France 2012: PS open primaries 2011 (first round)

The first round of left-wing open primary elections (primaires citoyennes) were held in France on October 9, 2011. These primaries will nominate the candidate of the opposition Socialist Party (PS) and its small ally, the Left Radicals (PRG) for the 2012 presidential elections. All French citizens were eligible to vote provided they pay a symbolic minimum fee of €1 and sign a declaration of vague left-wing values. These ambitious primaries, the first on such a large scale in France, were truly an historic first for the PS and for French politics in general. I had talked about the primaries and the six candidates in a preview post ahead of the first round.

The first round of the primaries were pretty much a success. 2,661,284 voters came out to vote. Using data from the 2010 regional elections (slightly inaccurate as these primaries also included New Caledonia, French Polynesia, Saint-Pierre-et-Miquelon and French citizens abroad as well as under-18 members of the PS or PRG’s youth wings), roughly 6.1% of registered voters (2010 numbers) came out to vote. For comparison’s sake, the PS’ candidate in the 2007 runoff against President Nicolas Sarkozy had won nearly 16.8 million votes. In the 2009 European elections, in which the PS hit rock-bottom and won only its core voters, its lists had received 2.8 million votes. It is certainly not record-shattering turnout, but it is a good turnout for the PS. For another comparison, in the closed primaries of 2006, only 179,412 Socialists had voted and in the 2008 election of the PS’ First-Secretary, 232,912 members were registered to vote. The PS said that it had expected 1-2 million voters, so 2.7 million voters is a good figure. And it also brings in €3 million or so… never a bad thing for any political party.

The results were as follows:

François Hollande 39.17%
Martine Aubry 30.42%
Arnaud Montebourg 17.19%
Ségolène Royal 6.95%
Manuel Valls 5.63%
Jean-Michel Baylet (PRG) 0.64%

Hollande, as the polls had predicted, came out comfortably ahead of his main rival, party boss Martine Aubry in the first round. The polls had not messed up, he was the true frontrunner. However, Hollande did not do as well as the polls had predicted, though not in the end by a large margin. His result, 39.2%, is strong but it is under the symbolic ‘40% line’ which most polls had predicted he would cross. He has an 8.75% margin over Aubry, which is also slightly less than what the polls had given him, and more importantly it is under the symbolic 10% margin which would have maintained his solid advantage. It is certainly not a defeat or even a major setback for his candidacy, but undoubtedly the slight underperformance on October 9 has broken Hollande’s strong momentum somewhat and changed the dynamics of the runoff. Hollande likely suffered from a demobilization of his potential electorate, which decided not to bother voting given how certain his victory seemed to them on the eve of the vote. The polls in general also underestimated the size of the gauche de la gauche electorate in the broader primary electorate, with the higher than expected turnout of the party’s left-wing playing against Hollande, who was the more centrist of the main candidates.

Martine Aubry won 30.4%, a showing which is a few points better than what polls had given her though not by any means a shocking overperformance. Yet it is for her and her supporters a strong result and, coupled with the smaller than expected gap between her and Hollande (8.75%), will revitalize them ahead of the runoff which is very open-ended. Her strong showing undeniably boosted her supporters’ morale. One of Aubry’s main advantages, especially in the circumstances of this runoff is that she has made herself the standard bearer of the “true left” of sorts, that is a clearer and more offensive left. Hollande’s main weakness is his centrist, feel-good image which makes him appear as a weak-willed, opportunistic and flip-flopping candidate to the party’s left.

Certainly the biggest surprise of this primary and one of the factors which boosts Aubry is the strong showing of Arnaud Montebourg, the young maverick figure of the PS’ left and the vocal proponent of démondialisation. Montebourg’s late surge into third place had been picked up by most pollsters, but the remaining undecideds and late-deciders broke heavily in his favour, meaning that pollsters all underestimated his performance by at least 3-4%. He won 17.2%, a very strong third and a result which, if played correctly, promises Montebourg a bright political future. Montebourg was clearly the only candidate who gained something from the three debates and the only candidate for whom the ‘official campaign’ had a major impact on his numbers (because he led what was probably the best campaign of the 6). Montebourg’s left-wing rhetoric of deglobalization, European protectionism, reindustrialization, institutional shakeup and fighting corruption struck a chord with the gauche de la gauche, which turned out in big numbers on October 9. His left-wing rhetoric appealed to those voters, who, in these times of economic crisis, found his radical leftism quite attractive and saw in him a refreshing change. Ségolène Royal’s campaign was all about appealing to those indignés, but in contrast to Montebourg’s well orchestrated campaign, hers was chaotic, overly populist and sectarian. Montebourg had a clear, well-managed campaign and he was able to defend his ideas with intelligence and charisma – which Royal failed to do.

Ségolène Royal was the major loser of the primaries. For the PS’ 2007 candidate, who had won the 2006 closed primaries in a landslide and had come within 102 votes of winning the party’s leadership in 2008, her phenomenal downfall is very bitter indeed. She had styled herself as the candidate of the indignés, but on this ground she found herself at a loss against Montebourg’s more credible and more reasonable discourse. Her chaotic, erratic and populist discourse of jumbled-up vote-winning goodies and random ideas did not convince. With 7% support she did very poorly, worst in fact than predicted by most pollsters – ironically enough in fact, having gone on a bizarre crusade against the pollsters when they started showing bad numbers for her.

Manuel Valls, with 5.6%, did about as well as a candidate on the PS’ right could expect. There is simply not a large base within the French left for a candidate who campaigns vocally for fiscal orthodoxy, against the 35-hours and against welfare-leeches. The story is a similar one for Jean-Michel Baylet, the sole non-PS candidate in these open primaries, who could not expect to do much better in a primary which was effectively a PS primary. His 0.6% are a paltry showing, but for Baylet, what counted more than the result was just participating to increase his party’s notability and remind the PS that he is a very loyal ally who expects to receive his fair share in upcoming negotiations for legislative seats for 2012.

Geographical analysis

What is fascinating about the map of this primary is how homogeneous Hollande’s support was, and also how the resemblances with internal PS party shenanigans are sparse. True, the map is certainly that of a left-wing primary, as Aubry’s isolated strongholds reveal. But it is quite different from the maps of internal PS party business, elections in which the support of the local federation’s bigwig will sway the whole federation your way. Some departments, like the Nord or Seine-Maritime certainly followed the orientation of the local federation boss. But in a lot of other cases, the endorsement of the local bigwig didn’t have much effect: Aubry lost the Bouches-du-Rhône despite Guérini’s support, Hollande lost in Lyon despite Collomb’s support and Montebourg lost in Guyana despite Taubira’s support. Right off the bat, this makes these primaries much, much cleaner and harder to rig.

Hollande received very homogeneous support throughout France, which is a strength for his candidacy. His strong base of support with provincial elected officials and local notables surely helped, but above all his support shows that he was the candidate of choice for the middle-classes (la France moyenne), employees or small businessmen. He is the candidate of rural France, the small and mid-sized towns and the  Hollande performed best, obviously, in his local stronghold of Corrèze where he received 86% of the votes, a local base which overflowed in a Chiraquian-Pompidolian manner into Haute-Vienne, Creuse, Cantal or Lozère. Hollande also performed well in regions such as the inner west or Brittany, politically moderate regions where the PS has gained in strength recently. While Hollande performed strongly in the quiet suburbs surrounding most major cities, he was not the favourite of urban voters and he lost most or performed comparatively weakly in most of France’s largest cities. Within cities, Hollande’s support was highest in the most affluent (and often most right-wing) neighborhoods. As the Paris inset of the main map shows, Hollande’s strength in Paris was concentrated in the city’s affluent west-end. A similar pattern can be seen in Lyon or Marseille. This is not to say that Hollande was purely the candidate of the affluent, in fact one of Hollande’s main strength here is his proven ability to appeal to a heterogeneous base to build relatively homogeneous levels of support throughout France. Not too surprising, perhaps, given Hollande’s conciliatory and moderate “normal president” image.

In contrast to Hollande’s homogeneous map, one of Aubry’s weaknesses as evidenced by her map is the relative confinement of her support to small bastions. Aubry won only four departments: the Nord, Pas-de-Calais, Seine-Maritime and Paris. In her native Nord, she won 54% and her local favourite daughter base also extends into Pas-de-Calais. In the Seine-Maritime, the support of the historical local left-wing bigwig Laurent Fabius probably played a key role. In Paris, Bertrand Delanoë’s support was probably not without effect. In the Seine-Saint-Denis, which she lost by 1%, Claude Bartolone’s support was also probably not without effect. In general, Aubry convinced a young, urban, generally well-educated and ‘trendy’ electorate. In contrast to François, the candidate of small-towns and the province; Aubry was the candidate of the cities. Though often losing the department in which they are located in, Aubry won most of France’s major cities: Paris, Lyon, Lille, Strasbourg, Toulouse, Grenoble, Rouen or Metz (and came within 2 votes in Rennes). A breakdown of the vote in Paris, Lyon or Marseille shows that Aubry carried the more working-class (and left-wing) neighborhoods but also the more trendy and central bobo areas. In middle-class suburbia or in rural France, Aubry struggled far more. In more urban and in general more politically left-leaning areas (the old left-wing areas in the Nord-Pas-de-Calais and Picardie, for example), Aubry did better against Hollande. Her map shows a little Green effect: she performed strongly in regions where Greens do well: Ille-et-Vilaine, Loire-Atlantique, Savoie, Isère, urban areas and even Alsace (!). This is the only visible effect of non-PS voters on the primary map in a consistent fashion, as the maps show very little perceptible Communist or far-left influence.

Arnaud Montebourg’s map was also relatively evenly balanced, ignoring massive favourite-son voting in his Saône-et-Loire stronghold (56%). Unlike Aubry, his local support spills over into neighboring departments: he performed well above average in next-door Jura (nearly 30%), Ain, Côte-d’Or and Nièvre. In good part, Montebourg’s support was quite rural or small-townish, doing especially well in isolated and ‘forgotten’ departments such as Alpes-de-Haute-Provence (24%), Lot (21%), Lozère (19%), Drôme or Ardèche (22%). These rural departments have been hit hard by a decline in local public services, and there is a lot of anger in these ‘forgotten’ rural confines against Sarkozy and the UMP in general. In other parts, Montebourg’s support was reminiscent of FN support: Montebourg performed above average in traditional frontiste strongholds such as Vaucluse, Bouches-du-Rhône, Var or Alpes-Maritimes. Small town or low-income protectionist voters who may flirt with the FN in other elections, or perhaps – in the Bouches-du-Rhône particularly – a vote against local corruption? In urban or old industrial areas, Montebourg performed about average, and in urban areas his support was usually correlated quite closely to Aubry’s support.

Ségolène Royal did not even save face with her results. She won only 18% in her political base in the Deux-Sèvres, a distant second behind Hollande. Even in her own political stronghold, Melle (in Deux-Sèvres), she was 10% behind Hollande… even Jean-Michel Baylet won his hometown by a big margin. She polled best, with 15-18%, in the region she governs, Poitou-Charentes but did poorly outside there with homogenously low support throughout the rest of France. Royal was popular and had targeted low-income suburban neighborhoods, but even in those top targets for her campaign, she barely polled 10% – at best.

Manuel Valls polled 11% in Essonne and 10% in the Hauts-de-Seine, and won in his hometown of Evry. While some may have thought his political implantation in a commune populaire like Evry and his law-and-order rhetoric might have helped him in other difficult suburbs, he did poorly in those (5% in Seine-Saint-Denis). His support was heavily concentrated in affluent neighborhoods and municipalities. Like Hollande, he did best in posh west-side Paris (over 15% in the 7, 8 and 16th arrondissements) or in the wealthy parts of Lyon or Marseille. He polled 23% in Neuilly-sur-Seine, hometown of a certain Nicolas Sarkozy. Outside Ile-de-France, Valls’ map has a spookily close correlation with that of the FN in the 1980s and early 1990s.

Jean-Michel Baylet managed to win something. Saint-Pierre-et-Miquelon, with 39% or 106 out of 269 votes. Saint-Pierre-et-Miquelon had elected a PRG deputy in 2007, who had backed Baylet in the primaries. In France, he polled best – 15% – in his native Tarn-et-Garonne where he is President of the General Council and Senator. He also won his hometown. But his result in Tarn-et-Garonne represents a full 11% of his national support, and it was one of his only two good showings, the other being the PRG stronghold of Haute-Corse where he won 14% (and more votes than in Paris!). His map, besides that, peaks at 2% support in the traditionally Radical departments of Lot and Hautes-Pyrénées or hardly impressive 1 percents in a lot of old Radical bases.

Predicting the unpredictable

What result for the runoff? The closer-than-expected margin in the first round between Hollande and Aubry will mean a closer-than-expected runoff. While Hollande probably remains the marginal favourite going into the runoff, his slight underperformance on October 9 has broken his momentum somewhat and given Aubry much more momentum then she had on October 8. Both Hollande and Aubry have things going for them, which makes predicting the final result rather perilous and meaning, in effect, that this race is very much open.

François Hollande’s main advantage is that he enters the runoff with a high floor. He can count on the support of at least two-thirds to three-fourths of Valls’ and Baylet’s voters (0.6% and 5.6%), which places his floor at roughly 46%. He has already received the endorsements of both Manuel Valls and Jean-Michel Baylet, both endorsements of low impact but important in that they give him a high floor. More surprisingly, he also received on Wednesday the surprisingly clear endorsement of Ségolène Royal, his ex. It is doubtful whether Royal’s endorsement will have much impact on the final result, given that her reduced electorate is largely composed of her die-hard fans and is rather left-leaning (and thus probably more towards Aubry than Hollande, all things being equal). But the endorsement of his ex is a nice momentum-booster for Hollande, showing him as the candidate most capable of rallying support from both sides of the playing field (Valls and Baylet on the right, Royal on the left).

The PS primary and the primary-related events of this week are all over the news in France, something which should increase turnout even more on October 16 where the motivation to turn out is even higher than last week. There are, in France, a good number of voters who only turn out in the runoff. It is hard to evaluate where these new voters will be coming from and how they will break between Hollande and Aubry. If it is true that some of Hollande’s potential voters abstained on October 9 because his victory was looking very likely, then some of those voters should logically come out to vote on October 16. Higher turnout from centrists or from PS sympathizers should help Hollande, while higher turnout from the radical left or the Greens should help Aubry. Aubry, in general, basically needs any new voters to prefer her over Hollande. Hollande is not as reliant on the support of new voters, and could win the primary while losing these new voters.

Martine Aubry’s main advantage is that her more left-wing positioning in the primary means that she is the most likely benefactor of Arnaud Montebourg’s support. Montebourg, very much a kingmaker, is unlikely to personally endorse any candidate. Though the maverick Montebourg had supported Aubry in Reims in 2008, the two have crossed swords on the issue of the primaries themselves (Montebourg being one of the earliest proponents of open American-like primaries) and most recently on the Guérini affair (Montebourg accusing Aubry of being soft on corruption within the PS). He has repeatedly called Hollande and Aubry two sides of the same coin. While some of his supporters from October 9 will likely not turn out in the runoff, most of his electorate will probably turn out a second time. This is where Aubry’s chances lie: she needs, absolutely needs, strong support from Montebourg’s voters to win the runoff. She appears to be the ‘natural’ choice for his voters. While they have their differences (Europe, corruption etc) she is more left-wing than Hollande and Montebourg’s voters probably harbour lingering doubts about Hollande’s left-wing values. They are more likely to see him as weak-willed, directionless and a flip-flopper. That is exactly why Aubry has been going after Hollande rather violently on these points.

In the final high-stakes TV debate held on Wednesday, Aubry very much targeted Montebourg’s voters by taking a more left-wing tone and attacking Hollande of being soft, weak-willed, a flip-flopper, incoherent or even ‘too rightist’ and ‘too system’. The debate was pretty much a sleep-inducing draw, with Hollande and Aubry both having their weaknesses and strengths and the ‘winner’ largely dependent on one’s perspective on the candidates. It is hard to evaluate what impact the debate will have, but I doubt it will have much impact. Aubry’s increasingly desperate tack to the left is clearly aimed at winning over Montebourg’s voters, and it is really hard to evaluate from my perspective whether or not her debate performance and her tougher and tougher attacks against Hollande will succeed in winning over Montebourg’s voters.

[last minute: Friday Oct. 14 >> Montebourg has announced his personal endorsement of Hollande. A very big blow to Aubry’s increasingly chaotic and desperate campaign, and a major boost for Hollande’s campaign, perhaps not as much in absolute vote terms but in terms of momentum and last-minute advantage. It remains to be seen whether or not Montebourg’s voters will vote like their guy in droves, but even if they don’t all vote like him, it remains a pretty crippling blow to Aubry. Her desperate and bizarre attempts to attract him has not work, but it might attract some of his voters. Hollande should be counted as the major favourite in this contest right now.]