elections and the composition of the executive is not as visible as it is in most European national parliamentary elections. However, the view that the Commission is ‘unelected’, unlike national governments, is overly simplistic. For example, British citizens do not directly elect the UK government: technically, the head of state (an unelected one at that) appoints a prime minister who appoints a government, which crucially relies on the confidence of the directly elected Members of Parliament in the House of Commons. Similarly, the Commission relies on the confidence of the EP; the EP itself has always had the right to dismiss the Commission. In the 1990s, the EP acquired the right to approve the appointment of the Commission, and indeed to elect its president. In fact, the EP took such votes even before such procedures were laid down formally in the treaties. The in-depth examination that candidate commissioners now receive from EP committees at hearings prior to their confirmation goes well beyond what ministers have to face in most European countries and for the last four Commissions has led to some candidates falling and being replaced (see Box 3.5).
Despite all this, few would consider the Commission to have an elected mandate. They might if the College of Commissioners were composed to reflect a majority coalition in the EP. Yet a Commission structured in this way is unlikely p. 150↵to emerge: most governments want to nominate as the commissioner from their country a member of their own political ‘family’. Again, this highlights the tensions in the EU—the Commission is supranational in essence but its College emerges from 27 national decision-making processes. The idea of the Commission being assembled like most national cabinets thus remains elusive—the link between the outcome of EP elections and the political balance of the EU’s ‘executive’ is distorted by the national appointment of commissioners, even though the Commission president can now veto nominations and decides on the portfolios that Commissioners get.
What has begun to change is that the vote on the president of the Commission is becoming more political and linked to the outcome of EP elections. This represents an attempt to establish a way of selecting the head of the executive that is linked to a democratic act (an EP election), rather than backroom bargaining among the leaders of the member states. That choice is increasingly important, as the president’s pre-eminence within the Commission has grown over time. The change to the treaties brought in by Lisbon refers to the ‘election’ of the president of the Commission by the EP. This vote is, as before, on a proposal of the European Council, but the latter must now take into account the results of the European elections in making its nomination (Article 17 TEU). This provision potentially makes the nomination similar to that of a head of state choosing a candidate prime minister who is capable of enjoying a parliamentary majority (see Box 6.7).
Box 6.7 Appointing prime ministers: There’s no single way to do it
Commission president: elected by the EP by an absolute majority on a proposal of the European Council, which must take account of the results of European parliamentary elections.
German chancellor: elected by the Bundestag by an absolute majority on a proposal of the Federal president.
UK prime minister: appointed by the Queen in light of advice as to who can secure a parliamentary majority in the lower chamber (House of Commons), but with no formal vote in parliament.
US president: chosen by an electoral college, whose members are elected in each state, normally as a function of which presidential candidate they support.
French prime minister: chosen by the directly elected president, without requiring a vote by parliament. However, the lower chamber (Assemblée) may dismiss the government by an absolute majority.
Swiss government: college of seven (with annual rotation of the president among them), elected by the two chambers of the parliament (and comprising members of all major parties).
Swedish prime minister: nominated by the speaker of the parliament and serves unless opposed by an absolute majority of (single chamber) parliament.
Italian prime minister: nominated by the president whose cabinet then requires approval by both chambers of the parliament (simple majority).
p. 151↵As a result, European political parties have started nominating their candidates for Commission president ahead of European elections. In 2014, the five main European political parties selected a Spitzenkandidat (a German word that means ‘lead candidate’) in an attempt to have their candidate named as Commission president. They argued that this could improve the transparency of the process and create a clear link between voting in EP elections and the selection of the Commission president, thus giving the post some form of democratic legitimacy. Televised debates were held between these candidates ahead of the elections (Dinan 2015). However, the process was controversial. The eurosceptic ECR Party refused to nominate a Spitzenkandidat, seeing the process as furthering and potentially legitimizing a more supranational—even quasi-federal—form of EU governance.
Despite heralding a form of pan-European political campaign (see Garcia and Priestley 2015), the process had ‘a limited impact on voter participation and voter choices’ (Hobolt 2014). Public attention generated by the process varied from country to country. Nonetheless, despite some misgivings from some of its members, the European Council proceeded to nominate the candidate of the largest party, Jean-Claude Juncker, who was duly elected by the EP (Christiansen 2016). The UK Prime Minister David Cameron strongly opposed this development and was joined by Hungary’s Prime Minister, Victor Orban, in voting against Juncker—the first case ever of a Commission president being nominated via a qualified majority vote in the European Council (Peterson 2017b).
In 2019, all the main parties again nominated candidates for Commission president ahead of the elections. However, ultimately the presidency was given to the (then) German Defence Minister, Ursula von der Leyen, who had expressed no interest in the position previously, and was not a candidate. A complex process of negotiation and bargaining led to this outcome (see Box 6.8). Some felt that this sounded the death-knell of the Spitzenkandidat system. But the largest political parties said they remain committed to it and will again put forward their candidates for Commission president ahead of the next elections in 2024. After all, in national contexts as well, it is not always a pre-announced party candidate who becomes prime minister: sometimes it is a new compromise figure (for example, Italy after the 2018 election), but it is the exception that proves the rule.
Box 6.8 How it really works: Picking a Commission president in 2019
There was good reason to predict that, following the 2019 EP elections, the Commission president would be appointed in the same manner as Juncker had been appointed (albeit not without controversy) in 2014. However, after the elections the European Council deadlocked on the decision. The EPP had won the most seats, but the PES had won more votes. The EPP candidate (Manfred Weber, leader of their Group in the EP) was seen as lacklustre and unable to secure a majority coalition. The PES candidate, Frans Timmermans, seemed more likely to get support from other parties who held the balance of seats. The most senior EPP head of government, Angela Merkel, at first accepted that it should be Timmermans, but a significant number of other EPP leaders did not follow her. Timmermans was also opposed by a number of governments from central Europe, unhappy about his criticisms of their policies on migration and on the rule of law.
After negotiations within the European Council that were explicitly on a party-to-party basis (the Croatian and Latvian prime ministers negotiated for the EPP, the Spanish and Portuguese prime ministers for the Socialists, and the Belgian and Dutch prime ministers for the Liberals), agreement was reached on a compromise candidate for Commission president, the German Defence Minister Ursula von der Leyen (as a German Christian Democrat this meant the presidency ultimately going to the EPP). The Socialists were compensated with Timmermans becoming First Vice-President of the Commission—able to choose his own portfolio—and Josep Borrell becoming the High Representative. The Liberal Charles Michel became President of the European Council.
p. 152↵The Spitzenkandidat process sheds considerable light on various aspects of the EU, in particular the tensions between supranationalism and intergovernmentalism. Juncker himself has called the process ‘not very transparent’. Many heads of state and government did not truly embrace the process in 2014—they saw it as an incursion onto their turf but were distracted by other things and found it hard to argue against a candidate as experienced as Juncker. The process revealed a lot about how the leaders of the EU’s various institutions seek to construct narratives and arguments that advance their position, and about different understandings of democratic legitimacy within the EU—should it come from the EU’s citizens voting in the EP elections, from the elected governments of the member states deciding in the Council and European Council, or from the interaction and tension between both? Eurobarometer (2021: 77) has found that a clear majority of EU citizens support the Spitzenkandidaten process.
6.4 Fundamental Rights, Values, and the Rule of Law
Democracy is frequently defined as rule by the majority. But in modern times, it is increasingly seen as going hand-in-hand with respect for minorities and for the rights of individuals. Governments and even elected parliaments can be challenged in the courts should they fail to respect fundamental rights. This feature can also be found in the EU. Initially it was exercised via case law: the CJEU acknowledged that the Union had to respect the fundamental rights that are common to the constitutional traditions of the member states. The Court recognized that all member states had signed the ECHR (of the CoE—see Box 6.1) and that it should be a source p. 153↵of law for its own deliberations. From 1993, the Maastricht Treaty entrenched this case law in the treaty itself.
With the Lisbon Treaty, the Union obtained its own Charter of Fundamental Rights (see Box 6.1), intended both to make those rights already contained in the ECHR visibly applicable to the Union but also to complement them with a range of other rights. The Charter was framed in such a way as to be binding in the field of EU law, and as such is binding on both the EU institutions, and the member states when applying European law. This means that decisions or acts of the Union can be struck down by the Court should they fail to respect the rights contained in the Charter. Furthermore, the Lisbon Treaty provided for the Union itself to accede to the ECHR. Although this has not yet occurred, it will give plaintiffs the right to appeal to the European Court of Human Rights should they fail to gain satisfaction from the CJEU, much in the same way as in member states an appeal can be made against the final judgment of a national court. In other words, the EU’s legal system will be subject to the same external yardstick as member states’ legal systems. Thus, in relation to the formal criterion of respecting fundamental rights, the EU system and procedures measure up well.
Questions about fundamental rights, values, and the rule of law have become increasingly prominent within the EU in light of developments in several member states, most notably Hungary and Poland (Holesch and Kyriazi 2020). Often called ‘democratic backsliding’—that is a gradual and deliberate process of de-democratization—it represents not just a policy challenge to the EU but also a potential existential crisis, calling into question the EU’s internal normative coherence and its standing in the world as a promoter of democratic values. Kelemen (2017) has referred to this development as the EU’s ‘other democratic deficit’. But it is not just an EU (or European) phenomena, and a wider literature explores the decline of democracy in various parts of the world (Diamond and Plattner 2015).
Minimalist definitions of democracy focus narrowly on elections and competing political elites, but broader definitions take in key features that have become associated with modern liberal democracy, such as the absence of corruption, the strength of civic society, an independent press/media, the existence of independent courts to uphold the rule of law, the protection of minority rights, and so forth (Diamond 2008; Luhrmann et al. 2017). Furthermore, great care is needed to preserve the informal rules of the game that help to maintain democracies—such as honouring the spirit as well as the letter of the law, exercising power in a restrained way, and recognizing legitimate political rivals. The gradual erosion of such rules can herald the beginning of the death of democracy (see Ziblatt and Levitsky 2018).
The governments of Hungary and Poland have not suspended elections or banned opposition parties—things that would be unambiguously anti-democratic—but stand accused of ‘enabling the erosion of key features that have become associated with modern liberal democracy’ (Gora and de Wilde 2020). In 2010, after winning a two-thirds ‘super’-majority of seats in the Hungarian parliament (on 53 per cent of the popular vote), Prime Minister Victor Orban and his Fidesz party set p. 154↵about making changes to the constitution and the laws that curtailed the independence of the judiciary, the media, various sectors of the economy, NGOs and universities, and the electoral process, as well as changing parliamentary procedures, making it a less effective scrutinizer of his government. In 2015, after winning power in Poland, the right-wing-populist Law and Justice Party (PiS) moved to annul appointments to Poland’s Constitutional Court and passed a law reducing the powers of that Court. In the years since, both governments have: filled the courts and media with pro-government appointees; actively tried to drive liberal NGOs, academics, and think tanks out of the country; and passed restrictive laws (in contravention of the EU Charter of Fundamental Rights) on abortion and gay and transgender rights.
As a result, Hungary and Poland have been characterized as ‘semi-authoritarian regimes’ (Dawson and Hanley 2016) and as ‘hybrid regimes’ that sit between democracy and autocracy (Krekó and Enyedi 2018), while Freedom House (2021) categorizes Hungary as only ‘partly-free’—in large part because of changes to electoral laws that give an ‘undue advantage’ to the governing party (OSCE 2014: 3). Orban has styled his approach ‘illiberal democracy’, a fusion of words that German Chancellor Angela Merkel said she ‘could not understand’. Juncker, when Commission President, greeted Orban at a 2015 summit in Riga with the words, ‘Here comes the dictator.’
The EU has struggled to respond effectively to these developments. The Commission initially tried to tackle the problem through the standard method of infringement procedures—the EU attempted to deal with Orban’s first media law as a failure to transpose a media directive appropriately (rather than as a rule of law issue), and with the Polish government’s moves against judges as a violation of EU employment law (rather than as a rule of law issue). Combined with the application of softer forms of social pressure—naming and shaming, appealing to a shared sense of values (see Sedelmeier 2017)—these measures hindered but did not prevent the ‘backsliding’ in either Hungary or Poland.
The treaty has a mechanism to deal with such situations (Article 7 TEU—see Sadurski 2010). The Commission, the EP, or one-third of member states can ask the Council to determine whether a member state is at risk of seriously breaching the EU’s values. The Council—acting by a four-fifths majority of its members and after obtaining the consent of the EP—is empowered to make that determination after hearing from the accused member state. However, determining the existence of a ‘serious and persistent breach’ of the EU’s values is the job of the European Council acting by unanimity (except the member state accused), on a proposal from either one-third of member states or from the Commission, and after obtaining the EP’s consent. If a member state is determined by the European Council to be in serious and persistent breach, the Council—acting by QMV—may decide to suspend certain rights of the member state, including voting rights in the Council.
The role of the European Council in this process—and the high thresholds involved—reflect both the ‘high politics’ nature of the issue and the desire by member states to guard themselves from such sanctions. The unanimity requirement p. 155↵means that a member state in breach of the EU’s values needs just one ally to shield it. The first step of the Article 7 process—asking the Council to determine whether a risk of a serious breach exists—has been triggered against both Poland (by the Commission in December 2017) and Hungary (by the EP in September 2018). In Poland’s case only 14 member states subsequently voted that Poland posed a ‘clear risk’ (four voted against and nine abstained).
In parallel to these developments, the Commission developed a Rule of Law Framework in 2014 to deal with ‘systemic threats to the rule of law’ in the EU but—lacking the power to impose sanctions—it is a toothless instrument. In March 2019, determined to find more effective leverage and mechanisms, the Commission won the EP’s support for linking EU budget payments to the rule of law in member states. Several member states had already voiced their support for such a linkage, but Hungary and Poland threatened to veto the EU’s 2021–7 Multi-annual Financial Framework (MFF) if it tied EU funding in such a way. An ‘interpretive declaration’ was agreed between the member states, referring the rule of law provisions in the MFF agreement to the CJEU before they can be applied. The attempt to sanction Hungary and Poland in this way is another example of the EU trying to deal with rule of law problems through instruments that are not explicitly designed for that purpose.
The EU’s attempts to contain ‘backsliding’ have thus been found wanting. It is a fine political calculation about how strongly to challenge it, especially given that challenging it too strongly could risk member states leaving the EU, weakening it economically, and potentially gravitating into other spheres of influence (for example, Russia’s or Turkey’s). Conversely, if the EU is a Union of values and a Union of (liberal) democracies, then the presence of illiberal democracies and ‘partly-free’ states within it risks undermining the EU’s identity and its normative power on the global stage (see section 8.5).
6.5 Conclusion
Questions about democracy in the EU are perennial. In a narrow sense they are evaluative questions about the democratic credentials of the EU’s institutions (Chapter 3) and policy-making process (Chapter 5). In a broader sense they are normative questions about the relationship between an increasingly powerful and visible EU and what some see as its missing demos. The increased political visibility of the EU and the ‘constraining dissensus’ make these questions important for the EU’s future. Numerous developments—from strengthening the EP to involving national parliaments more in the EU’s workings, from the right to petition the EP to the ECI, from hearings with candidate Commissioners to Spitzenkandidaten—have attempted to bolster the EU’s democratic legitimacy and its connection to its citizens. Challenged by developments in Hungary and Poland, the EU faces existential questions about how it ought to handle member states that do not uphold appropriate democratic standards.
p. 156↵We have seen that the EU system does fulfil fundamental democratic norms, but in a way that is more complex and less visible to the public than is the case at a national level. Like any political system, it has its own idiosyncrasies. Inevitably, its detailed functioning is different from what people are familiar with within their national system (which are themselves diverse). Those differences and complexities give rise to misunderstandings and can also be exploited by eurosceptic opponents. Nonetheless, the EU is unique in how far it goes to try to apply democratic principles at a level above the nation state. How successful it is remains open to debate. A study by Duina and Lenz (2017) concluded that the EU, at the very least, was one of the most democratically legitimate regional economic organizations. However, that simply reminds us that the EU is not merely a regional economic organization and questions remain about how to build new—and maintain existing—aspects of EU democracy.
Discussion Questions
How can democratic accountability be assured for those matters dealt with at a European level? Should it be via national parliaments scrutinizing their own government’s negotiating position, or via the EP, or both?
Does the relatively low turnout in EP elections matter?
What are the main challenges to citizen involvement in EU politics?
Is the Spitzenkandidaten system a good development?
Are fundamental rights sufficiently protected at European level?
Test your understanding of this chapter with additional self-test questions, with instant feedback.
Stay up to date on all developments in EU politics with biannual updates.
Further Reading
Weiler (1995) is a still classic exploration of demos in Europe, and Majone (2014) offers an interesting discussion of the ‘democratic deficit’. Siedentop (2002) explores whether representative government in possible in the EU. Cheneval et al. (2015) is a collection of papers exploring the concept of demoi-cracy in the EU. Habermas (2008) is a discussion of how decisions about Europe’s future can be put in the hands of its citizens. De Vries (2018) offers a comprehensive discussion of euroscepticism, and Schmidt (2020) considers whether the Eurozone crisis created a crisis of legitimacy. Kelemen (2020) discusses the recent authoritarian turn in some EU member states and Cianetti et al. (2018) consider ‘democratic backsliding’ in CEE, beyond Hungary and Poland. The European Economic and Social Committee (2021) has reported on civic dialogue and participatory democracy in the EU.
Web Links
The Platform for EU Interparliamentary Exchange (IPEX) facilitates the exchange of information between national parliaments and the EP concerning EU-related issues.
For data about voting in both the EP and the Council.
The website of the European Citizens’ Initiative.
The website of the EU Ombudsman.
The European political parties each have websites detailing their policies:
p. 158↵ European Greens.
Explore this topic further with additional web links to reliable content on EU politics.