Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, POLITICS (politics.oxfordre.com). (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy and Legal Notice).

date: 19 June 2018

Referendums in the European Union

Summary and Keywords

Referendums are frequently used to ratify European Union (EU)–related propositions. Since 1972 there have been in total 46 EU-related referendums, excluding third-country referendums on EU-related matters. While referendums are constitutionally mandated in some countries in order to ratify new treaties, other referendums are held for either normative or for political reasons.

Referendums deal with topics that are less familiar to voters, where key issues typically do not map onto domestic political cleavages. This means that we should expect that campaigns and the information they provide about the issues and the positions of political actors might matter more in framing issues than in first-order national elections. While there is by no means a scholarly consensus, recent research has shown, for instance, that an issue that dominates media coverage can impact how voters evaluate a proposition.

Finally, what do we know about voter behavior? While referendums on EU affairs have been criticized as being decided by “second-order” factors such as government popularity, there is evidence that when a proposition matters for voters, voting behavior is more dominated by issue-voting. Recent research has drawn on advances in cognitive psychology to investigate the impact of attitude strength and personality characteristics for voter behavior.

Keywords: European Union referendums, European Union, direct democracy, campaign effects, second-order elections, issue-voting, motivated reasoning, European Union politics

Different Types of EU Referendums

There are three main types of referendums: (1) on joining (or leaving) the European Union (EU), (2) ratifying new EU treaties or agreements, and (3) referendums on particular EU-related issues, like the Greek 2015 referendum on the bailout package. Since 1972, 46 referendums in total have been convened.1 A majority of referendums were convened by countries acceding to the EU, with the second major category being ratification of new treaties (e.g., the Maastricht Treaty).

As shown in Table 1, turnout has varied extensively across referendums. In situations where turnout was high, referendums can be seen as tools to alleviate the gap between EU-level governance and voters (de Vreese, 2007, pp. 5–7). In other referendums, turnout has been low in comparison to national parliamentary elections, suggesting that voters did not perceive the issue to be important or that they were even unaware of the referendum being held.

Table 1. EU Referendums

Year

Country

Subject

Required/Advisory

Percent yes-vote

Turnout

Ratified?

1972

Denmark

Accession

Required

63.3%

90.1%

Yes

1972

Ireland

Accession

Required

83.1%

70.9%

Yes

1972

Norway

Accession

Advisory

46.5%

79%

No

1972

France

Accession Treaty

Advisory

68.3%

60.3%

Yes

1975

United Kingdom

Accession Treaty (remain in)

Advisory

67.2%

64%

Yes

1986

Denmark

SEA

Advisory

56.2%

75.4%

Yes

1987

Ireland

SEA

Required

69.9%

44%

Yes

1989

Italy

Mandate for MEPs

Advisory

88.1%

85%

Yes

1992

Denmark

Maastricht Treaty I

Required

49.3%

83.1%

No

1992

Ireland

Maastricht Treaty

Required

68.7%

57%

Yes

1992

France

Maastricht Treaty

Advisory

51.1%

70%

Yes

1993

Denmark

Maastricht Treaty II

Advisory

56.7%

86.5%

Yes

1994

Austria

Accession

Required

66.6%

82%

Yes

1994

Sweden

Accession

Advisory

52.3%

70%

Yes

1994

Finland

Accession

Advisory

56.9%

83%

Yes

1994

Norway

Accession

Advisory

47.8%

89%

No

1998

Ireland

Amsterdam Treaty

Required

61.7%

56%

Yes

1998

Denmark

Amsterdam Treaty

Required

55.1%

76.2%

Yes

2000

Denmark

Euro opt-out

Required

46.8%

87.6%

No

2001

Ireland

Treaty of Nice I

Required

46.1%

35%

No

2002

Ireland

Treaty of Nice II

Required

62.9%

49%

Yes

2003

Sweden

Euro membership

Advisory

42%

83%

No

2003

Czech Republic

Accession

Required

77.3%

55%

Yes

2003

Estonia

Accession

Required

66.8%

64%

Yes

2003

Hungary

Accession

Required

83.7%

46%

Yes

2003

Latvia

Accession

Required

67.0%

73%

Yes

2003

Lithuania

Accession

Required

91.1%

63%

Yes

2003

Malta

Accession

Advisory

53.6%

91%

Yes

2003

Poland

Accession

Required

77.5%

59%

Yes

2003

Romania

Accession (change to constitution to join later)

Required

89.7%

56%

Yes

2003

Slovakia

Accession

Required

92.%

52%

Yes

2003

Slovenia

Accession

Required

89.6%

60%

Yes

2005

Spain

Constitutional Treaty

Advisory

76.7%

42%

Yes

2005

Luxembourg

Constitutional Treaty

Advisory

56.5%

89%

Yes

2005

France

Constitutional Treaty

Advisory

45.3%

69%

No

2005

Netherlands

Constitutional Treaty

Advisory

38.2%

63%

No

2008

Ireland

Treaty of Lisbon I

Required

46.6%

53%

No

2009

Ireland

Treaty of Lisbon II

Required

67.1%

59%

Yes

2012

Croatia

Accession

Required

66.7%

43%

Yes

2012

Ireland

Extra EU treaty (Fiscal Compact)

Required

60.3%

50%

Yes

2014

Denmark

European Patent Court

Required

62.5%

55.9%

Yes

2015

Denmark

JHA opt-out

Required

46.9%

72.0%

No

2015

Greece

Bailout terms

Advisory

38.7%

59%

No

2016

Netherlands

EU–Ukraine Association

Advisory

38.2%

32.3%

No

2016

Hungary

EU refugee quotas

Advisory

98%

40.4%

Rejected

2016

United Kingdom

Exit from the EU

Advisory

48.1%

72.2%

No

Note: The table does not include third-country referendums on EU-related matters.

Sources: European Parliament Research Service, 2016; Hobolt, 2009; UK, 2016.

Why EU Referendums Are Convened

The first scholarly debate relating to EU referendums is why they are convened. A key distinction here is between advisory or required referendums (Christin & Hug, 2002). Many referendums have been required, usually for constitutional reasons. For example, referendums are often required in Denmark when transferring sovereignty to the EU level. According to §20 of the Danish Constitution, measures transferring sovereignty have to be ratified either by the Parliament with a 5/6 majority or by voters in a binding referendum. In Ireland, in 1987 the Irish supreme court decided that a referendum was required before ratifying the Single European Act (SEA). After this, Irish governments have had to decide whether a new EU treaty required a referendum or not in relation to the constitution, resulting in every EU treaty since the SEA being the subject of a referendum.

Advisory referendums are convened by governments for two predominant reasons: either norm-related concerns relating to the need to ensure the democratic legitimacy of the country’s EU policies or for politically strategic reasons. Normative explanations relate to the ideational argument that actors attempt to follow “logics of appropriateness” in their decisions to convene referendums (Closa, 2007). These can be driven by normative pressures from either outside or inside one’s country. An external, norm-based dynamic can be seen in the wave of referendums held to ratify the Constitutional Treaty in 2005. Here the EU-level rhetoric about the EU’s “constitutional moment” and the need to legitimize the EU through more participatory forms of democracy led elites in several member states to decide to follow the example of other countries and hold referendums in order to be seen as doing “the right thing.” In the words of Closa, “Arguments and decisions taken elsewhere create a context in which domestic decisions . . . come to be seen as reasonable, logical and legitimate, easening thus the cost for governments to take such decision. Both the logic of justification and the imitation or mimesis generate a ‘rule of appropriateness’ about the proper way to ratify a Constitution” (2007, p. 1327).

There can also be domestic normative reasons for convening referendums—a form of normative path dependency (Closa, 2007). This ideational dynamic is best seen in the case of Denmark’s ratification of the Maastricht Treaty. After the first no vote in June 1992, the major yes parties (now including the Socialist People’s Party for parliamentary reasons) adopted the so-called “national compromise,” which mandated that the Danish government would ask for certain clarifications and exemptions from the common currency, defense cooperation, justice and home affairs, and citizenship, along with an attempt to push for more “openness” and subsidiarity. The Danish government took this deal to the Edinburgh Summit, where with the support of the U.K. presidency they received non-binding clarifications on these issues in the form of the “Edinburgh Agreement,” along with a promise of increased openness in the EC (Piris, 2006). However, when the new Social Democrat–led government—which took power in January 1993—had to decide how to ratify the Maastricht Treaty with the Edinburgh Agreement clarifications, they decided to send it to an advisory referendum despite enjoying the support of the 5/6 majority in parliament required for ratification. The predominant reason for this choice was the widespread expectation among Danish voters that they should have a say in ratifying the proposition with clarifications that a small majority had originally rejected. The Danes voted yes in the second, advisory referendum held in May 1993.

Other scholars have focused more on factors like partisan calculations and political strategy when explaining why referendums are convened (Bogdanor, 1994; Prosser, 2016). One explanation is that governments can use them as tactical weapons to strengthen their power, for example, in an attempt to create divisions in the opposition (Bogdanor, 1994; Prosser, 2016). This dynamic was seen in the 1992 French referendum on the Maastricht Treaty, which was convened by French President Mitterand partially as an attempt to undermine political opponents.

Governments can also try to avoid being punished for an unpopular policy (EU) in a coming election and decide to convene a referendum to shift the focus of EU unpopularity away from a forum where they risk losing seats (Prosser, 2016). Another, sometimes overlapping, strategic reason that governments decide to convene referendums is to use them to mediate a crisis within a party, allowing voters to decide an issue that splits party elites. This dynamic has been seen in the United Kingdom, where both membership referendums were to some extent attempts to resolve internal party divisions (1975 within the Labour Party, in 2016 within the Conservative Party). For example, then-U.K. Prime Minister Cameron’s promise to convene an advisory referendum on U.K. membership in the EU was an attempt to avoid dangerous splits among Euroskeptic and pro-EU Conservatives in the run-up to local elections in 2013. The promise of a referendum was included in the next Conservative party manifesto, which after the victory in the 2015 parliamentary elections led to the decision to actually convene a referendum.

Government can also decide to convene referendums in order to “tie their hands” in an attempt to strengthen their bargaining power in treaty negotiations based on a two-level game logic, at least in theory (Hug & König, 2002; Hug & Schulz, 2007). For the logic to work, voters have to be skeptical toward a new treaty, enabling the negotiator to make arguments like “we would love to accept what is on the table, but we need more to be able to sell the treaty to our skeptical voters at home.” Hug and Schulz (2007) find that governments that had scheduled a referendum on the Constitutional Treaty, and where voters were skeptical toward more integration, actually had more gains than their counterparts. However, there are clear limits to the applicability of this logic due to the risk of not being able to secure ratification even after securing “selling points,” and it is questionable whether countries like Denmark and Ireland can realistically utilize the tactic as a strategy for getting meaningful concessions when their bargaining power is limited due to their relatively small size.

Another strategic reason for governments to convene referendums is to use them strategically vis-à-vis the EU itself as a form of bargaining leverage in further negotiations in a policy area. In the 2015 Greek referendum on the third bailout package, the Syriza government in Greece convened a referendum on the terms of the agreement in an attempt to wrest more concessions from the Troika (EU Commission, European Central Bank, and the International Monetary Fund), with the Prime Minister Tsipras advocating a no vote (Triga & Manavopoulos, 2017). However, it can be argued that additional concessions from the Troika were unrealistic, meaning that the government would have to accept the terms irrespective of the outcome of the referendum. This suggests that the vote might have also been motivated by a desire by the government to be seen as protesting the deal before they were forced to buckle under and accept the terms.

In the Hungarian referendum in 2016, the government initiated a referendum on the EU’s mandatory refugee quotas, a measure that had been opposed by the Hungarian government (Pállinger, 2017). The referendum was a strategic tool to attempt to strengthen the government’s hand in further discussions of mandatory quotas in the EU, but the referendum did not succeed due to low turnout that were spurred by calls by the opposition to boycott the referendum in protest against the government’s strategy (Pállinger, 2017).

Campaign Effects

Once the decision to convene a referendum is taken, a political campaign starts, which can involve governmental ministries, political parties, political and economic interest groups, and ordinary citizens. EU referendums often deal with quite complex topics that do not typically map onto normal national political cleavages, meaning that this is a context in which we might expect that campaigns could matter more than they do in national election contexts (de Vreese, 2007; de Vreese & Semetko, 2004; Hobolt, 2006; Hobolt & Brouard, 2011; LeDuc, 2002).

A first critical question to ask before we proceed is: What exactly are “campaign effects”? Campaigns can be defined as all of the activities that provide information that: inform voters about the proposition under consideration, frame how issues are understood, prime particular aspects of the issue, and potentially persuade voters to change their underlying attitudes (Krosnick & Kinder, 1990; Lenz, 2009, 2013; Valentino, Hutchings, & White, 2002; Zaller, 1992). In democratic terms, a campaign should ideally enable voters to make a decision that corresponds with their underlying attitudes on the issue. But campaigns can also alter how voters perceive issues or even change their underlying attitudes.

The extent to which campaign effects matter can be illustrated using a continuum between situations where they have little effect (voter stability) or large effects (voter volatility) (de Vreese, 2007; LeDuc, 2002; see Figure 1). In situations where the issue is relatively unknown to voters and/or where partisan and ideological conflicts on the issue do not mirror more familiar patterns from regular national elections, voters lack signposts that can help guide them to forming opinions on the issue.

Referendums in the European UnionClick to view larger

Figure 1. The Importance (or Lack Thereof) of Campaigns

In this situation, campaign events and how issues are framed by media and political elites can be very important, shifting public opinion on the issue in the short term in ways that can affect the final outcome. Here voter attitudes are very volatile and malleable because they are not anchored in solid attitudes toward the EU that can be retrieved to help voters make an informed choice based on their underlying attitudes. In contrast, at the other extreme, in contexts where voters are familiar with EU-related issues, campaigns merely provide information that enable voters to decide based on their underlying, non-malleable issue attitudes, resulting in voter stability. In between are what LeDuc (2002) terms “opinion reversal campaigns,” where, for example, the frame of a well-known issue shifts during a campaign in ways that affect public opinion toward the issue.

There are two primary channels whereby the activities of political actors and the media can matter in referendum campaigns: (1) the provision of campaign material to voters about the proposition that provides information, frames the issue, and/or primes certain aspects of the issue and (2) through elite cues in the form of recommendations and other forms of endorsements.

First, campaigns have to provide information on the proposition in order for voters to be able to choose the option that matches their attitudes toward the issue (Bartels, 2005; Zaller, 1992). Unfortunately, despite the potential importance, there have only been a few studies of campaign effects, and these have concentrated on only a few cases (Denmark in referendums in 2000 and 2015, Netherlands and France in the 2005 referendums, Irish referendums). One of the reasons for this lack of research is that there are significant methodological and practical challenges related to studying campaign effects. Assessing them systematically requires expensive survey instruments that measure public attitudes repeatedly over the course of a campaign (e.g., either through panel designs or rolling cross-sectionals), coupled with the detailed analysis of media content during the campaign.

Most research on the importance of informational effects has looked at priming and framing through media coverage and political campaigning. For example, de Vreese and Semetko (2004) found in the 2000 Danish euro referendum that voter exposure to specific newspapers and public broadcasting had an impact on how they voted, although the magnitude of the effects found was not very large. Schuck and de Vreese (2008) found in the Dutch 2005 vote on the Constitutional Treaty that exposure to positive media coverage had an impact on voting behavior, making respondents more likely to support the proposition, other things equal.

Campaigns can also prime which aspects of an issue that voters find important. While campaigns probably cannot change the underlying attitudes of voters toward European integration, they can more plausibly impact short-term perceptions of the specific proposition that voters are being asked to decide on in a referendum. Hobolt and Brouard (2011), for example, found in the French referendum on the Constitutional Treaty that campaigns can prime certain dimensions of underlying attitudes of voters related to the EU, especially when the proposition is as complex and multidimensional as the Constitutional Treaty. In the French case, attitudes relating to the EU as a social threat became the most prominent dimension for voters in relation to deciding on the Constitutional Treaty, echoing the fact that economic concerns was a major topic of debate in the campaign (Hobolt & Brouard, 2011, p. 7). Garry (2013) found that campaigns can change voter perceptions of specific aspects of a treaty when comparing the two Irish referendums on the Treaty of Lisbon (the first resulted in a no, the second a yes). In the second referendum, fewer voters were concerned about the implications of the treaty for Irish neutrality and the potential loss of an Irish Commissioner.

Information provided by campaigns can also frame issues in terms of how voters perceive the benefits of the proposition in relation to what happens in the event of a no vote (the reversion point). In the Danish referendum on adopting the Euro in 2000, the yes side focused their arguments on the economic benefits of joining the Euro. However, when this argument was discredited by many experts, the yes side struggled to formulate a convincing argument for why Denmark should join the Euro (de Vreese & Semetko, 2004). We saw a similar dynamic in the 2015 Danish referendum on replacing the Justice and Home Affairs (JHA) opt-out with an opt-in model (Beach, 2017). The core yes-side argument was that the opt-in model had to be adopted in order for Denmark to remain in Europol after the legal base became community-based. However, when the no side and experts argued that a form of Europol membership was also possible in the event of a no vote because Denmark could negotiate an intergovernmental “parallel agreement” that would allow participation, this left the yes side without clear arguments in favor of removing the opt-out.

Second, elite cues can provide voters with heuristic shortcuts that can enable them to make an informed choice “as if” they had all of the relevant knowledge that would enable them to choose the option that best matches their underlying attitudes. The importance of elite cues relates to the fact that most voters do not possess deep “expert” understanding of the issues involved in an EU referendum, nor do they have motivations to undertake an exhaustive information search in order to update their information. Instead, voters tend to rely on low-cost cognitive shortcuts and different heuristics in order to make sense of the issues in referendums (Bowler & Donovan, 1998; Downs, 1957; Hobolt, 2006; Lupia & McCubbins, 1998).

In research on Californian referendums, Bowler and Donovan (1998) found that different types of voters use different types of information to calculate whether they should vote yes or no based on their underlying interests. Highly educated voters often use what can be termed more “objective” information available during a campaign, but even less educated voters can make reasonable decisions based upon information from TV advertising, editorials, and conversations with friends and colleagues, enabling them to vote “as if” they were fully informed (Bowler & Donovan, 1998; Lau & Redlawsk, 2006; Lupia, 1992, 1994). One method that less informed voters can utilize to gauge the veracity of statements put forward in a campaign about the potential impact of a measure from different sources is to look at “who is behind it” (Lau & Redlawsk, 2006; Lupia, 1994). Applied to EU referendums, Hobolt (2006) has argued that voters can infer their own position based on whether it is endorsed by a group they like or dislike. She finds in the 1994 Norwegian accession referendum that when voters have knowledge of party endorsements on a proposition, they are just as competent in choosing an outcome that maps onto their own underlying attitudes toward the EU as voters with more detailed knowledge of the proposition.

However, a reliance on cues and endorsements can also lead voters to make a decision that reflects the position of the party they normally vote for instead of their own issue attitudes (Hobolt, 2006; Kriesi, 2005). In the Danish Maastricht referendums in 1992 and 1993, Hobolt (2006) finds that reliance on party endorsements also means that parties can frame the meaning of the choice voters face—either by making the proposition appear to be more attractive or by making the reversion point in the event of a no vote less attractive. When comparing the two referendums, she finds evidence that yes parties succeeded in the second vote in 1993 in framing the proposition as more beneficial and the implications of a second no vote as more negative, making supporters of yes parties more prone to support the proposition, other things being equal in comparison to the first referendum.

For party endorsements to matter, parties also have to be united in their messaging. In EU referendums, parties often find it difficult to “formulate a clear campaign message and mobilize what is often an internally divided party to stay ‘on message’ during the campaign” (de Vreese, 2007, p. 10). When a party sends mixed messages, voters who identify with the party are unable to use endorsements as a heuristic to figure out how they should vote in a referendum.

There are significant avenues still to be explored relating to how different types of voters respond to campaigns. The general literature on voter behavior has long identified heterogeneity, understood as a situation where voters with different characteristics respond differently to campaigns. Zaller (1992) told us that reasoning processes differ depending on whether voters are politically sophisticated or not, but there has been relatively little investigation of the effects levels of political sophistication on how referendum campaigns impact different types of voters (although see later for a few examples).

Recent research in political psychology also suggests that we might expect that citizens with strongly held attitudes would tend to engage in more motivated reasoning, where voters selectively recruit and evaluate evidence so that it matches their prior beliefs (Epley & Gilovich, 2016; Holbrook, Berent, Krosnick, Visser, & Boninger, 2005; Taber, Cann, & Kucsova, 2009; Visser, Bizer, & Krosnick, 2006). In an EU referendum context, this would imply that voters with strongly held attitudes: (1) require less information overall to make a decision that maps onto their underlying issue attitude (e.g., Druckman, 2012; Holbrook et al., 2005; Houston & Fazio, 1989; Schuck & de Vreese, 2008); (2) are less susceptible to cues and endorsements than less motivated voters (Hobolt, 2006); and (3) overall are less susceptible to persuasion and arguments because they give greater weight to information that matches their prior beliefs when they evaluate incoming information from a campaign.

Based on evidence from studies of EU skepticism and public opinion toward the EU, we should expect that anti-EU attitudes are affect-based, meaning that voters would hold them more strongly (Boomgaarden, Schuck, Elenbaas, & de Vreese, 2011). By contrast, pro-EU attitudes are typically utilitaristic and therefore are likely to be less strongly held. Pro-EU voters therefore are in theory less motivated to deploy the cognitive resources required to “protect” their existing beliefs, meaning that we might expect that they are more predisposed to utilize second-order behavior where they either stay home (see next section) or utilize party recommendations and/or level of trust/satisfaction with the proposer of the proposition (i.e., the government). Given the lower salience of the EU issues in general for these pro-EU voters, we should expect stronger campaign effects within this group. At present there is some evidence from the 2015 Danish opt-out referendum (Beach, 2017) that suggests that Euroskeptical voters were able to better utilize information provided by the campaign to figure out whether voting no corresponded to their underlying issue attitudes. For example, on election day the percentage of undecided and, therefore, non-mobilized pro-EU voters was three times higher compared to the group of undecided Euroskeptical voters. Although the same information was available to every voter, there were larger shifts in vote intention and learning for voters with Euroskeptical attitudes throughout the campaign. This provides some evidence for asymmetric campaign effects due to motivated reasoning, where a group of voters with strong, affect-based attitudes needed less information to decide how to vote.

Voter Behavior—Do Voters Actually Answer the Question They Are Asked?

What factors determine voter choice in EU-related referendums? The core debate in the literature is whether citizens actually decide in relation to their underlying attitudes toward the EU and/or the proposition itself (issue-voting) or whether they decide based on other considerations like governmental popularity because they do not believe EU-related propositions are important enough to expend the cognitive resources required to actually understand the issue and make an informed decision (second-order). This section explores the debate between second-order and issue-voting theories and the evidence for both, illustrating the developing consensus about the contextual conditions under which issue-voting dominates. However, even when issue-voting dominates in high-salience referendums, there is also evidence that voters hold a status-quo bias, and in instances where a vote is close, voter dispositions to keep what they know instead of opting for more unsure gains can tip the balance toward preserving the status quo. Finally, more recent research has opened up for the role of emotional dynamics might play alongside more “rational,” material considerations in relation to issue-voting.

Many scholars and commentators claim that EU referendums are perceived by ordinary citizens as abstract and relatively unimportant affairs, meaning that we should not expect voters to expend the cognitive resources required to evaluate the proposition and the reversion point in relation to their underlying attitudes towards European integration. Instead, voters are expected to treat an EU referendum as a “second-order” election, where they express their level of satisfaction with the performance of the incumbent government by voting yes or no (Franklin, 2002; Franklin, Marsh, & Wlezien, 1994; Ivaldi, 2006; Reif & Schmitt, 1980). A popular government will be able to steer a proposition it has proposed to ratification because a majority of voters will trust that the government will only endorse it if it is good for the country, and vice versa (Franklin et al., 1994, p. 102). If this second-order dynamic holds, this implies that referendums are not useful tools for democratic legitimization of EU affairs because they are decided on non-EU–related matters.

While there is some disagreement about the exact causes and mechanisms of the second-order election thesis, at the core is the argument that when voters feel little is at stake, they either abstain from voting (resulting in lower turnout), or if they decide to vote, they merely utilize heuristics drawn from first-order, national affairs such as attitudes towards government performance or attitudes to national political issues (Hix & Marsh, 2011; Hobolt & Brouard, 2011; Marsh & Mikhaylov, 2010). Franklin, van der Eijk, and Marsh (1995) claim that as public opinion is ill-informed about issues in referendums, in a parliamentary regime nothing will be more salient to voters than the standing of government, meaning that any proposal by a government will be colored by this fact. Franklin et al. (1995) claim that partisan attachments are the primary factor in determining outcomes because voters will decide merely to follow the recommendations of the party that they otherwise support, but at the same time party preferences will typically track with governmental popularity because voters who supported a government in the last election, but who are unhappy with its current performance, might indicate in a survey during the referendum that they would vote for an opposition party. Irrespective of the specific mechanisms, the results of the second-order dynamics are the same: when going to the polls in European referendums, voters focus on how they feel about national politics rather than how they feel about European integration.

In contrast, other scholars claim that voter behavior in referendums on EU matters is similar to normal elections, where voters decide based upon their attitudes towards the issue itself, thereby engaging in issue-voting (Hobolt, 2006, 2009; Merrill & Grofman, 1999; Svensson, 1994, 2002). Other things being equal, we can expect that when a referendum is required, voters will perceive that it is more important because the outcome is binding (Hug, 2002).

Early versions of the issue-voting model in regard to EU referendums argued that voter choice was based upon voters’ general attitudes toward European integration (Svensson, 1994, 2002), whereas more recent formulations focus more explicitly on voter attitudes toward the specific question they are posed (e.g., Garry, 2013; Hobolt, 2006, 2009). While it is usually not made explicit in issue-voting models, it is assumed in many studies that these issue-related attitudes reflect voter evaluations of their utility gains in relation to EU integration based upon the socio-economic predispositions of the individual voter, in that those groups stand to gain the most from integration support it and vice versa, other things equal (Gabel, 1998; Hooghe & Marks, 2005). However, other research has suggested that voter attitudes toward the EU can also be driven by identity concerns, with voters who fear immigration and are strongly attached to their national identity having more negative attitudes toward the EU (Hobolt & de Vries, 2016; Hooghe & Marks, 2005; McLaren, 2006). (We refer readers with further interest in the drivers of EU attitudes to consult Hobolt & de Vries, 2016.)

Hobolt (2006, 2009) has proposed the most ambitious issue-voting model regarding EU referendums. Hobolt draws upon a rational choice proximity model, where the median voter decides based upon the relative location of their ideal point (IP) toward EU integration in relation to their perceptions of the placement of the proposition and the reversion point (RP) in the event of a no vote.2 When the proposition is closer to the median voter’s IP than the RP, the voter will vote yes, and vice versa.3

While engaging in issue-voting, voters might also be relatively risk-averse, preferring the certainty of what they know to the uncertainty of change. Therefore, there is often a status quo bias in voter behavior in referendums. This argument can be found as far back as 1932, where Schumacher, in a case study of Oregon initiatives, attested that “the elector, when in doubt, is inclined to vote no” and there was a tendency for voters “to continue the status quo” (1932, p. 251). More recently, other scholars have also argued that referendums have a potential conservative bias toward the status quo. For example, Bowler and Donovan (1998) propose that the electorate in referendums use a no vote as an anchoring reference point. Lupia (1994) has demonstrated that voters often have more information about the status quo than the change that will be initiated by the proposal at the ballot. The logic is that they know what they have, and changes to the status quo involve greater uncertainty (risk). Hobolt (2009) has included this factor in her model as a propensity for voters to prefer the status quo except in circumstances where it is very clear for voters that the reversion point in the event of a no vote is very harmful.

Several authors have suggested that the context of the referendum might matter for which dynamic is dominant. Franklin (2002) suggests that second-order dynamics might be particularly dominant in low-salience referendums, whereas when a question has high political salience we should expect voting based upon the issue (Franklin, 2002; see also Szczerbiak & Taggart, 2004a, 2004b). Hug (2002) has claimed that second-order dynamics are more likely in referendums that have been convened for strategic reasons by governments and when the outcome is non-binding on the government. The non-binding nature of the referendum enables voters to more easily use the referendum to send a signal to unpopular political elites, other things equal.

What evidence do we have that issue-voting or second-order factors dominate voter decision-making in EU referendums? As referendums across Europe have become politically contentious, a scholarly consensus has begun to develop that suggests that issue-voting is increasingly the prevalent dynamic in EU referendums (Garry, Marsh, & Sinnott, 2005; Hobolt, 2009; Schuck & de Vreese, 2008). Simply put, there is not a lot of evidence that suggests that unpopular governments proposing referendums are less successful than their more popular counterparts.

Schuck and de Vreese (2008) found that in the 2005 Dutch no vote to the Constitutional Treaty, the strongest predictor for how voters decided was their general EU skepticism, which was a factor through which political attitudes and predispositions (unrelated to the specific issues) exerted influence. They did find that voters who had little faith in the political elites that supported the referendum tended to have more skeptical attitudes toward the EU and that voters that were strongly attached to Dutch national identity were also more skeptical toward the EU. Additionally, voters who feared immigration and globalization were also more skeptical. They conclude that EU skepticism is a mediator of influence of other, non-EU–related political attitudes and predispositions, although they did find some second-order effects also (Schuck & de Vreese, 2008, p. 117).

There is also strong evidence for issue-voting in Danish referendums (Hobolt, 2006, 2007, 2009), Irish referendums (Garry, 2013; Garry et al., 2005), and in the wave of accession referendums in 2003 (Szczerbiak & Taggart, 2004b). Comparing the two Maastricht referendums in Denmark in 1992 and 1993, Hobolt (2006) finds that issue-voting dominated in both but that partisan cues did affect voter assessments of the benefits of the proposition and the location of the reservation point in the event of a no vote. Voters with identical EU attitudes shifted their evaluation of the proposition due to information from parties that claimed that (1) the Maastricht + Edinburgh agreement package sent for ratification in the second referendum was better for Denmark than Maastricht on its own (i.e., closer to the median voters ideal point) and (2) the consequences of voting no a second time would be Danish exit from the EU. The difference between the first no and the second yes vote on the same treaty can therefore be explained by a shift in voter perceptions of the consequences of a yes and no vote.

Garry (2013) found that issue-voting dominated in the two Irish Lisbon Treaty referendums and that voter choice was dominated by attitudes toward the specific proposition and not just underlying and general EU attitudes. And in a journal special issue on the accession referendum in 2003, the editors concluded based on the case studies that issue-voting dominated across the cases but that elite cues did matter depending on how knowledgeable voters were of EU affairs and how credible elites themselves were (Szczerbiak & Taggart, 2004a, 2004b).

Recent work in the field of issue-voting has begun to investigate whether there is heterogeneity in voter behavior, with some types of voters behaving differently from others. Drawing on Zaller (1992), Hobolt (2009) found that politically sophisticated/knowledgeable voters were more likely to have substantive opinions on the issues at stake and thereby were more likely to engage in issue-voting. Garry (2013) did not, however, find evidence of this in the two Irish Lisbon referendums. Another form of heterogeneity was detected in his analysis. Garry (2013) assessed whether economic concerns are more salient for voters with low “human capital.” The logic is that voters who are potentially exposed to competition created by the EU’s free movement of workers would be more skeptical toward more integration in a referendum. There was some empirical support for the thesis.

Another form of heterogeneity has also been assessed by Garry (2014). Building on research from political psychology on the role of emotions—in particular work that has found that voters who are “anxious” are more likely to engage in deliberation in contrast to “angry” voters, who are more likely to rely on partisan attitudes (e.g., Petersen, 2010). Angry voters are more likely to make choices based on second-order factors, whereas anxious voters will tend to be risk-averse and seek out more information in an attempt to make a choice based on their attitudes toward the issue. Garry (2014) provides empirical evidence supporting this hypothesis about emotions. In line with risk aversion, anxious voters were more likely to support the Fiscal Compact than angry voters. Of those who felt angry during the referendum campaign, “a strong and highly statistically significant relationship emerges between dissatisfaction with the government and voting No” (Garry, 2014, pp. 246–247). The final result was a yes vote because there were more anxious voters than angry voters.

Recent EU Referendums—Anything New Under the Sun?

There have been three no votes in EU referendums in 2015 and 2016, including the dramatic British no vote in June 2016 to leave the EU. Are we witnessing new voter dynamics? Is voter decision-making more influenced by a broad lack of confidence in political elites and institutions among large groups of voters, resulting in a new form of second-order dynamics? The following will discuss only the British and Danish no votes, as the extremely low salience of the Dutch 2016 referendum on the EU’s association agreement with Ukraine makes it difficult to draw any conclusions about voter decision-making, with only 32.3% of voters turning up.

That a majority of British voters opted to leave the EU was a shock for many observers, although polls in the run-up suggested that it might be close. Many observers expected that the well-known status quo bias would kick in when it was a 50/50 situation, pushing enough undecided voters arriving at the polls to tip the vote toward a remain result. However, a small majority voted to leave the EU, with a turnout of 72.2%. Overall, Hobolt (2016) found that issue-voting dominated voter decision-making, which was not surprising given the high salience of the issue. Hobolt (2016) found that the remain arguments about the economic costs of leaving the EU resonated with pro-EU voters, whereas the leave side’s anti-immigration arguments were effective in mobilizing anti-EU voters. Voters who “benefited” from increased internationalization (better educated, younger, and more well-off voters) tended to vote remain, whereas the “losers of globalization” tended to vote leave. This suggests that voter decision-making was driven by some of the same concerns that have led to increased support for right-wing populist parties throughout Europe, both in national elections and in the EP election of 2014.

Steenbergen and Siczek (2017) also found support for issue-voting, but they also investigated whether voter levels of risk propensity affected how they voted. The theory was that more risk-averse voters would prefer to keep what they know and vote Remain, whereas voters who responded in surveys that they were willing to take risks also were more likely to vote for Brexit. Their findings suggest that the emerging political psychology literature, which looks at the impact that different underlying personality traits can have, is also relevant in understanding voter behavior in EU referendums.

Related to this, there is some evidence from recent studies of the importance of attitude strength and motivated reasoning from Danish referendum in 2015. In this referendum, the proposition was to replace the Danish Justice and Home Affairs (JHA) opt-out with a British-style opt-in protocol. However, a majority of Danes voted no despite the referendum being relatively high salience (turnout of 72%). Issue-voting dominated decision-making by voters, but there was also a difference in how voters with strong attitudes behaved. The theory of motivated reasoning suggests that voters with strongly-held attitudes will selectively recruit and evaluate information in a manner that gives greater credence to information that matches their pre-existing beliefs. Based on survey data (see Beach, 2017), there is evidence to suggest that voters with strongly held attitudes needed less information to make a decision, and were more / less responsive to arguments that were consistent / inconsistent with their underlying attitudes. This suggests that providing more information in a campaign will not necessarily shift voters, but instead enables them to figure out how they should vote based on their pre-existing EU attitudes.

Conclusions

Research on EU referendums can be split into three distinct questions: (1) Why are referendums convened?; (2) Do campaigns matter?; and (3) What factors determine voter decision-making?

Governments convene referendums for a number of reasons. There are situations where they are constitutionally mandated, but in other situations they are voluntary. Here research has suggested both norm-based and strategic reasons for convening referendums. There can be situations where convening a referendum can be seen as the “right thing to do,” best seen in the wave of referendums to ratify the Constitutional Treaty in 2005. But governments also use referendums strategically—for example, the decision by then-British PM Cameron to convene a referendum on EU membership to reduce pressure on the Conservative party from the UK Independence Party (UKIP).

Campaigns do matter, but the preponderance of evidence suggests that the effects deal mainly with providing information about the proposition that can enable voters to determine whether they should support it or not based on their underlying attitudes toward the EU. One key channel for this information is through partisan endorsements that provide voters with a useful heuristic to decide “as if” they had full information. Campaigns can also matter by priming particular aspects of the issue.

When making decisions, there is a strong body of evidence that indicates that in salient referendums, voter decision-making is dominated by issue-voting. There is some disagreement about whether issue-voting is driven by general EU attitudes or more proposition-specific attitudes, and there is debate about what drives EU attitudes (economic/materialist or identity-based concerns). However, there is still significant work to be done in incorporating insights from political psychology into the study of voter decision-making in EU referendums.

References

Atikcan, E. Ö. (2015). The puzzle of double referendums in the European Union., Journal of Common Market Studies, 53(5), 937–956.Find this resource:

Bartels, L. M. (2005). Homer gets a tax cut: Inequality and public policy in the American mind. Perspectives on Politics, 3(1), 15–31.Find this resource:

Beach, D. (2017). Denmark. In F. Mendez & M. Mendez (Eds.), Referendums on EU matters (pp. 174–187). Brussels: European Parliament Policy Department for Citizens’ Rights and Constitutional Affairs.Find this resource:

Berinsky, A. J., & Lewis, J. B. (2007). An Estimate of Risk Aversion in the U.S. Electorate. Quarterly Journal of Political Science, 2(2), 139–154.Find this resource:

Bogdanor, V. (1994). Western Europe. In D. Butler & A. Ranney (Eds.), Referendums around the world. The growing use of direct democracy (pp. 25–97). London: Macmillan.Find this resource:

Boomgaarden, H., Schuck, A. R. T., Elenbaas, M., & de Vreese, C. H. (2011). Mapping EU attitudes: Conceptual and empirical dimensions of Euroscepticism and EU support. European Union Politics, 12(2), 241–266.Find this resource:

Bowler, S., & Donovan, T. (1998). Demanding Choices: Opinion, Voting, and Direct Democracy. Ann Arbor: University of Michigan Press.Find this resource:

Bowler, S., & Donovan, T. (2002). Do voters have a cue? Television advertisements as a source of information in citizen-initiated referendum campaigns. European Journal of Political Research, 41, 777–793.Find this resource:

Buch, R., & Hansen, K. M. (2002). The Danes and Europe: From EC 1972 to Euro 2000—elections, referendums and attitudes. Scandinavian Political Studies, 25(1), 1–26.Find this resource:

Christin, T., & Hug, S. (2002). Referendums and citizen support for European integration. Comparative Political Studies, 35(5), 586–617.Find this resource:

Closa, C. (2007). Why convene referendums? Explaining choices in EU constitutional politics. Journal of European Public Policy, 14(8), 1311–1332.Find this resource:

Downs, A. (1957). An Economic Theory of Democracy. New York: Addison-Wesley Publishing.Find this resource:

Druckman, J. N. (2012). The politics of motivation. Critical Review, 24, 199–216.Find this resource:

Epley, N., & Gilovich, T. (2016). The mechanics of motivated reasoning. Journal of Economic Perspectives, 30(3), 133–140.Find this resource:

European Parliament Research Service. (2016, May). Referendums on EU issues. Research briefing. Available online.

Franklin, M. N. (2002). Learning from the Danish case: A comment on Palle Svensson’s critique on the Franklin thesis. European Journal of Political Research, 41(6), 751–757.Find this resource:

Franklin, M. N., van der Eijk, C., & Marsh, M. (1995). Referendum outcomes and trust in government: Public support for Europe in the wake of Maastricht. West European Politics, 18(3), 101–117.Find this resource:

Franklin, M. N., Marsh, M., & Wlezien, C. (1994). Attitude towards Europe and referendum votes: A response to Siune and Svensson. Electoral Studies, 13(2), 117–121.Find this resource:

Gabel, M. J. (1998). Economic integration and mass politics: Market liberalization and public attitudes in the European Union. American Journal of Political Science, 42(3), 936–953.Find this resource:

Garry, J. (2013). Direct democracy and regional integration: Citizens’ perceptions of treaty implications and the Irish reversal on Lisbon. European Journal of Political Research, 52(1), 94–118.Find this resource:

Garry, J. (2014). Emotions and voting in EU referendums. European Union Politics, 15(2), 235–254.Find this resource:

Garry, J., Marsh, M., & Sinnott, R. (2005). Second-order versus “issue-voting” effects in EU referendums: Evidence from the Irish Nice treaty referendums. European Union Politics, 6(2), 201–221.Find this resource:

Hix, S., & Marsh, M. (2011). Second-order effects plus pan-European political swings: An analysis of European Parliament elections across time. Electoral Studies, 30, 4–15.Find this resource:

Hobolt, S. B. (2006). How parties affect vote choice in European integration referendums. Party Politics, 12(5), 623–647.Find this resource:

Hobolt, S. B. (2007). Taking cues on Europe? Voter competence and party endorsements in referendums on European integration. European Journal of Political Research, 46(2), 151–182.Find this resource:

Hobolt, S. B. (2009). Europe in question: Referendums on European integration. Oxford: Oxford University Press.Find this resource:

Hobolt, S. B. (2016). The Brexit vote: a divided nation, a divided continent. Journal of European Public Policy, 23(9), 1259–1277.Find this resource:

Hobolt, S. B., & Brouard, S. (2011). Contesting the European Union? Why the Dutch and French rejected the European Constitution. Political Research Quarterly, 64(2), 309–322.Find this resource:

Hobolt, S. B. & de Vries, C. E. (2016). Public Support for European Integration. Annual Review of Political Science, 19(1), 413–432.Find this resource:

Holbrook, A. L., Berent, M. K., Krosnick, J. A., Visser, P. S., & Boninger, D. S. (2005). Attitude importance and the accumulation of attitude-relevant knowledge in memory. Journal of Personality and Social Psychology, 88(5), 749–769.Find this resource:

Hooghe, L., & Marks, G. (2005). Calculation, community and cues: Public opinion on European integration. European Union Politics, 6(4), 419–444.Find this resource:

Houston, D. A., & Fazio, R. H. (1989). Biased processing as a function of attitude accessibility: Making objective judgments subjectively. Social Cognition, 7(Spring), 51–66.Find this resource:

Hug, S. (2002). Voices of Europe: Citizens, referendums, and European integration. Lanham, MD: Rowman & Littlefield.Find this resource:

Hug, S., & König, T. (2002). In View of Ratification: Governmental Preferences and Domestic Constraints at the Amsterdam Intergovernmental Conference. International Organization, 56(2), 447–476.Find this resource:

Hug, S., & Schulz, T. (2007). Referendums in the EU’s constitution building process. Review of International Organizations, 2, 177–218.Find this resource:

Ivaldi, G. (2006). Beyond France’s 2005 referendum on the European Constitutional Treaty: Second-order model, anti-establishment attitudes and the end of the alternative European utopia. West European Politics, 29(1), 47–69.Find this resource:

Kriesi, H. P. (2005). Direct Democratic Choice. Lanham: Lexington Books.Find this resource:

Krosnick, J. A., & Kinder, D. R. (1990). Altering the foundations of support for the president through priming. American Political Science Review, 84(2), 497–512.Find this resource:

Lau, R. R. & Redlawsk, D. P. (2006). How Voters Decide: Information Processing in Election Campaigns. Cambridge, U.K.: Cambridge University Press.Find this resource:

LeDuc, L. (2002). Opinion change and voting behavior in referendums. European Journal of Political Research, 41, 711–732.Find this resource:

Leeper, T. J., & Slothuus, R. (2014). Political parties, motivated reasoning, and public opinion formation. Political Psychology, 35(S1), 129–156.Find this resource:

Lenz, G. S. (2009). Learning and opinion change, not priming: Reconsidering the priming hypothesis. American Journal of Political Science, 53(4), 821–837.Find this resource:

Lenz, G. S. (2013). Follow the leader?: How voters respond to politicians’ policies and performance. Chicago: University of Chicago Press.Find this resource:

Lupia, A. (1992). Busy voters, agenda control, and the power of information. American Political Science Review, 86(2), 390–403.Find this resource:

Lupia, A. (1994). Shortcuts versus encyclopedias: Information and voting behavior in Californian insurance reform elections. American Political Science Review, 88(1), 63–76.Find this resource:

Lupia, A., & Matsusaka, J. G. (2004). Direct democracy: New approaches to old questions. Annual Review of Political Science, 7, 463–482.Find this resource:

Lupia, A., & McCubbins, M. D. (1998). The democratic dilemma. Can citizens learn what they need to know? Cambridge, U.K.: Cambridge University Press.Find this resource:

Marsh, M. (2007). Referendum campaigns: Changing what people think or changing what they think about? In C. H. de Vreese (Ed.), The dynamics of referendum campaigns in international perspective (pp. 63–83). London: Palgrave.Find this resource:

Marsh, M., & Mikhaylov, S. (2010). European Parliament elections and EU governance. Living Review in European Governance, 5, 4.Find this resource:

McLaren, L. (2006). Identity, Interests and Attitudes to European Integration. London: Palgrave MacMillan.Find this resource:

Merrill, S., III., & Grofman, B. (1999). A unified theory of voting. Cambridge, U.K.: Cambridge University Press.Find this resource:

Pállinger, Z. T. (2017). The Hungarian migrant quota referendum of 2016 in the context of Hungarian direct democracy. In F. Mendez & M. Mendez (Eds.), Referendums on EU matters (pp. 156–173). Brussels: European Parliament Policy Department for Citizens’ Rights and Constitutional Affairs.Find this resource:

Petersen, M. B. (2010). Distinct emotions, distinct domains: Anger, anxiety and perceptions of intentionality. Journal of Politics, 72(2), 357–365.Find this resource:

Piris, J.-C. (2006). The constitutional treaty: A legal analysis. Cambridge, U.K.: Cambridge University Press.Find this resource:

Prosser, C. (2016). Calling European Union treaty referendums: Electoral and institutional politics. Political Studies, 64(1), 182–199.Find this resource:

Ray, L. (2003). Reconsidering the link between incumbent support and pro-EU opinion. European Union Politics, 4(3), 259–279.Find this resource:

Reif, K. H., & Schmitt, H. (1980). Nine second-order national elections: A conceptual framework for the analysis of European election results. European Journal of Political Research, 8(1), 3–44.Find this resource:

Schuck, A. R. T., & de Vreese, C. H. (2008). The Dutch no to the EU constitution: Assessing the role of EU skepticism and the campaign. Journal of Elections, Public Opinion and Parties, 18(1), 101–128.Find this resource:

Schumacher, W. (1932). Thirty years of the people’s rule in Oregon: An analysis. Political Science Quarterly, 47(2), 242–258.Find this resource:

Sniderman, P. M., Brody, R. A. & Tetlock, P. E. (1991). Reasoning and Choice: Explorations in Political Psychology. Cambridge, U.K.: Cambridge University Press.Find this resource:

Steenbergen, M. R., & Siczek, T. (2017). Better the devil you know? Risk-taking, globalization and populism in Great Britain. European Union Politics, 18(1), 119–136.Find this resource:

Svensson, P. (1994). The Danish yes to Maastricht and Edinburgh. The EC Referendum of May 1993. Scandinavia Political Studies, 17(1), 69–82.Find this resource:

Svensson, P. (2002). Five Danish referendums on the European Community and European Union: A critical assessment of the Franklin thesis. European Journal of Political Research, 41(6), 733–750.Find this resource:

Szczerbiak, A., & Taggart, P. (2004a). The politics of European referendum outcomes and turnout: Two models. West European Politics, 27(4), 557–583.Find this resource:

Szczerbiak, A., & Taggart, P. (2004b). Conclusion: Towards a model of (European) referendums. West European Politics, 27(4), 749–777.Find this resource:

Taber, C., & Lodge, M. (2006). Motivated scepticism in the evaluation of political beliefs. American Journal of Political Science, 50(July), 755–769.Find this resource:

Taber, C. S., Cann, D., & Kucsova, S. (2009). The motivated processing of political arguments. Political Behavior, 31(2), 137–155.Find this resource:

Triga, V., & Manavopoulos, V. (2017). The Greek bailout referendum of 2015. In F. Mendez & M. Mendez (Eds.), Referendums on EU matters (pp. 119–141). Brussels: European Parliament Policy Department for Citizens’ Rights and Constitutional Affairs.Find this resource:

UK. (2016). Election Results. Available online.

Valentino, N. A., Hutchings, V. L., & White, I. K. (2002). Cues that matter: How political ads prime racial attitudes during campaigns. American Political Science Review, 96(1), 75–90.Find this resource:

Visser, P. S., Bizer, G. Y., & Krosnick, J. A. (2006). Exploring the latent structure of strength-related attitude attributes. Advances in Experimental Social Psychology, 38(6), 1–67.Find this resource:

Visser, P. S., Krosnick, J. A., & Simmons, J. P. (2003). Distinguishing the cognitive and behavioral consequences of attitude importance and certainty: A new approach to testing the common-factor hypothesis. Journal of Experimental Social Psychology, 39(2), 118–141.Find this resource:

de Vreese, C. H. (2007). Context, elites, media and public opinion in referendums: Why campaigns really matter. In C. H. de Vreese (Ed.), The dynamics of referendum campaigns in international perspective (pp. 1–20). London: Palgrave.Find this resource:

de Vreese, C., & Semetko, H. (2004). Political campaigning in referendums: Framing the referendum issue. London: Routledge.Find this resource:

Zaller, J. (1989). Bringing converse back in: Modeling information flow in political campaigns. Political Analysis, 1(1), 181–234.Find this resource:

Zaller, J. (1992). The nature and origins of mass opinion. Cambridge, U.K.: Cambridge University Press.Find this resource:

Notes:

(1.) This excludes the French 1972 referendum on enlargement of the EEC as well as also Swiss and Liechtenstein’s referendums on EFTA and EEA.

(2.) While it cannot be reasonably expected that average voters engage in a synoptic evaluation of the proposition in relation to their IP and the expected RP, voters can utilize heuristics like cues from referent persons to calculate their position “as if” they had expended the analytical resources to calculate their utility function (Hobolt, 2009; Lupia, 1992, 1994; Lau & Redlawsk, 2006; Sniderman, Brody, & Tetlock, 1991).

(3.) Berinsky and Lewis (2007) question whether voters ever have more than general notions about the location of a proposition and the RP, but many other scholars contend that while voters do not have the information to actually engage in a fully synoptic utility calculation, they can utilize different heuristics when issues are salient, like cues from referents to enable them to make a decision “as if” they had engaged in this type of calculation (Hobolt, 2009; Lupia, 1992, 1994).