A Red Letter Day: Investigating the Renaissance of the French Far Left in the Presidential Election

Transcription

1 A Red Letter Day: Investigating the Renaissance of the French Far Left in the 2012 Presidential Election Paper prepared for presentation at the 2014 MPSA Conference, Chicago, IL, 3-6 April. Pavlos Vasilopoulos, Centre de Recherches Politiques (CEVIPOF), Sciences Po Paris. Laurie Beaudonnet, University of Montreal, Bruno Cautrès CNRS - Centre de Recherches Politiques (CEVIPOF), Sciences Po Paris. Bruno.cautres@sciencespo.fr 1

2 Introduction In the first round of the 2012 French Presidential election, the candidate of the far-left Front de Gauche (FDG), Jean-Luc Mélenchon, managed to gain an impressive 11.1 percent of the vote. This was the highest far-left vote share in French presidential elections for over 30 years and the impressive electoral performance of communist candidate Georges Marchais in the 1981 presidential election, who then gathered 15.3 percent of the vote. This FDG success epitomized an upward tendency that started after the 2009 European Election, the party s first electoral bid since its creation in 2008, and as vote intention polls indicate are likely to continue in the 2014 French municipal and European elections. Yet, even though there have been extensive studies on the electoral support of the French far right (e.g. Hainsworth 2004; Mayer 2013; Schields 2013), research on the resurgence of the French far left remains scarce (with the notable exception of Mayer 2007; Sperber 2010). For the case of FDG, Cautrès (2013) conducted a sociodemographic analysis of the party s electoral basis at the aggregate level, but no study so far has investigated the newly-shaped French far left voters attitudinal profile. This article aims to fill this gap by investigating the sociodemographic and ideological profile of the French far left voter in the 2012 Presidential Elections using individual level data. To this direction, we use a threefold explanatory framework focusing on a) sociodemographic and political value determinants of the French far left vote b) the role of attitudes toward Europe and c) the impact of the economic crisis on the success of Mélenchon. At the same time we are interested in highlighting possible similarities of the newly formed FDG electorate with voters of other European far-left parties. Overall, our results illustrate that the reason behind the FDG electoral success in the 2012 presidential election can be summarized in three points: First, in its ability to secure the vote of both segments of the traditional far left and postmaterialist electoral pools. Second, by becoming the voice of the citizens who while not against the idea of the European Union, are 2

3 critical of its current ideological direction. Finally a third factor was the deterioration of voters economic situation, which had a modest yet significant effect in the FDG electoral rise. All in all, our results point to the conclusion that the French far left voter profile appears somehow divergent from the respective profile of other European far-left voters. The rest of the article is organized as follows: In the next section, we discuss the Front de Gauche programmatic characteristics and campaign strategy in the context of the 2012 Presidential election. Next, we investigate the specific demographic and attitudinal profile of the newly formed French far-left voter vis-à-vis other electorates. Based on these findings, we draw some conclusions about the transformation of the French electoral competition and put the Front de Gauche s success in perspective regarding the next elections and the rise of the new far left in European party systems. The Front de Gauche in the 2012 presidential elections: Merging the old and the new left The Front de Gauche (Left Front, FDG) was created in 2009 as a united front on the far left of the French political landscape. The 2005 referendum on the European Constitutional Treaty divided the French left and gave an increased importance to the leftwing critic of the European project. The FDG is a product of this opposition to a neo-liberal Europe perceived by voters as a threat to the French social model. In order to build on this dynamic for the 2009 European elections, the Left front to Change Europe, FDG, was created as a coalition of the Communist Party (Parti Communiste Francais, PCF), the Left Party (Parti de Gauche, PG) and several other [political ideology] political groups such as République et Socialisme, FASE, Gauche Unitaire (United Left) with the clear goal to unify the far left and to present a left-wing alternative to the current European project. 3

4 The 2012 FDG manifesto, L humain d abord (Human being comes first) 1, which the party and its candidate Jean-Luc Mélenchon used for the presidential and legislative campaigns, perfectly illustrates this mix of old and new left issues and values. Four topics are particularly prominent in the party s manifesto: redistribution, opposition to financial markets, the environment, and institutional change (with the creation of a 6 th republic). On the one hand, the two first themes belong to the traditional economic dimension of the left-right cleavage in France (Tiberj 2013), with measures such as increasing taxes for the wealthiest segments of the society, expanding social protection, opposing financial markets, favoring state taking over banks. This is typical of the anti-capitalist stance of PCF. On the other hand, the third and fourth themes are typical of the new politics. Several elements of FDG came from the Green party (the PG s motto for instance is Écologie, Socialisme, République ) and support environmental protection, alternative energies and anti-growth measures. However, the party s stance is somewhere in between: it does not explicitly position itself against nuclear energy (but call for a referendum) and advocate an alternative way of producing, based on new technologies and energies, re-industrialization and training. Indeed, these political ideas stand out as a merging of the traditional communist stances in favor of nuclear energy (in defense of manual worker jobs) and a re-industrialization of France, and new left stances (from PG and other entities) that clearly have a post-materialist dimension (in favor of sustainable development, anti-growth movement and alternative energy). In that regard, the expression used in the manifesto La planification écologique (Ecological planning, with planning resonating with economic planning in the USSR) perfectly summarize this mix of communist and post-materialist values that FDG aims at representing. Finally, the call for a new political regime, based on participative democracy and increased powers of the Parliament, stands out as a post-materialist claim for more transparent and representative 1 The manifesto can be accessed at 4

5 institutions. 2 In the 2012 Presidential campaign, J.L. Mélenchon thus positioned himself on the left of the Socialist candidate, François Hollande. His aim was to capture votes from the two leftist parties, Lutte Ouvrière (LO) and the Nouveau Parti Anticapitaliste (NPA), by using the typical anti-capitalist and Marxist terminology of the Communist party. At the same time he aimed both at the post-materialist votes, concerned with the environment (the typical electorate of Europe Écologie - Les Verts, the Green party, EELV) and sustainable development, as well as the disappointed Socialist voters, who are in favor of more radical solutions to the economic crisis and the fight against inequalities. The content of FDG s TV ads for the legislative campaign highlights this specific combination of issues 3 : Regarding the traditional left-wing issues, the dominant theme in FDG claims is social justice, it concentrates a third of the observations (a number that goes up to more than 45% if we add references to the welfare state). The economy comes second and represents 21% of the observations. Among these quasi-sentences, about 70% belong to Marxist rhetoric or call for a controlled economy. Of course this is not as much as LO and NPA, for whom Marxist quasi-sentences are three time more numerous, but FDG is the only other party who uses this rhetoric, thus directly targeting the traditional extreme left electorate, voters who usually choose LO and NPA. Accordingly, we expect voting for Jean- Luc Mélenchon, the FDG candidate in the presidential race, to be positively associated with preferences for redistribution. Regarding the post-materialist claim about the environment and anti-growth, without surprise, the party for whom the issue is the most present is EELV, followed by PS and by FDG, for whom this dimension is the fourth one in size (representing 2 Source : Programme du front de gauche, l Humain d abord (2011), (last accessed on February 25th 2014). 3 Source : French TV ads dataset, Legislative elections, Making Electoral Democracy Work Project. Data consists in quasi-sentences coded using the Comparative Manifesto coding (the original material consists in the 8 official TV ads issued by FDG during the 2012 official legislative campaign). As the analysis presented in this article is centered on the presidential elections, it would have been better to rely on TV ads for this specific campaign. Unfortunately such data are not available. Additionally, FDG conveyed the same message and used the same program for both elections, making the legislative campaign informative of its overall strategy. 5

6 about 5% of the observations). Consequently, we expect FDG electorate to be more postmaterialist culturally liberal than other electorates (with the exception of Green voters). The third most covered issue in FDG s strategy is the European Union. It represents 10% of the observations in the data previously described. Since the EU issue was increasingly salient during the 2012 French presidential campaign, we believe it played a key role in the rise of the French far left. It has been argued that far left parties are generally Eurosceptic, a term that encompasses both those who want the EU to dissolve and those who want to see it change and is typically measured by asking respondents whether their country s membership in the EU is a good or a bad thing. (e.g. March and Rommerskirchen 2012). Yet, it has been argued that this is an overgeneralization that has little explanatory value for tapping the multidimensionality of attitudes toward European integration, especially in France (Cautrès 2012). While LO and NPA, who opposes the European Union as an agent of globalization and neo-liberalism, can easily be said to be hard Eurosceptics, the case of FDG is not that simple and the content of TV ads shows this in a very significant way: On the one hand, the party is very vocal in its opposition to the EU in its current state, denouncing the influence of financial markets and the importance given to Monetary and Economic Union at the expense of social policies and political union, opposing the independence of the European Central Bank and austerity measures. On the other hand, it clearly calls for an expanded European social model, based on harmonized rights and expanded European public services, and is in favor of increasing the power of the European Parliament. For instance, after PS, FDG is the party that covers the European issue the most (concentrating 20% of the statements, against 35% for PS, almost 7% for the centrist MoDem and 10% for the right-wing UMP). But 90% of FDG statements about Europe are negative, focusing on denouncing neo-liberal Europe. This is a strong difference with PS, for whom negative statements only represents 30%, but also with LO and NPA, which statements about Europe are only negative, and surprisingly 6

7 with EELV, who do not mention Europe at all. FDG s position thus could be qualified as soft Eurosceptic (Taggart 1998): opposing the current policies or institutional features of the European Union, while support European integration as project, or as what Wessels defined as Critical Europeans (2007): supporting European integration but advocating for a different political orientation. This specific position goes in line with the initial goal of FDG (offering a left-wing alternative for the European project) and its intention of capitalizing on the rejection of the European Constitutional Treaty, and is clearly different from hard Euroscepticism as expressed by the extreme-right party, the Front national,(national Front, FN). Several studies have showed that citizens are becoming increasingly conscious of the social and economic consequences of European integration and that many of them do not simply support / oppose the EU anymore, but support / oppose a specific political orientation of the European project, with the main division being between a social- v. an economicoriented Europe. This logic is particularly strong in protective welfare states like France (Brinegar and Jolly 2005; Cautrès 2005 and 2012; Dalem 2005; Beaudonnet 2013). By vocally positioning itself on this issue, using the rhetoric of resistance to neo-liberal Europe and calling for increased power of the EU in order not only to protect but diffuse the French social protection system, FDG is very likely to attract the Euro-critical segment of the left electorate, for whom the Euro-enthusiast PS and EELV are too complaisant with the limits of the current system. Indeed, it has been showed that EELV in particular lost these votes in the 2012 presidential elections, to the advantage of FDG (Beaudonnet and Vasilopoulos 2014). Consequently, we expect a two-way relationship between voting for FDG and attitudes towards Europe: on the one hand, voting for FDG is likely to be positively correlated with having voted No in the 2005 referendum and being critical of the Common currency, the Euro. This critical stance about European integration is likely to strongly differentiate FDG voters from PS and EELV voters. On the other hand, we expect FDG voters to be supportive 7

8 of European integration to some extent, and to be on this dimension radically different from FN voters. An additional factor that contributes to the explanation of the rise of the new French far left is the impact of the European crisis on the national economy. The fact that the deterioration of the national economy affects negatively the incumbent and increases support for the challenger is one of the most well established facts in political behavior literature, having been tested across countries over extended periods of time (e.g. Anderson 2000; Lewis-Beck and Paldam 2000; Lewis-Beck and Stegmeyer 2007). The basic distinction in economic voting theories is between sociotropic and egocentric economic voting. Sociotropic economic voting taps evaluations on the past or future state of the national economy. Egocentric economic voting on the other hand taps evaluations on respondents personal financial situation. Sociotropic economic voting has been found to have a higher explanatory power in voting choices (Anderson 2000) a finding that Lewis-Beck also highlights for the case of France (1997). However, research on the impact of economic evaluations on far left parties remains scarce. In one of the few studies that tap the relation between far left voting and economic conditions, Bowyer and Vail (2011) show a significant effect of retrospective sociotropic economic considerations on the probability voting for the German far left party, Die Linke. More recently, two studies further illustrated an association between other indexes of economic hardship and far and radical left voting. March and Rommerskirchen s aggregate-level analysis of 39 far and radical left parties illustrates an association between levels of unemployment and radical left voting (2012). At the individual level, Visser et al. showed that there is higher probability of endorsing far left ideology among low income layers and the unemployed (2014). In the light of the above, we are particularly interested to assess the extent to which the deterioration of economic conditions in France has been pivotal in the rise of the FDG, especially vis-à-vis voters of the other anti-systemic pole in the French 8

9 political system, the far right Front National, which also appears to gather support among those who hurt the most from economic distress (Mayer 2011; 2013). We thus expect voting for FDG to be positively correlated with either being a manual worker or unemployed, and with a negative evaluation of the economy (whether this evaluation is sociotropic or egocentric). Data and Measurement Data come from the CEVIPOF 2012 Election Study, a face-to-face survey that was conducted on a nation-wide representative sample (N= 2504) one week after the second round of the Presidential election. In order to analyze the demographic and attitudinal profile of the French far left voter we use multinomial logistic regression, a method suitable for treating categorical dependent variables with more than two categories and that has been particularly deemed appropriate for the study of the French voter (Dow and Endersby 2004). We hence assess the impact of a series of variables on the likelihood of casting a vote for Jean-Luc Mélenchon against voting for Nicolas Sarkozy (Conservative- UMP), Francois Hollande (Socialist- PS), Marine Le Pen (Extreme right- FN), Francois Bayrou (Centrist- MoDem), Eva Joly (Green- EELV) 4. We construct two models: Model A controls for demographic characteristics (age, gender, education) and the variables lourdes (occupation, income, church attendance, and being urban). These variables have been repeatedly described as the long term forces of the vote in France, stemming from macro-historical cleavages (Boy and Mayer 1993) and can be safely considered as causally prior to attitudinal variables. Due to sample limitations, 4 For the sake of parcimony, abstainers and votes for other candidates were excluded of the analysis.other candidates were Philippe Poutou (Nouveau Parti Anticapitaliste, NPA), Arthaud (Lutte Ouvrière, LO), Jacques Cheminade (Solidarité & Progrès, SP), and Nicolas Dupont-Aignan (Debout la République, DR). Results including Abtainers and one answer category for these other candidates did not yield different results. Available on request. 9

10 occupation has been recoded in two categories: manual workers and unemployed 5. Investigating the impact of these variables allows us to determine how well the newly shaped French far left voter fits to the socio-demographic profile of the traditional French communist left as well as the new European far left. In the second model (B) we introduce a number of ideological and attitudinal items. In particular we tap economic left-right predispositions through a four-point item measuring attitudes toward income redistribution ( To establish social justice, it should be taken from the rich and given to the poor ). Furthermore, we tap libertarian-authoritarian values through a four-point item measuring attitudes toward death penalty in France ( Death penalty should be restored in France ). What is more, the model accounts for postmaterialist attitudes, to investigate whether the new French far-left electorate sets priority on the improvement of material well-being or quality of life 6. We tap attitudes toward the European Union through three items: The first is the typical item used in public opinion surveys to measure general support for EU membership ('Is France s membership to the EU a good or a bad thing'). Yet, as we mentioned earlier we have reasons to expect that citizens who cast a vote for Mélenchon are not against the idea of Europe but rather against its current political direction (the soft euroscepticism hypothesis). Hence we explore a possible multidimensionality of Mélenchon`s voters attitudes toward the EU with two additional items. The first taps voting behavior in the 2005 European Constitutional Treaty referendum, separating between those who voted for, those who voted against and respondents who abstained or were too young at the time to vote. This item has previously been found to have explanatory power in the electoral behavior of French voters (Beaudonnet and Vasilopoulos 2013). The second item accounts for attitudes toward the euro, by asking respondents whether 5 The analysis was repeated using only the two major candidates and Marine Le Pen, illustrating that respondents occupation is not a statistically significant predictor of the FDG vote, with the exception of the occupational categories included in the analyses. 6 Although the question wording is not identical to the original item used by Inglehart (1990), we believe that this question adequately taps the value conflict between personal economic progress and quality of life, which is central to the materialist/post-materialist cleavage (e.g. Dalton 2006, Inglehart 2008). 10

11 they 'feel that we are stronger economically because France is part of the Eurozone'. This item was chosen to capture the most salient dimension of European integration in these times of crisis: the economic one. Additionally, ideological self-positioning is measured with the leftright self-positioning scale, recoded in five positions ranging from extreme left to extreme right. Finally, we examine the impact of economic distress by drawing on economic voting theory. Economic voting was measured by the typical retrospective egocentric and sociotropic evaluation items (e.g. Anderson 2000; Lewis-Beck and Stegmaier 2007) asking respondents to assess whether a) their personal financial situation and b) the situation of the French economy has improved or deteriorated during the last 12 months 7. Exact wordings and codification details for all variables are provided in the Appendix. 7 Unfortunately our model cannot assess the impact of prospective economic evaluations on the vote as the fact that the survey was carried out after respondents were aware that Hollande was France s new president created a strong reverse causation effect. 11

12 Results Table 1 reports the results of the sociodemographic profile of the far-left voter in the first round of the 2012 French Presidential election. We report relative risk ratios (RRR), meaning that values lower than 1 indicate an increase in the probability of voting for Mélenchon while RRR higher than 1 indicate an increase in the probability of voting for the other candidate. Overall, and unlike available comparative evidence (Visser et al. 2014) that show that far left ideology is disproportionately endorsed by low-income layers, manual workers and the unemployed, our results point to a rather mixed demographic profile of the French far left voter in the 2012 Presidential election. Starting with age, Mélenchon voters appear to be younger than voters of Sarkozy and Hollande (the candidates of the two dominant French parties) as well as voters of centrist candidate Bayrou. This is to in opposition with the results of aggregated analysis of the FDG vote, which showed that the departements (territorial units) where FDG was strong were characterized by older population (Cautrès 2013). Regarding the impact of education, far left voters appear more educated than voters who cast a vote for the two right wing candidates (who traditionally belong to lower educational strata). On the contrary, highly educated people vote more for centrist François Bayrou and Green party candidate Eva Joly (although the difference here is only significant at the 0.1 level). As for gender, Mélenchon voters appear to be predominately male when compared to Sarkozy and Hollande voters but gender does not seem to play a role when compared with the other candidates of the electoral bases. Moving on to the investigation of the impact of the variables lourdes, church attendance has the highest explanatory power, as French far left voters tend to attend church significantly less than voters of all other candidates except Hollande s and Joly s. Location has the expected effect: living in a big city does not make any difference for Melenchon, Hollande, Joly and Sarkozy voters, but does for Le Pen and Bayrou. Indeed both candidates typically seek to attract rural segments of the 12

13 society. Surprisingly, income is not a strong predictor of far-left voting, apart from differences between Bayrou and the traditionally wealthy French center right voters (Cautrès 2003, Lewis-Beck et al. 2012), no significant relation is found in the rest of the comparisons. Finally, it is worth noting that being a manual worker or unemployment does not exert a significant influence on the French far left vote in any of the comparisons with the exception of Marine Le Pen. Manual workers and the unemployed, the typical electoral basis of the French communist party, appear to be more likely to vote for Le Pen instead of Melenchon, however the relationship is only significant at the 0.1 level and should be interpreted with caution Two preliminary conclusions can be drawn from the sociodemographic analysis of the Mélenchon vote: First, results suggest that Mélenchon voters do not reflect the typical characteristics of traditional communist left voters. Indeed, since they are poorer than right wing voters and do not attend church, FDG voters are more educated than typical Communist voters and manual workers are not over-represented (this going in line with the attraction of workers for FN over the past decade) (Mayer 2011). What is more, they only partly resemble the sociodemographic characteristics of the economically insecure European far left voter. The most important result of this analysis is that there is no statistically significant difference between PS, EELV and FDG electorates, from a socio-demographic point of view (with the exception of women and older voters that are more likely to vote for the Socialists). This echoes the results of the aggregate analysis of FDG vote (Cautrès 2013): Jean-Luc Mélenchon was successful in the tradidional territories of PS and EELV: the three parties share the same electorate. Despite its radical stance and opposition to the Centre-left position of the PS candidate, Francois Hollande, FDG is positioned de facto, by its electorate, within the governmental left (`gauche de gouvernement`). {TABLE 1 ABOUT HERE} 13

14 Table 2 presents the results of the full model assessing the impact of both demographic and attitudinal items. Overall, results further illustrate that the French far-left voter has a distinct ideological profile that does not correspond to the traditional left voter tradition. Mélenchon voters appear in favor of vigorous redistribution policies compared to Sarkozy (RRR= 0.67) and Bayrou voters (RRR= 0.76), but differences are statistically indiscernible when compared to voters of all other candidates.. Furthermore, and rather unsurprisingly, French far Left voters appear more libertarian compared to the two Right wing candidates (RRR= 1.61 when compared to Sarkozy and 2.08 when compared to Le Pen) and Bayrou (RRR= 1.40). However, when compared to Hollande, and Joly voters the respective coefficients fall short of achieving statistical significance. On the other hand, it is far Left voters strong postmaterialist attitudes that decidedly distinguish them from the rest of the electorate, with the exception of traditionally postmaterialist Green voters (Beaudonnet and Vasilopoulos 2014). Taken together, these three items point to an unusual attitudinal voter profile that stands away from traditional left values for income redistribution focusing more on issues of self-expression and libertarianism. Moving on to the impact of Europe on French the far-left vote, results suggest that Mélenchon voters have a distinct profile when it comes to attitudes toward the European Union: there are no differences in support for France s membership to the EU between Mélenchon voters and voters of the rest of the candidates with the exceptions a) of the Far Right electorate (RRR= 0.45), which is traditionally hard Eurosceptic and b) of the incumbent Sarkozy whose electorate appears more Europhile. What separates far left voters from voters of the two mainstream and the centrist candidates is the economic dimension of support for the EU: Voters of Hollande, Sarkozy and Bayrou are significantly more likely to suggest that they have benefitted economically from France s membership to the Eurozone (RRR= 1.64; 1.93 and 2.01 respectively), while respective differences between Mélenchon, 14

15 Le Pen and Joly voters are not statistically significant. [This finding can be attributed to ].What is more, a striking difference that separates French far left voters from the rest of the French electorate is voting behavior in the 2005 referendum. Mélenchon voters are far more likely to have voted against the European Constitutional Treaty. This finding is not surprising since as we mentioned earlier it is in alignment with the party s history and its particular soft Eurosceptic criticism on the EU. Overall, results confirm our theoretical expectations regarding the European dimension in the French far left vote: Findings based on all three items we measured to tap Euroscepticism among the French far left electorate point to the existence of a soft Eurosceptic electorate that unlike the traditional French communist and other European radical left electorates, is not against the idea of a united Europe, not less favorable toward their country s membership in the EU compared to voters of other parties that are considered Europhile parties, but it is rather critical of EU s current direction, as it is underlined by their vote in the ECT referendum and their attitudes toward the common currency. Moving on to the investigation of the economic crisis on the far left vote, we find mixed evidence. On the one hand, we find that sociotropic economic evaluations have a marginally significant effect in the expected direction when comparing far left voters to those who cast a vote to incumbent Sarkozy (RRR= 0.59, p < 0.05). However, evaluations of the state of the French national economy during the past 12 months do not appear to exert a significant influence for voting the far left candidate in any of the rest of comparisons. Evaluations on personal financial situation on the other hand while do not have a statistically significant effect between Mélenchon and Sarkozy, appear to be marginally statistically significant when we compare far left with Green voters (RRR= 0.44, p < 0.05), while the effect is most striking when we contrast voters of the two extremes of the French political 15

16 system: All else equal, the deterioration of personal economic situation makes voters 89 percent more likely to cast a vote for Marine Le Pen compared to Mélenchon. Finally, it is worth noting that Mélenchon voters do not place themselves on the left end of the left right scale. While appearing more left-wing than voters who cast a ballot to the center and right candidates, far left voters do not place themselves to the left of Green and Hollande voters. Conclusion This paper illustrated that the increased electoral gains of the French Far Left party in the 2012 Presidential Election were a product of three main factors: (i) Its increased appeal among heterogeneous social layers that combined different segments of the French left electorate, (ii) its ability to become the voice of a politically moderate electorate which while not against the idea of European Integration is becoming increasingly critical of the current vision for the EU and (iii) the deterioration of France s economy. These three factors point to the conclusion that the rise of the new French far left is not a temporal phenomenon, but on the contrary pave the ground for future successes for the French far left. As for the first factor, the fact that both FDG and the PS electorates do not differ in terms of ideological identification and attitudes toward the economy gives the FDG a major advantage ahead of upcoming elections. As the French socialist government appears to respond positively to EU s calls for increased austerity combined with pro-business economic measures in to increase the economy s competitiveness, the FDG has only to gain by attracting the vote of the increasingly dissatisfied PS voters. At the same time, EU s pressure on the French government increases the French left voters alienation toward EU s leadership and at the same time makes the question of the direction of the Union increasingly salient in the minds of French electors. According to our findings this is likely to further boost the FDG 16

17 electoral share, as Melenchon s party bases its support on a significant mass of French voters who while in favor of France s EU membership are becoming increasingly dissatisfied with the ideological direction the Union is taking. Finally, the fact that voters who thought that France s national finances had deteriorated during the past year were equally likely to vote for Hollande and the rest of French minor opposition parties (including FDG) compared to Sarkozy illustrates that the poor state of the French economy under the economic crisis is likely to bring further future electoral gains for Melenchon s party, especially if we consider that the deterioration of personal finances appear to favor the FDG compared to the ideologically proximate French Green party. Viewed in a comparative perspective, our findings extend the still underdeveloped literature regarding the newly formed European far left voter. Two conclusions can be extracted in this regard. First, our results only partly confirm previous analyses that found a link on economic hardship and far left voting. For example, in contrast with their German counterparts (Bowyer and Vail 2011), the French far left does not attract more votes among those who have been hurt most by the crisis such as manual workers or the unemployed, but rather its electoral support cuts across these segments. All in all our results show that for the case of France economic hardship appears to be one of several dimensions of the far left vote. Variables such as postmaterialism and attitudes toward the European Union had a stronger impact on the Far Left vote compared to the traditional explanatory framework focusing on economic hardship and even economic left-right issues. A second conclusion concerns the rise of Eurosceptic parties under the current economic crisis. As our analysis indicated disregarding far left parties as Eurosceptic is an overly simplified taxonomy that fails to tap the multidimensionality of critical attitudes toward Europe. Our analysis indicated that once we control for variables tapping opposition with the current direction of the EU, French far left voters do not generally appear to disagree with their country s membership in the Union 17

18 compared to other parties. What instead vastly differentiates them is voting behavior in the 2005 referendum and their attitudes toward the Euro. In this context our analysis offers additional evidence confirming the distinction between soft and hard Euroscepticism and underlines the need to add additional measures of critical attitudes toward the EU so as to be able to measure the heterogeneity of critical attitudes toward the EU under the current economic crisis. 18

19 References Anderson, Cristopher J. (2000). Economic Voting and Political Context: A Comparative Perspective, Electoral Studies, 19 (2-3): Beaudonnet, Laurie (2013). Preferences for European Social Policy in Times of Crisis. Politique Européenne, numéro 42. Beaudonnet, Laurie and Pavlos Vasilopoulos. (2014). Green Parties in Hard Times: The Case of the EELV in the 2012 French Presidential Election, Party Politics, 20 (2), Bowyer, Benjamin T., and Mark I. Vail. (2011). "Economic insecurity, the social market economy, and support for the German left." West European Politics 34, (4): Boy, D. and Mayer, N. (1993). The changing French voter. In Boy, D. and Mayer, N., eds., The French Voter Decides, Ann Arbor: University of Michigan Press, pp Brinegar, Adam P., Seth Jolly (2005). Location, Location, Location: National Contextual Factors and Public Support for European Integration. European Union Politics 6(2): Cautrès, Bruno (2004). Old wine in new bottles? New wine in old Bottles? Class, religion and vote in the French electorate. The 2002 elections in time perspective, in Michael S. Lewis- Beck, ed. The French Voter: Before and After the 2002 Elections. Basingstoke: Palgrave Macmillan, Cautrès, B. (2005). Les clivages socio-politiques sur l'intégration européenne et le vote du 29 mai Cahiers du CEVIPOF, 42, Cautrès, B. (2012). Multidimensionality of EU attitudes in France: An issue for the understanding of the politicization of attitudes towards the EU. European Integration online Papers (EIoP), 16(2). Cautrès, B. (2013). Le vote Mélenchon : le poids de l héritage communiste et les limites d une dynamique." In Pascal Perrineau (ed.) Le Vote Normal : Les élections présidentielle et législatives d avril-mai-juin 2012,, Paris: Presses de Sciences Po Dalem, Alexis. (2005). L économie, le peuple et la Constitution. Le «non» français en perspective. Raisons politiques 20(4): 73. Dow, Jay K., and James W. Endersby. (2004). "Multinomial probit and multinomial logit: a comparison of choice models for voting research." Electoral studies 23 (1) : Lewis-Beck, M.S., Nadeau, R. and E. Belanger. (2012). French Presidential Elections. New York: Macmillan. Lewis-Beck, Michael S., and Martin Paldam. (2000). "Economic voting: an introduction." Electoral studies 19 (2):

20 Lewis-Beck, Michael S. and Mary Stegmaier. (2007). Economic Models of Voting. In Dalton, Russel J. and Hans-Dieter Klingemann, eds., The Oxford Handbook of Political Behavior. Oxford University Press, Oxford, March, Luke and Charlotte Rommerskirchen. (2012). Out of Left Field? Explaining the Variable Electoral Success of European Radical Left Parties, Party Politics, forthcoming. Mayer, Nonna. (2011). "Why Extremes Don't Meet: Le Pen and Besancenot Voters in the 2007 French Presidential Election." French Politics, Culture & Society 29 (3): Mayer, Nonna. (2013). "From Jean-Marie to Marine Le Pen: Electoral Change on the Far Right." Parliamentary Affairs 66 (1): Sauger, N., Brouard, S., & Grossman, E. (2007). Les Français contre l'europe? Les sens du référendum du 29 mai Shields, James. (2013). "Marine Le Pen and the New FN: A Change of Style or of Substance?." Parliamentary Affairs 66 (1): Sperber, Nathan. (2010) Three Million Trotskyists? Explaining Extreme Left Voring in France in the 2002 Presidential Election, European Journal of Political Research, 49: Taggart, Paul. (1998). A Touchstone of Dissent: Euroscepticism in Contemporary Western European Party Systems. European Journal of Political Research 33(3): Tiberj, V. (2013). Values and the Votes from Mitterrand to Hollande: The Rise of the Twoaxis Politics. Parliamentary Affairs, 66(1), Visser, Mark, Marcel Lubbers, Gerbert Kraaykamp, and Eva Jaspers. (2014) "Support for radical left ideologies in Europe." European Journal of Political Research (forthcoming). Weßels, Bernhard. (2007.) Discontent and European Identity: Three Types of Euroscepticism. Acta Politica 42(2-3):

21 Appendix Table 3: List of Independent Variables Variable Type Wording Coding Age Continuous Gender Dummy 0: Male 1: Female Urbanity Dummy 0: Rural / semi rural 1: Urban Education Dummy 0: Up to Baccalauréat. 1:University attendance or higher Income Ordinal Income quartiles Church Attendance Ordinal Death penalty Ordinal Death penalty should be 1: Totally disagree restored in France 2: Disagree 3: Agree Income redistribution Ordinal To establish social justice, it should be taken from the rich and given to the poor Post-materialism Dummy What do you prefer: Work more and earn more, or mork less and earn less? EU membership evaluation Voting behavior at the 2005 ECT Referendum Dummy Categorical Do you think that France s membership to the EU is a good, neither good nor bad, or a bad thing? Euro attitudes Dummy I feel that we are stronger economically because France is part of the Eurozone. Ideological identification Retrospective egocentric economic evaluations Retrospective sociotropic economic evaluations Dummy On a scale from 0 to 10, where 0 is the left end, and 10 the right end, where do you stand? Dummy Dummy Over the past 12 months, do you think that your financial situation has improved, worsened, stayed about the same? Over the past 12 months, have you felt that your personal financial situation has 4: Totally agree 1: Totally disagree 2: Disagree 3: Agree 4: Totally agree 1: More leisure time over higher earnings. 0: Higher earnings over more leisure time 1: Good thing 0: Neither good nor bad, or a bad thing 0: voted yes 1: don t know, abstained, was too young to vote 2: voted no 0: Disagree 1: Agree 1: for positions 0 to 4 in the 10-point left-right scale 0: for other positions 1: Stayed about the same / improved 0: Got worse 0:Stayed about the same /improved 1: Got worse 21

22 Table 1 : Sociodemographic determinants of the Mélenchon vote against voting for other candidates and abstaining. Hollande Sarkozy Le_Pen Joly Bayrou Age 1.491*** 1.617*** *** (0.120) (0.140) (0.097) (0.179) (0.162) Gender 1.765*** 1.525** (0.263) (0.244) (0.183) (0.220) (0.242) Higher education ** 0.321*** * (0.167) (0.113) (0.076) (0.703) (0.347) Manual workers (0.247) (0.192) (0.344) (0.516) (0.283) Unemployed (0.367) (0.472) (0.563) (0.866) (0.419) Income *** *** (0.086) (0.123) (0.098) (0.169) (0.157) Attends church *** 1.593* *** (0.192) (0.510) (0.336) (0.594) (0.610) Urban ** * (0.158) (0.128) (0.101) (0.621) (0.127) Constant 1.837*** 1.382* *** 0.486*** (0.268) (0.214) (0.219) (0.033) (0.095) Pseudo-R Observations 1,954 Dependent variable : Vote choice in the first round of the 2012 French Presidential election. Entries are Relative Risk Ratios (RRR), RRR standard errors in parentheses - *** p<0.001, ** p<0.01, * p<

23 Table 2: Sociodemographic and attitudinal determinants of the Mélenchon vote against voting for other candidates and abstaining. Hollande Sarkozy Le_Pen Joly Bayrou Age 1.644*** 1.822*** 1.343* *** (0.181) (0.239) (0.192) (0.235) (0.238) Gender 1.797*** 2.253*** (0.307) (0.480) (0.234) (0.287) (0.330) Higher education * (0.186) (0.172) (0.195) (0.439) (0.453) Manual workers (0.263) (0.203) (0.340) (0.521) (0.326) Unemployed (0.432) (0.546) (0.721) (1.027) (0.517) Income * * (0.093) (0.143) (0.133) (0.192) (0.163) Attends church * * (0.173) (0.399) (0.344) (0.661) (0.473) Urban * ** (0.190) (0.142) (0.128) (0.571) (0.127) EU membership: good * 0.512** (0.231) (0.389) (0.131) (0.513) (0.302) The Euro has benefitted us 1.604** 1.724* * (0.292) (0.393) (0.217) (0.647) (0.457) 2005 ECT referendum : abstained 0.432** 0.434** *** (0.110) (0.129) (0.261) (0.244) (0.109) 2005 ECT referendum : Voted No 0.214*** 0.147*** 0.432* 0.223** 0.187*** (0.052) (0.046) (0.143) (0.121) (0.060) Income redistribution: in favor *** * (0.111) (0.074) (0.109) (0.383) (0.095) Death penalty: in favor *** 2.135*** ** (0.106) (0.190) (0.254) (0.212) (0.182) Post-materialist 0.555** 0.198*** 0.309*** 2.436* 0.437** (0.101) (0.055) (0.090) (0.971) (0.115) France s economy : worse * (0.203) (0.139) (0.213) (0.270) (0.240) Personal finances : worse ** 0.438* (0.162) (0.226) (0.446) (0.184) (0.201) Left-Right scale *** 6.932*** *** (0.126) (1.349) (1.040) (0.281) (0.470) Constant 4.541*** 3.360** 2.812* 0.175* (1.534) (1.371) (1.231) (0.133) (0.867) Pseudo-R Observations 1784 Dependent variable : Vote choice in the first round of the 2012 French Presidential election. Entries are Relative Risk Ratios (RRR), RRR standard errors in parentheses - *** p<0.001, ** p<0.01, * p<