Participation in elections to the European Parliament

An econometric analysis of the effects of media mobilisation and social capital on voter turnout

José Luengo-Cabrera, Nakita Patel & David Schlosberg

Characterised by significantly low turnouts, elections to the European Parliament (EP) have been dubbed as being of ‘second-order’ due to their inferior importance vis-à-vis national elections (Reif & Schmitt 1980). Previous research (Reif 1984; Van der Eijk & Franklin 1996; Van der Brug et al. 2007; Hix & Marsh 2011; Hobolt & Wittrock 2011) shows that although they determine the composition of a supranational parliament, the underlying motivational drivers for participation in EP elections reside in the national realm. For this particular reason, we base our investigation on national attitudinal indicators considered to have a predominant influence on EP voting behaviour.

Using data from the 2005 British Social Attitudes (BSA) survey, we yield an econometric model to measure the magnitude and significance that a combination of socio-economic factors had on the turnout of the 2004 United Kingdom (UK) elections to the EP. While we include some variables that are prominent throughout the literature, the focus is on the relationship between media mobilisation, social capital and voter turnout. Our results show that greater levels of media mobilisation and social capital increased the likelihood of voter participation.

Measured in terms of valid votes as a percentage of the registered electorate, the average historical turnout for the pre-Brexit UK elections to the EP is 33.8% (1979–2014). In the elections of 2004, 2009 and 2014, voter turnout was 38.2%, 34.7% and 35.6%2, respectively — with 2004 being the year marking the highest turnout since the UK’s first participation in 1979. However, these figures remain significantly low compared to national ones.

With an average historical rate of 70% (1970–2010), voter turnout to the UK general elections of 2001 (59.5%), 2005 (61.4%) and 2010 (65.1%) demonstrate that EP elections fall short of being democratically representative. This investigation will explore the nature of the discrepancy between voter turnout at national and European elections, with a particular focus on the positive externalities derived from implementing policies that increase the levels of media mobilisation and social capital at a time of asserted EU-wide political apathy (Bilska 2009; Malkopoulou 2009; Delwit 2013).

This paper is organised as follows. First, we will provide a literature review on the determinants of political participation, followed by an exploration of the prevalent theories on voting behaviour — with particular attention to the way in which media mobilisation and social capital play a role in prompting voters to the polls. As we set the ground in mapping out the factors that appear to be salient drivers of voter participation, we provide a diagnostic of our dataset and guide readers through a breakdown of the variables of interest.

Subsequently, we introduce our econometric model and present the results. We then scrutinize the robustness and external validity of our model — stressing the underlying caveat that although the use of statistics to evaluate the nature of political processes and inform policymaking is at best imperfect — it is a useful tool nonetheless. By highlighting the possible biases emanating from the statistical limitations of our dataset, we attempt to provide qualitative reasoning as to why we believe our model has a certain degree of empirical legitimacy.

Finally, we turn to policy recommendations. We suggest a series of initiatives that European governments could implement to bolster voter participation in future EP elections. We conclude with the evaluative idea that the European Union (EU) continues to suffer from a ‘democratic deficit’ (Crombez 2003; Peter & de Vreese 2004). This implies that there is much room for improvement, particularly in light of the viability to boost voter participation via policies tailored to enhance citizenry knowledge of the EU and the role they can play as active regulators of the EU’s decision-making power.

On the determinants of political participation: theory & evidence

The concept of political participation is one that has generated a vast amount of debate among political scientists, not only by the crucial role that it plays in any society, but by the difficulty in conceptualizing it and, most importantly, quantifying it. Political participation can be defined as “the actions by which individuals take part in the political process” (Axford et al. 2002:120) yet this clearly indicates that the term is ambiguous, especially when we question what we mean by action and what could in actual fact be considered as political.

Regardless of the diverse views on human nature, it is undeniable that any individual citizen seeks, at the very least, security and well-being. In turn, political participation may be seen as a natural stimulus that is incited whenever societal well-being has deteriorated or is threatened by the actions of the governing elite. In the context of any established democracy whereby its citizens have tacitly given their consent to abide by a social contract and be ruled by a government responsible for the well-functioning of a society, the “collective endeavour to promote change in any direction and, by any means, not excluding violence or illegality” (Benny 2003:165) is the way in which citizens mobilise to influence public policy whenever their rights and freedoms are not being guaranteed by the government.

Any citizen within a democracy has the right to partake in politics. Whether citizens get involved with a political party or participate in demonstrations, the scope for political engagement is extensive, but the ability to measure its intensity is rather limited. While the “relative strengths of political parties can be measured by their electoral performance, the strength or influence of social movements is more difficult to quantify” (Benny 2003:165). Therefore, it is the quantification of participatory levels that poses a major obstacle and this is highly correlated with the difficulty to define it.

In this investigation, political participation will be defined as any sort of social movement whereby individuals congregate and act together to engender change or preserve the status-quo in their society. The problem remains, as aforementioned, that given the ambiguous scope of its definition, it becomes elusive to quantify. Fortunately though, electoral participation appears as a subset that facilitates measurement given that statistical data on voter turnout can be collected. A fundamental question therefore lies in the analysis of how many citizens (out of the total population) vote and why. This will often depend on a multitude of factors that may influence citizens’ proclivity to be politically active. Undoubtedly then, personal or collective endowments such as levels of education, living standards and the degree of citizens’ reliance on the delivery of public goods will tend to determine their propensities to participate in politics. In turn, factors such as social exclusion or injustice may prompt a sudden surge in political protest in the form of a collective gripe for change. Political participation may then be considered as a dependant variable on a number of socio-economic factors that precondition and influence its intensity.

The question regarding political efficacy may largely determine why individuals may decide to engage in politics (McCluskey et al. 2004). In addition, the influence of an overtly devoted welfare state may incline citizens to be incrementally passive (Hvinden & Johansson 2007). Activeness in politics is then a matter of social stimuli that will tend to determine the degree to which individuals partake in politics — either to demand change in policies or to preserve the prevailing modes of governance. Any given stimulus with a threshold strong enough to enable individuals to become politically active may then alternate the levels of political participation (McCluskey et al. 2004). Public policies, the discretionary power of an elected government or rational economic interests may be important drivers for citizens to become active participants in political processes. Nowadays, with an overwhelming majority of states being democratic, political scientists are concerned with the health of democracies — the amount of regular input that citizens are able to provide so as to ensure the accountability and effectiveness of those granted the power to govern.

The allegedly deteriorating health of European democracies can be approximately evidenced by the considerable decline in voter turnout across countries in both national and EP elections — with the latter demonstrating a steeper declining trend, as shown in Figure I. Measured by the degree of prevailing “social networks, norms and trust that enable participants to act together more effectively to pursue shared objectives” (Putnam 1994: 664); social capital seems to have declined in Europe. The idea is that European citizens seem to have become increasingly individualistic and apathetic. They appear to have become decreasingly active in social congregations, thereby weakening the potential for collective action (Olson 1965).

However, this evaluation of a progressive decline in political participation suffers from intertemporal inconsistency. Indeed, there has been an evident revival in political participation across Europe in the wake of events triggered by the recent financial crisis, with European citizens mobilising against austerity policies at a time of anaemic economic growth and excessive public debt. This may explain the post-2009 flattening in the EU average EP curve in Figure I.

In spite of the increasing role played by the EU in all aspects of European citizens’ lives, the power to implement economic and social policy remains largely within the purview of national authorities. As a result, citizens continue to view their governments as the primary sovereign agents in the realm of policymaking, thereby perceiving the EP as a supra-national legislative body whose authority is subordinate. As a result, EP elections are generally not considered of primordial importance and this is evidenced by the significantly low historical turnouts in EP polls relative to turnout in national elections (except for the cases of Luxembourg, Belgium, Greece (and Cyprus); where voting is compulsory) — as shown in Figure II. Acknowledging this reality, a considerable amount of literature has been centred on the nature of voting at EP elections. In complement, our investigation will focus on the determinants of voter turnout at EP elections, with particular attention to the positive externalities generated by increased levels of media mobilisation and social capital.

Voting is by far the most widespread and conventional form of political participation. It is the act by which members of a democratic society are legally empowered to elect the governing body that is to preside over a sovereign state. Voting is commonly seen as a habit that citizens learn throughout their lives; influenced by personal attributes such as education, financial endowment and political party affiliation (Plutzer 2002). But the act of voting is mainly about public interest - it concerns the communal act carried out by citizens in nominating their political representatives through the ballot box and subsequently providing recurrent input (via popular referenda, pressure groups or social movements) to regulate the way in which the government distributes public goods, implements the rule of law and manages the economy. Therefore, direct participation in politics has become the norm in representative democracies, to the extent that citizens have become increasingly empowered to have a say on the policies that affect their daily lives. A paradox however emanates out of the reality that in spite of being incrementally empowered, citizenry participation in politics has continued to decline in past decades.

A fundamental trend that has been characteristic in post-war Europe has been the relative decline in political participation. Since the 1960s, party membership began to witness a steady decline, with single-issue movements (i.e. nuclear non-proliferation, environment) attracting an increasing amount of supporters; prompting a subsequent rise in the frequency of social protests in the 1980s and 1990s (Inglehart & Catterberg 2002). Already in the 1970s, politics had started to become individualized (Dalton 2004), whether in terms of political participation or campaigning styles (Bosancianu 2014). Organized politics began to progressively lose its appeal, with political party and trade union membership witnessing a gradual decline as a result of their perceived inability to address the challenges brought by the forces of globalization. As a result, political actors began to recur more and more to alternative modes of political communication, evidenced by the rise of “mass appeals through television and campaign public relations” (Bosancianu 2014:6) — directed at large to an ever more apathetic citizenry.

In the past two decades, political entrepreneurs have increasingly relied on the advantages brought by information & communication technology (ICT); most notably the internet and, in recent times, social media. To date, the whole spectrum of political actors - from political parties to trade unions, civic organizations and denominated “terrorist” groups - are capitalizing on the proliferation of online communication platforms to disseminate information and bolster political engagement, whether peaceful or violent. While ICT has brought obvious benefits in the form of reduced information asymmetries and increased ability to hold political representatives accountable, it has contributed to a relative drop in the levels of politically-motivated communal gatherings. With the rising phenomenon of clicktivism, citizens (especially young ones), have become more politically active online, but not necessarily offline.6 In spite of the popularity of youth-led global advocacy organizations like Avaaz and their success in organizing cross-issue mass demonstrations in major cities across the world, a well-documented fact is that global voter turnout continues to be lowest for citizens in the 18-30 age cohort; a worrying reality when considering that they are estimated to constitute 50.5% of the world population (UNESCO, 2012).

In the case of EP elections, high rates of youth absenteeism is often attributed to the failure of political parties in addressing youth issues. Indeed, in a 2013 Flash Eurobarometer on European youth participation in democratic life, the most frequently cited reasons for not voting at EP elections were largely attributable to the belief that their vote would not change anything and that the EP does not sufficiently deal with problems that concern the young voters (among others), as shown in Figure III below.

Acknowledging the aforementioned decline in social capital and the rising importance of ICT, it is important to consider the way in which fluctuations in the supply of these two can have considerable effects in altering the levels of electoral participation. With the idea that the salience, competitiveness and exposure of an election can mobilise voters (Rosenstone & Hansen 1992), we generally expect greater turnouts where campaigns are more intense, where media coverage is tailored to the issues of the elections in question and where the perceived sense of collectiveness and urgency for change within a citizenry is greatest.

Prompting voters to the polls

Media mobilisation

By lowering the costs of information-gathering, news media (whether print, online or televised) tends to (re)generate general interest in politics, particularly in the lead to elections. This often prompts nation-wide and trans-European televised debates brought about by a burgeoning of forum talks organized by political parties and civic organizations that usually enhance citizenry interest in politics. As a result, greater media exposure and mobilisation tends to increase the likelihood of participation in elections (Banducci & Semetko 2003). A greater volume and velocity of media coverage also increases the perceived benefits of voting, principally by signaling to voters that the outcome of the election is important (Cox 1999). Despite the breadth of literature on theories of media malaise, which argues that exposure to news media can discourage learning about politics, erode trust in political leaders and dampen political mobilisation (Newton 1999), the existing empirical evidence to the contrary suggests a positive relationship with voter turnout (Norris 2000). Moreover, Norris and Sanders (2003) find significant effects of television news, broadsheet newspapers and surfing party websites on knowledge gains concerning politics, with tabloids being found to elicit significantly fewer knowledge gains when compared with broadsheets.

Most recently, with the exponential growth of citizens and public servants joining online platforms like Twitter and Facebook, social media has become one of the most effective tools for political communication. With a rising number of politicians opening online accounts, voters have gained direct access to their publicised thoughts. This has effectively enabled voters to read summaries of candidates’ manifestos, opinions, reactions and policies ahead of the polls in question. In turn, political parties are recurring more and more to social media outlets to gain a better understanding of the electorate’s demands - in the hopes of aligning their campaign strategy and policies as close to the median voter as possible.

With nascent studies evaluating the extent to which social media has bolstered the visibility of politicians vis-à-vis the electorate (Vergeer et al. 2011, 2012), there has not yet been compelling evidence regarding the positive correlation between online mobilisation and offline participation (Bosancinau 2014). Although causality links are yet to be consolidated, the 2009 and 2014 EP election campaigns shed light on the rising reliance on social media platforms as popular means for electoral ends. Notwithstanding the counterproductive effects of information overload, it will become increasingly harder to refute the idea that social media is contributing to the enhancement of the connectivity between the electorate and those who are running for office – even though this will not necessarily prompt voters to the polls. In the case of EP elections, some even claim that this has had counterproductive effects.

From media mobilisation to social capital

Individuals with political party affiliation should in theory be more inclined to vote. As such, we expect party partisanship to act as a strong motivator for participation. This effect should be compounded by the expectation that the hurdle in moving from non-voter to voter is higher than the hurdle in moving from a national election to an EP election (Banducci 2005). Politically active individuals are aware of the importance of political participation and will tend to encourage peers to vote, thereby generating unquantifiable positive externalities in regards to local and nationwide participative encouragement.

However, we also need to look at the individual factors that make individuals more likely to vote. The Lijphart Thesis suggests that those who lack resources such as education and income are less likely to vote (Lijphart 1997). We therefore expect education and income biases to be greatest in a low turnout election like the EP. Education reduces the costs of information-processing and increases feelings of civic duty through a socialisation process. Education also helps to determine one’s level of prior political knowledge. If the assumption of decreasing returns to knowledge- acquisition holds, this should mean that, ceteris paribus, exposure to media and campaigning will generate the greatest knowledge gains amongst those with lower levels of education (Norris and Sanders 2003). Therefore not only do we expect those with higher education qualifications to be more likely to vote, we also expect the effects of media mobilisation and social capital to be stronger for individuals with lower educational qualifications. Given the intricate relation between personal endowments and media mobilisation, this provides a good segway into exploring how the levels of collective trust, empathy and solidarity have an additional effect on the propensity for individuals to vote.

Social Capital

Despite the rise of the modern electoral campaign model characterised by the widespread adoption of political marketing techniques (Swanson & Mancini 1996), we still expect membership to political parties, trade unions and civic associations to stimulate voter participation. Much social capital literature (Putnam 1995; Portes 2000; Adler & Kwon 2002) emphasizes the importance of social bonds in turning individuals into political stakeholders. Putnam (1995) suggests that voter turnout, news exposure and membership to leisure/civic/political associations are all interrelated. This tripartite relationship is a core focus of this investigation.

In our model presented below, we use home ownership as our key proxy for social capital as we abide by the assertion that “homeowners invest more in their neighbourhoods and engage more in civic activities”. Home ownership weaves individuals into the fabric of their community; they are more likely to both know and interact more with their neighbours and political representatives than transient home renters. As such, compared with the latter, they will tend to be more inclined to vote. Despite the second-order nature of EP elections - whose outcome might seem too dislocated from local communities - the correlation between homeownership and political participation at a national level has been found to be a strongly significant one (DiPasquale & Glaeser 1999; Engelhardt et al. 2010McCabe 2013; Rohe 2013) and we believe this relationship should trickle down to the communal level.

Considering the wider context of transnational issues such as environmental policy, we expect membership to environmental groups or strong sympathies with the cause to increase the probability of voting. Communal interest in policy that is influenced and directed at the supranational level should increase an individual’s perceived benefits from voting. This may also be true of those who have strong sentiments (positive or negative) with regards to other trans- national/European phenomena such as immigration and monetary policy. Sensitivity to European policy should also be affected by age, which has proven to be a strong predictor of turnout across a wide range of studies (Blais 2000; International IDEA 2002).

All in all, individuals’ degree of integration within their local community is seen as an important factor in determining their propensity to vote. The stakes are perceived to be higher when individuals have a sense of belonging to a collectivity, particularly in view of the fact that they are likely to use elections (whether local, national or supranational) as a means to express their (dis)satisfaction with the government’s actions. This implies that the levels of social capital matter for the way in which communities engage in collective action – participating in polls to reward or punish their elected leaders is just one example. We therefore expect social capital to act as a vehicle for electoral mobilisation, particularly under the assumption that groups with strong links will be driven to further their collective interests whenever they perceive that the elections in question are salient.

In the UK, for example, citizens that have more to gain/lose from policies directed at the European level will be more likely to act in concert and participate in EP elections. One clear example is the assertion that British farmer communities that benefit from the EU’s Common Agricultural Policy are more likely to mobilise to ensure that they elect representatives that have pledged to preserve and further British farmers’ interests in Brussels or Strasbourg. As a result, and in spite of the second-order nature of EP elections, we see high levels of social capital as an ex-ante determinant of citizens’ propensity to partake in these elections; particularly when their livelihoods depend largely on the formulation and implementation of policy at the EU level.

Having mapped out our theoretical expectations, we proceed to explore how we selected our dataset and constructed our model. The end-goal is to test whether our ex-ante projections on causality align with those of the model’s outcome.

Data and Statistical Synopsis

We use data from the 2005 BSA survey; enabling us to construct various attitudinal measures. The percentage of respondents who voted in the 2004 UK elections to the EP from the BSA dataset was 36.6% - a good approximation to the official 38.2% turnout. We therefore deemed the survey to be a sufficiently, albeit imperfect, representative sample of the population in Britain aged 18+.

Our variables for media mobilisation were proxied using a variety of political news sources such as television (politicstv), internet (net) and broadsheets (broadsheet) that represent the respondents who used these as means to inform themselves on political issues. For social capital, proxies were represented by measures of political interest (highpolint) and home ownership (ownhome). Figure IV shows the average percentage of surveyed respondents who owned a home (69%), had high political interest (33%) and used political news conveyors, from which internet usage stands out with a 56% of respondents, relative to the 24% and 22% for television and broadsheet users, respectively.

Our interest is to find out the percentage of this set of respondents who voted. We find that 79% of homeowners voted whilst the figure was 49% for those who had high political interest. Out of the respondents who regularly watched politics on TV, 24% of them voted whereas it was a figure of 30% and 59% for those who read broadsheets or obtained information from the internet, respectively — as shown in Figure V. This implies the idea that the internet was the most popular conveyor of political information for voters in the 2004 UK elections to the EP.

In addition, we want to find out the relationship between our variables of interest and a variety of other variables that influence them. As shown in Table I, we find positive correlations between high levels of education (highed) and (highincome) with our respective measures of media mobilisation and social capital, except for television usage, which exhibits a rather low correlation. The elderly cohort (55+) is positively associated with homeownership and high political interest, it is the exact opposite for the youth cohort (18–24). This implies that civic engagement is likely to be stronger amongst the older age cohort. Furthermore we find a positive correlation between the youth cohort and watching the news on the internet — a relation which is negative for the elderly cohort — suggesting that the internet could have been the means by which to increase youth participation at the 2004 UK elections to the EP the most; in light of the fact that a staggering 67% of people aged 18–24 did not vote.

This diagnostic sets the ground for the construction of our econometric model. Combining knowledge of prevailing theories on voting behaviour and the preliminary results obtained from variable interactions in the BSA survey, we proceed to measure the magnitude of the effects that our variables of interest had on the propensity of individuals to vote at the 2004 UK elections to the EP.

Model

We constructed a Probit model generating a numerical value for the likelihood to vote in EP elections, after controlling for selected factors believed to affect a respondent’s voting proclivity. Within a sample of people who voted in the 2004 UK elections to the EP, the model was able to quantify the marginal effects that these factors had on the dependent variable (votedeu). In other words, it assigned a numerical magnitude that represented the probable change in value of the dependent variable for each unit change in the explanatory variables of interest. 9

We present our model in a series of four regressions, as seen in Table II (p. 20). Regression (1) includes all of the socio-economic control variables10 that we predict to have an effect on voter turnout. Regression (2) includes the effects of media mobilisation variables and regression (3) adds the effects of social capital. Regression (4) below is the aggregated, complete model:

Results

As expected, factors such as political activity and affiliation, media consumption, social capital, education, income, age and gender all showed the expected causal direction and significance (although the statistical significance of income and education was diminished with the addition of factors in the augmented regressions). We analyse our results using a 15% significance level, based on the known limitations of our dataset and the implications they had on our results, as explored in the model caveats section further on.

We found a positive and significant relationship between media mobilisation and voter turnout. Our findings show that individuals that watched news on current affairs were 12.5% more likely to vote than those that didn’t. Furthermore, someone who read broadsheets was 8.7% more likely to vote than a tabloid reader — whilst the figure was 3% for internet users versus non-users.

For political activity, it can be seen that whether an individual supported a political party or had high political interest had a significant and positive effect on voting. The marginal effect of high political interest implies that, ceteris paribus, individuals with high political interest were 10.9% more likely to vote than those with lower levels of political interest. In turn, those who supported a party were 15.9% more likely to vote than non-supporters.

We expected strong views towards EU policy to have a significant impact on voting. We found that individuals with strong sentiments (positive or negative) towards the Euro were 13.8% more likely to vote than someone who was euro-apathetic. In addition, individuals with negative views on EU immigration policy were found to be 5.7% less likely to vote than pro-immigration individuals. This approximately implies that the anti-immigration group is more likely to boycott EP elections in light of the fact that EU policies on immigration have largely been tilted towards greater trans-European mobility, as evidenced by the implementation of the Bolkenstein Directive or the creation of the Schengen area. It is probable that anti-immigration groups would be attempting to undermine the validity of EP elections by abstaining from voting as a protest to the UK government on its position vis-à-vis immigration policy. However, given the recent (re)surge of populist right-wing parties advocating nationalistic policies11 in major countries like the UK and France, were we to re-run our regression with 2009 and 2014 electoral data, the causal direction is likely to be inversed; with anti-immigration voters much more likely to vote given that their asserted interests are embodied in the policies promoted by parties like the UK Independence Party (UKIP) or the Front National.

We anticipated higher levels of income and education to have a positive impact on voting behaviour, which was confirmed by our results but fell short of statistical significance in the final regression; thereby preventing us from deriving any valid conclusions based on statistical inference. In line with our expectations, our variables for social capital (home ownership, trade union membership, membership to an environmental group) all showed that high levels of civic engagement increased the likelihood of voting. Our results confirmed this hypothesis as a home owner was 6.5% more likely to vote than a renter, while a trade union and environmental group member was 7% and 15.4% more likely to vote than non-members, respectively.

Using interactive dummies to measure the additional effects of broadsheet readership on home owners, we found that home renters that read broadsheets were 12% less likely to vote than individuals who owned a home and read broadsheets. This entails, approximately, that the marginal effect of broadsheet readership on voter turnout was higher for those with higher levels of social capital.

Based on decreasing marginal returns to education, we expected the impact of additional information contained in broadsheets to have a greater influence on (and thus affect voting more) a high-school educated person than someone with a tertiary education, since the former are assumed to be less information-saturated. Interacting broadsheet readership with education, we found that an individual who read broadsheets and had a high-school education was 3.2% more likely to vote than an individual who read broadsheets and had attended university. This result however proved to lack adequate statistical significance, thereby not allowing us to reject the null hypothesis with sufficient confidence.

Overall, the model output proved to align well with our ex-ante hypothesis on the direction of causality. However, many data limitations played a hindering role on the validity of the model. This was mainly due to the very nature of the survey used. The BSA survey was broad, focusing on extensive social attitudes rather than drawing information specifically about attitudes towards Europe and the EP elections (granted that this was not the principal objective of the BSA survey anyways). Our dataset had a lot of missing values as the survey was split into four versions; generating missing data and reducing the model’s explanatory power. We think that this is captured in our low Pseudo R-squared value of 0.1425 for our final model, suggesting it had the potential to improve by 85.75%. In light of this and the unavailability of a better dataset at the time the statistical analysis was conducted, our results continue to be relevant although their validity for informing policy will remain suggestive. As a result, our statistical output needs to be complemented with context-specific qualitative analysis. This is a requisite if we are to discern and make full sense of the multitude of confounding factors that affected voter turnout but were not captured by the data.

Model Caveats

External validity: The trap of British idiosyncrasy

“External validity asks the question of generalizability: To what populations, settings, treatment variables and measurement variables can this effect be generalized?” Campbell (1966)

Like all investigations in social science research, this one must ask itself the question of whether the econometric results from the sample used are generalizable to a wider population; namely the EU population as a whole.

This investigation faces two main problems of external validity: 1.) it is based on a sample of a national attitudinal survey as opposed to an EU-wide attitudinal survey and 2.) it is based on a sample of a UK attitudinal survey.

The first problem is one that would be confronted when using any national attitudinal survey and the challenge would be to inquire into whether it would be valid enough to make EU-wide inferences on the basis of the results of a national survey. This question is difficult to answer due to the multifarious idiosyncratic characteristics and domestically-restrained developments that take place in each EU country before, during and after EP elections. As such, the external validity of this investigation is by default limited but only because it is confronted with an inevitable trade-off: use a national survey to capture attitudes towards nation-specific developments or use an EU-wide survey that predominantly uses cross-country aggregated averages that may not account for specific attitudes that can only be captured through national surveys. Although there exists a middle-way in the form of a hybrid survey accounting for both national and cross-national attitudes, this method has proven to be financially cumbersome, thereby favouring an either/or approach. Although some progress has been made with the more elaborate Eurobarometer reports appearing in recent years, they still fall short of the somewhat ideal trans-European surveys capable of capturing all of the factors and effects that are inevitably omitted in current practice. This however, is a problem confronted in all fronts of social science research. Therefore, the authors acknowledge and allow for such imperfection, esteeming this study to be a marginal contribution to the understanding of voting behaviour in EP elections while taking full responsibility of the deficiencies emanating from this imperfection. We however attempt to address this issue via the complementary qualitative analysis provided throughout this investigation, while remaining humble on the validity of our assertions which, as aforementioned, remain suggestive in nature.

The second one is trickier and indeed more sensitive as the UK is notorious for its euroscepticism. One only has to look at the historical positions taken by the UK government on the EU since the times of Ms. Margaret Thatcher; most notably the infamous UK rebate over British contributions to the EU budget and more recently to the popular rise of UKIP or the electoral pledge made by Mr. David Cameron on holding a referendum on EU membership in 2016/7 if the Conservative Party wins the 2015 general elections. Somewhat more objectively, a 2010 Eurobarometer report shed light on the fact that UK citizens were, according to the survey results, “divided” in regards to their support for the EU - with 42% of respondents having a “rather positive image of the EU” and almost as many (39%) having a “rather negative one.

As we explore in the sub-section below, the choice of dataset was mandated by the necessity to capture nation-specific attitudinal factors due to the underlying assertion that EP elections are in effect about national developments. The particular choice of the UK was guided by the academic intrigue to try and unravel the sort of policies that could be implemented to increase, even if by a minimum amount, an ostensibly Eurosceptic citizenry’s participative proclivity in EP elections. Against the backdrop of a rather inevitable, albeit slow, power transfer from national to supranational authority, the election of parliamentarians of a supranational legislative body is becoming increasingly important. In turn, the low participatory rates in the concerned elections will become gradually troublesome as we witness the progressive necessity to deal with transnational issues at a European level.

In sum, we caution on the limited external validity of our statistical results but we encourage future investigations to ensure the integration of both nation and EU-specific surveyed attitudes to their best of their abilities, with the recommended attempt to construct models that can account for the socio-economic idiosyncratic characteristics of the EU member states in question.

Choice of dataset

In regards to the discretionary choice of the dataset, at the outset, the two same questions arise once again: 1.) why a survey based on national attitudes and 2.) why a UK survey.

The decision to employ a national survey was principally due to the underlying endeavour to match our dataset selection with the assertions evidenced in the existing literature on the subject. If EP elections are indeed primordially influenced by domestic factors, employing an EU-wide survey could undermine the effective capture of nation-specific attitudes towards their national governments and the EU. We deemed that the EU-wide datasets were deficient in the sense that they were omitting context-specific attitudes of the surveyed citizens on the domestic issues that were of concern at the time the survey was conducted.

In turn, the recourse to a UK attitudinal survey arose from the authors’ curiosity to find out what could be done to increase participation in the most Eurosceptic country of the EU or, in other words, exploring the country where the potential for increased participation in EP elections was, a priori, greatest. This led to a surprisingly positive outcome in what regards the scope to increase British citizens’ propensity to vote at EP elections in future, something that we deem to be a noble endeavour. We explore this in our policy recommendations section below.

Endogeneity

We are aware that data limitations may have caused our model to fall victim of the problem of endogeneity. In accordance with the reward-punishment model of voting (Kramer 1971), we believe our data may have produced a positive bias on the respondents’ propensity to vote because the data was collected during a period of relative economic growth in the UK; preceding the latest financial crisis. This was a time during which the UK’s citizens’ belief that they were benefiting from being a member of the EU was at a record high (for UK standards, biensûr) of 39%.13 In prosperous times, it is believed that voters use second-order contests to manifest their support for the incumbent government if it is perceived to be managing the economy well (Tilley et al., 2008). Hence, since the survey was conducted during a period of economic prosperity, the suspected bias emerges from the possibility that voter turnout may have been endogenous to their evaluation of the state of the economy rather than by their evaluation of the EU. This bias could potentially be corrected with an instrumental variable: correlated to the voters’ perceptions based on their knowledge of the EU, but independent of their evaluation of the incumbent government’s management of the economy. The authors were unable to find a variable that could align with these criteria.

In addition we deem that the strongeuro variable may have also suffered from a positive endogeneity bias as there exists the possibility that other EU-related attitudinal measures were inexistent in the survey; causing a potential over-evaluation of the effects of this variable. We stress future investigations on this subject to ensure the inclusion of a greater variety of EU- related attitudinal measures in order to avoid the problems emanating from omitted variable bias, which will largely depend on the extent to which offices of national statistics are able to cooperate with Eurostat on the collection of EU-specific attitudinal measures.

Following this attempt to provide an overview of the main limitations of our econometric model, we deem this to be an opportune moment to explore a non-exhaustive review of policy recommendations, based principally on the results of this investigation, while relying on the findings of post-electoral surveys carried out by the European Commission.

Policy Recommendations

How to bolster voters’ proclivity to vote at EP elections?

The results presented to the reader shed light on the sort of policies that can be implemented to buttress the propensity of citizens to participate in the election of national representatives to the EP. Given our focus on media mobilisation and social capital, we will limit ourselves to making recommendations on the basis of these two phenomena.

Mobilise the media industry

For media mobilisation, the underlying premise relies on the documented reality that, on average, European citizens do not feel sufficiently informed about the EU and the role/impact of the EP more concretely. For this reason, there is a need for the media industry to mobilise further so as to disseminate the relevant information to trans-European audiences. Granted the appraised independence of the media, one can make the assumption that the discretionary power of media enterprises, particularly private ones, allows them to convey information that they deem most relevant (and indeed most commercially attractive). When their main source of revenue depends largely on publicity - which itself is determined by how entertaining or interesting their print/online/television news is to an audience - this may lead to possible distortions in the a priori objective distribution of information. This is in large part a hindering reality for participation in EP elections given its alleged second-order nature.

If these elections are not considered salient by the citizenry, media outlets are bound to tailor their information to the preferences of their audience. More often than not, media outlets may tend to minimize the exposure of informed debates and objective information on political developments - sometimes to the benefit of sensationalist reporting on candidates’ private lives or unfounded scenario projections based on the victor of the concerned elections. In the run-up to elections, and in particular general elections, the attention is sometimes overwhelming to the extent that it may in effect discourage citizens from voting. The confluence of bigoted accusations, elusive promises and exaggerated worst-case scenarios if the “other party” wins often trump the more sensible and objective manner in which information on the political environment could be presented. Having said that, there is evident scope for the media industry - with the help of public authorities through systems of subsidy - to improve its ability to abide by its objective of delivering a wide array of information and, in this particular case; information related to European politics.

With several studies alluding to the positive effect of media mobilisation on voter turnout (Baek 2009; de Vreese & Semetko 2002; Norris 2006; Bilska 2012), evidence that has been largely documented by the Eurobarometers over the past years points to the fact that low voter turnout at EP elections is largely attributable to a low level of knowledge and interest on the EU. Given that EP electoral campaigns do not tend to be as salient as national ones, the media industry tends to adapt to this preference-identity, thereby matching the ex-ante low interest in EU politics with a relatively low coverage of the subject matter. This has led to a condition whereby the ability to bolster voter turnout may indeed depend exclusively on the media’s ability to inform the public on the developments concerning EP elections. But we find ourselves in a reinforced cycle of low media mobilisation engendering low voter turnout and vice-versa — an unfortunate Catch 22 situation. The authors, and many other academic researchers for that matter, believe that there is viable scope to escape this vicious trap. In addition, the authors deem this to be a democratically-motivated endeavour as it concerns the enhancement of the EU’s oft-criticized democratic apparatus. A high degree of media mobilisation and visibility of cross-cutting EU issues ahead of EP elections is therefore a desired outcome, particularly when we acknowledge that the media industry is the primary vehicle of information for the citizenry, and especially when dealing with such “remote issues as the EP elections” (Bilka 2012:2)

Based on our results, improving circulation of broadsheets to low-income areas stands out as a viable and meaningful policy strategy. To achieve this, EU governments could further subsidise the distribution costs of broadsheets in municipal areas or circulate free information leaflets prior to the elections since EU-related information is clearly a crucial vehicle for bolstering EP voter turnout. In addition, news and features on the election can be broadcasted more during primetime viewing hours. With 42% of UK citizens complaining that there is too little media coverage of the EU,15 there is an evident information gap that the UK government needs to fill given the UK population’s reputation for being the EU’s don’t know, don’t care electorate. The European Commission’s efforts in boosting the outreach capacity of its flagship Europe Direct service is an encouraging one. The authors of this investigation would therefore recommend that the current system allowing citizens to request detailed information on particular subjects be enhanced – either via an increased staffing of Europe Direct’s telephone and internet services or by the opening of more information service points in main cities across the EU.

We know that national coverage of EP elections primarily focuses on national actors (Van der Brug et al., 2007) so national actors can themselves influence voting behaviour by stressing the importance of EP elections, via social media for example. There is evidence that this was done during the 2009 elections (Vergeer et al. 2012). However, the problem remains with the reality that national politicians campaigning ahead of EP elections will tend to divert attention towards national issues as a means to increase the salience of the election, but this is largely in detriment of the voters’ ability to gauge what is at stake in relation to EU policies.

Although the causal effects of social media on voter turnout are yet to be validated, there is growing acceptance that they at least contribute to increasing citizens’ exposure to political information, particularly the youth cohort (Bosancianu 2014). We primarily recommend that EU governments promote the use of social media for electoral purposes, with both national and EU public policy institutes participating in sharing their analytical views on the political dynamics surrounding an EP election. In second instance, we would recommend national authorities to capitalize on social media platforms and facilitate partnerships between social media platforms and EU information services, so as to ensure that EU-related information reaches as wide an audience as possible ahead of EP elections.

Given that the youth cohort was found to be less likely to vote than the other age cohorts, policies could be aimed at engaging young people by increasing their exposure to EU-related information on the internet — especially in popular social networks like Facebook and Twitter — two online platforms where the young cohort takes the lion’s share in terms of usage. This was effectively the case in the 2009 and 2014 EP elections, with recent studies (Vergeer et al. 2012; Bosancianu 2014) shedding light on the tentatively positive effects of social media campaigning on voter turnout by the youth cohort. With the significant electoral weight of this cohort, national authorities and EU institutions have a greater role to play in engaging young people to become more active in providing their input during EU democratic processes, which are not confined only to elections every five years. The continued success of the Erasmus programme, the increasing amount of youth-led European civic organizations that have flourished in the past decade or the coordinating platform of the European Youth Forum (bringing national youth councils and non-governmental EU youth groups under one roof) are all encouraging signs that there are indeed attempts by national and EU authorities to engage young EU citizens in having a say in European politics. The authors encourage the continuation of these efforts and call for regular revisions of the financial allocations for these youth programmes, particularly in light of the fact that the recent EU budgetary cuts have significantly affected the Erasmus programme.

Strengthen Social Capital across the EU

Based on our results, home ownership had a significantly positive effect on voter turnout. This has been backed by numerous past evaluations validating the correlation between home ownership and voting (Rohe 2013; Di Pasquale & Glaeser 1999; Engelhardt et al. 2010), asserting that homeownership exerts positive externalities in the form of social benefits — one being increased political participation. As a consequence — and although this measure is unviable in times of fiscal austerity — EU governments could (in future) push for policies tailored to facilitating and incentivizing citizens to own a home. This could be done directly through subsidized mortgages via government-backed issues. Such subsidies could be targeted to low- income cohorts if homeownership turns out to be lower for them. Young citizens are likely to constitute a large proportion of the low-income-transient-renters cohort, thereby implying that policies need to be adapted to the financial peculiarities of young people, taking into account their shorter credit history or their lower endowment of tangible assets for the acquisition of mortgages.

Against the backdrop of the nefarious global consequences of the recent subprime mortgage crisis, we however caution against a lax approach to this policy recommendation. Indeed, if mortgages are to be subsidized to low-income households, the authors recommend governments to strengthen their capacity to regulate and audit commercial banks more strictly to prevent them from granting mortgages haphazardly. In an endeavour to avoid the systemic problems of moral hazard, the authors recommend that governments improve the capacity to independently supervise and regulate the distribution of subsidized mortgages to low-income households, whether by private or public financial institutions. This would require an enhanced legislative framework for the financial sector, with an emphasis on the necessity to improve the screening scrutiny of mortgage buyers, but without making it so stringent that it crowds out the low- income cohort as a whole.

This may entail making recourse to microcredit (yes, microcredit in Europe). Microcredit addresses the gap that has been widening between lenders and borrowers long before the latest financial crisis. With the rising importance of Small & Medium Enterprises (SMEs) in Europe and the burgeoning of start-up businesses in niche cities like London, Paris and Berlin; the need for access to credit by small enterprises is a pressing one. With the reality that SMEs can rarely access loans under €250 000, microcredit is seen as a viable alternative to allow the small-business sector to grow. Given that France and Romania are the only EU countries with specific legislation on microcredit, there is huge potential for other EU countries to regulate the supply of credit via microfinance institutions (MFIs). Currently, most EU member states restrict almost all lending to banks (European Commission 2007:4). This entails that legislation is too constricting and calls for national legislators to open up space for the operation of a range of different financial institutions, including non-banks and MFIs. These would in turn concentrate solely on services relating to the provision of credit for specific target groups that have been traditionally excluded from access to credit. By overcoming the market imperfections resulting in the selective exclusion of citizens lacking sufficient financial collateral (low-income households in general but young people, unemployed people and the elderly more specifically), microcredit emerges as a viable alternative to address the negative spillover effects engendered by this structural financial exclusion.

Given the reality that many MFIs cannot currently operate profitably in Europe (due to their high administrative costs and risk of default in relation to their profit margins), the bottleneck generated by this limitation can be attenuated with the help of EU governments. Whether via grants or tax incentives for investing in MFIs, governments can reduce asymmetries in the financial markets by aligning credit supply with demand through their support for the operations of MFIs. This plays into the idea that EU governments could do more to implement legislation regarding the financial guarantee schemes for MFIs to ensure their solvency, liquidity and commitment towards their actual and potential borrowers. Although MFI interest rates will be inevitably high, EU governments could provide subsidies to MFIs so that they can apply risk- adjusted interest rates. These, alongside government cost-covering lending to MFIs, should allow them to supply credit to otherwise ineligible homeowners possible. The social benefits of reducing credit market imperfections through the fomentation of MFIs is a well-documented reality (Besley 1994; Morduch 1999; Stiglitz et al. 2006; Gaiha 2013) that applies just as much to European countries as it has to exemplary ones like Bangladesh and Mexico.

Returning to our model results, since all of our social capital variables relating to civic and labour organization membership suggested a positive effect on voting, this entails a more prominent and active role for local governments to encourage participation in collective activities by members of local communities across Europe. This could be done through increased public expenditure and administrative assistance in reducing the financial and administrative burden for civic organizations to be set up. But once again, this measure is unlikely to be implemented in times of fiscal austerity, although some would argue that Keynesian policies is just what Europe needs to solve its most recent crisis.

Given that the asserted lack of information on the EU and its benefits has been one of the most cited reasons16 to abstain from voting in EP elections, local governments could benefit from greater assistance in setting-up EU information offices and employing civic associations to organize information sessions and distribute leaflets across municipal neighbourhoods. This would enable citizens across the EU to gain a better understanding of the functioning of the EU and the relative benefits that membership has brought to Europeans as a whole.

While there are numerous other policy recommendations to be made, we deem these to be the ones that the authors can suggest with a given degree of confidence as they emanate directly from the results of our investigation. While there are wide-ranging policy implications, we hope that the ones explored suffice for future investigations to elaborate further and contribute to informing policymakers on empirically valid recommendations capable of bolstering electoral turnout in EP elections.

Conclusion

The EU has been incessantly criticized for its democratic deficit. This investigation has attempted to provide a comprehensive analysis of the determinants of voter turnout at EP elections. It has demonstrated the significantly positive effects that media mobilisation and social capital have on citizens’ propensity to vote and has proposed a series of policy recommendations that could be implemented in future.

With the increasing role played by ICT, the authors encourage the use of communication technologies to increase citizenry knowledge of how the EU works and how policies affect their livelihoods. Reducing the prevalent information gap could indeed be thought of as a viable policy to reduce the democratic deficit. In turn, with the apparent decline in social capital over the past decades, the authors encourage all policies aiming at a systemic (re)generation of social capital in Europe, particularly at the municipal levels. With the assertion that increased levels of social capital, via different channels, can lead to increased levels of political participation, the authors remain optimistic on the viability of the recommended policies to enhance collective action amid European citizens, in an endeavour to enhance the democratic machinery of the EU.

While we remain clear on the suggestive nature of our results, we strongly encourage further investigations to dig deeper and, while relying on statistical analysis for measuring effects is indeed important, it is nonetheless imperative to acknowledge its limitations in capturing the causal effects of real life. For this, we stress the need for complementary qualitative analysis based on observational data and population-wide attitudes towards domestic and European developments.

Although benefiting considerably from its membership, the UK citizenry is known to be largely cynical towards the EU. For this reason, British citizens are entitled to be provided with more information and incentives to increase their say in the functioning of the only directly elected body of the EU and discover how best to reap advantage of its membership. For this, we believe that the UK government should aim to increase the presence and importance of EU issues for subsequent EP elections, especially when considering the citizens’ embedded bias towards thinking that less is at stake than there actually is.

While the EU is yet to become the supranational democratic entity that it has always aspired to be, the next five years present themselves as a window of opportunity for the further democratisation of the Union. The hopes are that the Juncker Commission will abide by its promises and a newly elected EP by the expectations of its electorate. Time will tell whether the EU is able to live up to its aspirations.

Bibliography

Adler, P.S. & Kwon, S. (2002) “Social Capital, Prospects for a New Concept”, Academy of Management Review, 27(1), pp. 17–14

Axford, B. Browning, G.K., Huggins, R., Rosamond, B. & Turner, J. (2002) Politics: An Introduction, 2nd edition, Routledge publications

Baek, M. (2009) “A comparative analysis of Political communication Systems and Voter Turnout” American Journal of Political Science, 53 (2), pp. 376–393

Banducci, S. (2005) “Media Exposure, Media Content, and Voter Mobilization Implications for the Lijphart Thesis” Paper presented at the annual meeting of the American Political Science Association, Marriott Wardman Park, Omni Shoreham, Washington Hilton, Washington, DC, Sep. 1, 2005

Banducci, S. & Semetko H. (2003) “Media and Mobilisation in the 1999 European Parliamentary Election” Europe, Parliament and the Media, pp. 189–204.

Bennie, L. G. (2003) Social Movements: Understanding Democratic Politics: An introduction, Sage Publications Besley, T. 1994. “How do Market Failures Justify Intervention in Rural Credit Markets?” The World Bank

Research Observer, 91(1): 27–48.

Bilska, M. (2012) “Voter turnout in the 2009 European elections: Media coverage and media exposure as explanatory factors”, Paper prepared for presentation at the EPOP 2012 Conference, “European Integration” panel, University of Oxford, 7–9 September 2012

Blais, A. (2000) To vote or not to vote? The merits and limits of rational-choice theory, University of Pittsburgh Press Pittsburgh, PA

Bosancianu, C.M. (2014) “Online mobilisation and offline participation in European Elections”, GeT MA Working Paper №4, Department of Social Sciences, Humboldt-Universität zu Berlin

Campbell, D.T., Stanley, J.C. (1966) Experimental and Quasi-Experimental Designs for Research, Rand McNally. Crombez, C. (2003) “The Democratic Deficit in the European Union: Much ado about Nothing”

European Union Politics, 4(1), pp. 101–120
Cox, G (1999) ‘Electoral Rules and Electoral Competition’ Annual Review of Political Science, Vol. 2, pp. 145-

61
Dalton, R.J. (2004) Democratic Challenges, Democratic Choices: The Erosion of Political Support in Advanced

Industrial Democracies, (New York: Oxford University Press)
Delwit, P. (2013) “The end of voters in Europe? Electoral Turnout in Europe since WWII”, Open Journal

of Political Science, 3(1), pp.44–52
De Vreese, C. H. & H. A. Semetko. (2002) “Cynical and Engaged: Strategic Campaign Coverage, Public

Opinion, and Mobilisation in a Referendum”, Communication Research, 29(6), pp. 615–641.
Di Pasquale, D. & Glaeser, E.L. (1999) “Incentives and Social Capital: Are Homeowners Better

Citizens?” Journal of Urban Economics Vol. 45, pp. 354–384

Engelhardt, G.V., Eriksen, M.D., Gale, W.G. & Mills, G.B. (2010) “What are the social benefits of homeownership? Experimental evidence for low-income households”, Journal of Urban Economics, 67 (3)

Gaiha, R. (2013) “Credit, Microfinance and Empowerment”, Invited contribution to Expert Group Meeting: Policies and Strategies to Promote Empowerment of People in Achieving Poverty Eradication, Social Integration and Full Employment and Decent Work for All, 10 -11 September 2013, United Nations Secretariat Building, New York.

Hix, S. & Marsh, M. (2011) “Second-Order effects plus pan-European political swings: ananalysis of European parliament elections across time”, Electoral Studies, 30 (1), pp.4–15

Hobolt, S.B. & Wittrock, J. (2011) “The second-order election model revisited: An experimental test of vote choices in European Parliament elections”, Electoral Studies, Vol. (30), pp. 29–40

Hvinden, B. & Johansson, H. (2007) Citizenship in Nordic Welfare States, (eds. Routledge)
Inglehart, R. & Catterberg, G. (2002) “Trends in Political Action: The Developmental Trend and the

Post-Honeymoon Decline”, International Journal of Comparative Sociology, Vol. 43, pp. 300–316 International Institute for Democracy and Electoral Assistance (2004) Voter Turnout in Western Europe since

1945, International IDEA publications, ISBN 91–85391–00-X
Lijphart, A. (1997) “Unequal Participation: Democracies Unresolved Dilemma”, American Political Science

Review 91 (1), pp. 1–14.
Malkopoulou, A. (2009) “Lost voters: Participation in EU elections and the case for compulsory voting”,

CEPS Working Document no. 317
McCabe, B.J. (2013) “Are Homeowners Better Citizens? Homeownership and Community Participation

in the United States”, Social Forces, 91(3), pp. 929–954.
McCaffrey M., & Salerno, JT. (2011) “A Theory of Political Entrepreneurship”, Modern Economy, Vol. 2,

pp. 552–560

McCluskey, M.R., Deshpande, S., Shah, D.V. & McLeod, D.M. (2004) “The Efficacy Gap and Political Participation: When Political Influence fails to meet Expectations”, International Journal of Public Opinion Research, 16 (4)

Morduch, J. 1999. “The Microfinance Promise”, Journal of Economic Literature 37 (4), 1569–1614.

Newton, K. (1999) “Mass Media Effects: Mobilisation or Media Malaise?”, British Journal of Political Science, 29(4)

Norris, P & Sanders D., (2003) ‘Message of Medium?Campaign Learning during the 2001 British Election’ Political Communication, Vol. 20, pp. 233–262

Norris P. (2000) A virtuous circle: political communications in post-industrial democracies, Cambridge University Press, Cambridge

Norris, P. (2004). “Global Political Communication. Good Governance, Human Development,
and Mass Communication.” In: Esser, F. & B. Pfetsch, eds. The Comparing Political Communication.

Theories, Cases and Challenges. Cambridge: Cambridge University Press

Olson, M. (1965) The Logic of Collective Action: Public Goods and the Theory of Groups (Revised edition, 1971) Harvard University Press

Peter, J. & C.H. de Vreese (2004) “In search of of Europe: A cross-national Comparative study of the European Union in National Television News”, The Harvard International Journal of Press/Politics, Vol. 9, pp. 3–24

Plutzer, E. (2002) “Becoming a Habitual Voter: Inertia, Resources, and Growth in Young Adulthood.” American Political Science Review, Vol. 96 pp. 41–56.

Portes, A. (2000) ‘The Two Meanings of Social Capital’, Sociological Forum, 15(1) pp. 1–12

Putnam R. (1995) ‘Bowling Alone: America’s Declining Social Capital’, Journal of Democracy 6(1), pp. 65–78

Reif, K., and H. Schmitt (1980) “Second-Order National Elections — a Conceptual Framework for the Analysis of European Election Results” European Journal of Political Research 8 (1) pp. 3–44.

Reif, K. (1984) “National election cycles and European elections, 1979 and 1984”, Electoral Studies, 3(3), pp. 244–255

Rohe, W.M. (2013) “Reexamining the Social Benefits of Homeownership after the Housing Crisis”, Paper originally presented at Homeownership Built to Last: Lessons from the Housing Crisis on Sustaining Homeownership for Low-Income and Minority Families — A National Symposium held on April 1 and 2, 2013 at Harvard Business School in Boston, Massachusetts.

Rosenstone, S & Hansen J. (2002), Mobilization, Participation and Democracy in America, Longman, New York

Stiglitz, J., Emran, S. and Morshed, M. (2006) “Microfinance and missing markets”, CU Working paper.

Swanson, D.L., Mancini, P. (1996) “Patterns of modern electoral campaigning and their consequences”, in Swanson, D.L., Mancini, P. (Eds) Politics, Media, and Modern Democracy: An International Study of Innovations in Electoral Campaigning and Their Consequences, Praeger, London

Tilley, J., Garry, J. & Bold, T., (2008) “Perceptions and Reality: Economic Voting at the 2004 European Parliament Elections” European Journal of Political Research, Vol. 47, pp. 665–686

Van der Brug, Wouter., Van der Eijk, Cees. (2007) European Elections and Domestic Politics: Lessons from the past and scenarios for the future University of Notre Dame Press, Indiana

Vergeer, M., Hermans, L. & Sams, S. (2011) “Is the voter only a tweet away? Micro-blogging during the 2009 European Parliament election campaign in the Netherlands”, First Monday, 6 (8)

Vergeer, M., Hermans, L. & Cunha, C. (2012) “Web campaigning in the 2009 European Parliament elections: A cross-national comparative analysis”, New Media Society, 15 (1) pp. 128–148

Reports:

1. European Commission (2007) The Regulation of Microcredit in Europe, Expert group report 2. Eurobarometer Reports — # 62, 162, 320, 375, 318

Online Resources

1. House of Commons Library, Research paper 04/50 http://www.parliament.uk/commons/lib/research/rp2004/rp04-050.pdf, June 23rd 2004
2. UK Electoral Commission — http://www.electoralcommission.org.uk/faq/elections/what-was- the-turnout-at-recent-general-elections
3. Eurobarometer 62, Public Opinion in the European Union — National Report on the United Kingdom, Autumn 2004, ec.europa.eu/publicopinion/archives/eb/eb62/eb62_en.html
4. European Electoral Studies (EES) — Study 2004 http://www.ees-homepage.net
5. United Nations Education, Scientific and Cultural Organization (2012) Statistics on Youth
6. The Economist