France 2012: PS open primaries (runoff)

The second round of left-wing open primary elections (primaires citoyennes) were held in France on October 16, 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. I had talked about the candidates in a preview post and covered all the results of the first round here.

At the outset of the first round, the frontrunner of the campaign, former party boss François Hollande had come out ahead with 39.2% ahead of 30.4% for Martine Aubry, his main rival and current party boss. Hollande’s showing had been a bit weaker than originally expected, and Aubry’s performance was conversely better than expected. The main surprise of the first round came from the strong showing of Arnaud Montebourg (17.2%), the young standard-bearer of the party’s left and ‘deglobalization’. The party’s 2007 candidate, Ségolène Royal, won only 7% while the young standard-bearer of the party’s right, Manuel Valls, won 5.6%. PRG boss Jean-Michel Baylet won only 0.6%. Aubry’s hopes for the surprisingly open runoff rested on the transfer of Montebourg’s left-wing voters to her candidacy, more left-wing than Hollande’s consensual centre-left candidacy. Support from new voters and a good majority of Royal’s equally left-wing voters was also necessary. However, despite a very aggressive offensive against Hollande from the left (Hollande as ‘inconsistent’, ‘flip-floppy’, ‘unclear-wishy-washy’, ‘the system’s candidate’ or even ‘right-wing’), Aubry’s campaign failed. Royal, surprisingly, endorsed her ex (following the logical endorsements of the centrists Valls and Baylet). Then Arnaud Montebourg, on a personal level, endorsed Hollande, despite his ideas being generally closer to those of Aubry rather than those Hollande. Following these pretty crippling blows, Aubry needed two things to win: that Royal and Montebourg’s voters reject in large numbers the endorsement of their candidate and vote for Aubry instead, and that the runoff sees a major spike in turnout through the heavy participation of non-PS left-wingers (Greens, Communists, far-left) which favoured Aubry more than Hollande. A tall order, but not seen as impossible. Despite Hollande’s big momentum during the entre deux tours, almost all observers predicted a fairly close runoff though with Hollande favoured.

Turnout was, logically, up from the first round. With 9407/9425 polls confirmed (the rest were likely cancelled) 2,860,157 voters turned out, which is 6.6% of registered voters in France (2010 numbers, the same warning applies concerning this data as in round one). This up nearly 199k from the first round, a 7.5% increase. The metropolitan department with the biggest increase was Haute-Corse (+26.85%) followed by Corrèze (+18.77%). The Nord, Pas-de-Calais, Seine-Maritime, Haute-Loire and Pyrénées-Atlantiques also saw turnout increase by over 15%. Only Saône-et-Loire (-5.53%) and Tarn-et-Garonne (-0.22%) saw turnout drop.

François Hollande 56.57%
Martine Aubry 43.43%

In the end it wasn’t even close. Hollande, after underperforming in the first round, overperformed most polls in the runoff. While Hollande’s victory was pretty likely, nobody really thought that he’d be able to win by such a big margin – despite the big momentum going his way with the endorsements or personal support of all other four first round candidates.

Turnout increased by just about the right amount for Hollande – a more significant boost in turnout, which would likely have come from the non-PS left, would have been more likely to benefit Aubry. As I had hypothesized following the first round, a good number of potential Hollande voters seemed to have opted out of the first round because he was the big favourite, but they certainly turned out for the runoff. Then his undeniable big mo’ had a rather important effect. It might have pushed wavering voters to his side, and it also could have even taken a few first round Aubry voters who wanted to vote for the winner and guarantee him a large majority.

The most surprising aspect of the vote is also that those who had voted for the other four candidates in the first round followed the endorsements or opinions of their candidate rather loyally. In the case of Valls and Baylet voters, this is not surprising. But more surprising is the case of Royal voters. I had thought they would be more likely to abstain in very large numbers, and a good number probably did – turnout increased by less than the national average (+7.5%) in her Poitou-Charentes base (15-18% for her in the first round), meaning that a good number of her voters didn’t turn out but this was compensated by new voters. But those of her voters who did vote seem to have gone to Hollande by a rather significant margin. Aubry needed those Royal voters to win, and given the likely proximity of Royal voters with the party’s left, it wasn’t a pipe dream. But the numbers, especially in Poitou-Charentes where Hollande beat Aubry by over 30% in all four departments, show that her voters probably went in bigger numbers to Hollande rather than Aubry. In her native Deux-Sèvres, where she had won 18% in the first round, Hollande won 72.2% – one of his biggest wins outside his Limousin stronghold. In Melle, Royal’s home base where she had won 32% in the first round, Hollande won 77.1%. The Poitou-Charentes perhaps isn’t the best of examples: a lot of Royal’s vote here was probably native-girl rather than any ideological proximity to her, and the Poitou-Charentes, like the inner west or Brittany fits the profile of Hollande’s locales: provincial and politically moderate. But it is hard to conduct a good study, given that outsider her home turf, her results were uniformly low to the point where it’s hard to say if the runoff results were influenced by her voters.

Equally surprising was the case of Montebourg’s voters, who were very much the ‘kingmakers’ of the runoff and a crucial block for Aubry to win if she was to win. First off, a good number of them probably did not turn out. Some wavering voters (perhaps some FN voters) voted for Montebourg as a pure ideological or protest vote, and had no affinity for either Hollande or Aubry (in fact, they probably saw them as one and the same). Others were traditional PS voters, but very much on the party’s eurosceptic left and did not vote for either Hollande or Aubry (broadly similar ideologically, especially on Europe or such things). Turnout dropped by 5.5% in his native Saône-et-Loire (56% for him in the first round), which is a pretty huge number considering that turnout overall increased by some 7% in the whole of France. But a lot of his voters did turn out, and they rather surprisingly seem to have followed their candidate’s personal endorsement of Hollande rather loyally – though overall the split might have favoured Aubry by a hair (hard to say). Hollande won Saône-et-Loire by a bigger margin than he won nationally, taking 59.6%. In Montret, where Montebourg had won 97% (and Hollande 3%), Hollande won 69% (turnout dropped from 118 to 106). In Montebourg’s Frangy-en-Bresse bastion, where he had won 85%, Hollande won 61%. Even if Montebourg’s voters, overall, might have favoured Aubry by a hair, it is the very fact that Hollande – the ‘centrist’, ‘flip-flopper’ and ‘soft left’ candidate – could take so much Montebourg voters, voters who voted for the candidate farthest to the left and the proponent of ‘deglobalization’. Though some Montebourg voters expressed anger at Montebourg’s surprising personal endorsement of Hollande (the type of Montebourg voter who would not turn out), a surprisingly large amount of them opted for Hollande despite all of Aubry’s attacks on Hollande from the left. Montebourg himself said that he endorsed Hollande primarily because he came out ahead on October 9 and would have endorsed Aubry if she had come out ahead (in other words, he wants a cabinet position), this feeling of “unite around the favourite” seems to have been shared by a large number of left-wing voters. In the end, more than any questions over Hollande’s socialist pedigree, a lot of undecided or on-the-fence voters voted out of a desire to defeat Sarkozy, which Hollande is widely seen as the most capable of doing.

Aubry also played her runoff campaign very badly. The rapid string of endorsements from Valls, Baylet and particularly Royal didn’t help her, but she had the momentum on her side coming out of October 9 and she allowed herself to completely lost that. True, some of it wasn’t her fault, but she probably hurt her cause more than helped her cause when, starting during the Wednesday debate, she turned to a chaotic, aggressive, left-wing and frankly desperate tone against Hollande. It was hard to evaluate from a neutral perspective how her attacks on Hollande as being “the soft left”, or generally a centrist wishy-washy flip-flopper of questionable left-wing pedigree would help her. She did attack Hollande on his weak-point (that of being too centrist and ideologically unclear), but she did so in an increasingly desperate and ‘wtf’ way. She went off the deep end when she started calling Hollande “the candidate of the system”, which is a pretty amusing thing for the incumbent party boss backed by the party old guard to say. Her poor showing shows that she did not help her case with her attacks on Hollande.

Although the UMP, which resorts every day to more and more awful strawman arguments, would like to make you think that this shows an awfully divided party because “obviously Hollande should have won with 70%” (arguments that make psephologists cry); this is a good result for the party itself. If the race had been close, like Reims in 2008, then there would have been chaos and the personal feuds would have resisted and played out. Instead, such a definite victory is a big boost for Hollande. It gives him a big legitimacy boost and makes the continuation of major feuds unlikely. The Aubry camp certainly isn’t too enthused about the result, but some key Aubry supporters like Fabius have already made their moves to cozy up with Hollande.

One of the things which had very much crippled Royal’s campaign in 2006-2007 despite her landslide closed primary victory was the division of the party between her campaign, led by Royal and her lieutenants; and a rather hostile party establishment and old guard which barely raised a finger for her. Certainly Royal was and remains a very polarizing person within PS circles, and unlike Hollande has a lot of enemies. And the 2006 primary had left its wounds on the party. The old guard came out to help Royal only late in the campaign, and did so with little enthusiasm. While the Aubry camp might be gloomy and hardly enthused about their rival’s victory (and there are clear scars), Hollande has so far handled his victory better than Royal did in 2006. Right after he won, he staged a very powerful and symbolic victory celebration alongside Valls, Royal but also Aubry and Fabius. The danger for him and the PS remains a division of the party between the campaign and the party, especially given how the party apparatus is led by one Martine Aubry.

Geographic analysis

Those who hoped for a ‘bluer’ map than that of the first round will be pretty disappointed. Hollande won all departments which had voted for him on October 9, except for the Somme which went to Aubry by a handful of votes. He also picked up St. Pierre et Miquelon, which Baylet had amusingly won. The patterns of the runoff are broadly similar to those of the first round (well, obviously). The map clearly shows Hollande’s solid ‘provincial’ or rural implantation. His support ripples out from his Correzian bastion, which gave Hollande an “African-like” 94%, and extends into the rest of the Limousin and Massif Central to give his map a distinctively chiraquian-pompidolian flair. But he also won strong support in the rest of the left-leaning and rural departments of the old Radical southwest, and polled remarkably well in Royal’s Poitou-Charentais base (not all surprisingly: as I said above, this is a politically moderate region). He also did well in the bulk of Auvergne, the Languedoc-Roussillon (perhaps a frêchiste vote for Hollande against Aubry, who had gone on a crusade against the frêchiste feds of the region) and generally in rural Champagne, Touraine, Berry, Orléanais, Lorraine and even a good part of Bourgogne.

Only the more urbanized and Greenish departments of the Rhône-Alpes, PACA, Ile-de-France and surprisingly Alsace are a bit weaker for him, in general. His support was lowest in the Nord-Pas-de-Calais, the political home base of Martine Aubry and industrialized Picardy and Haute-Normandie (the latter of which is the political home base of Laurent Fabius).

In the first round, I had noted a rather stark urban-rural divide between Aubry and Hollande, with Aubry being the urban candidate against the ‘rural’ Hollande. The same pattern is replicated in the runoff, though not as pronounced. The gap between Aubry and Hollande in Paris, which Aubry won as in the first round, was smaller than in the first round. Lyon seems to have voted for Hollande, and Strasbourg went to Aubry by only 54 votes. I haven’t found data for Toulouse, Rouen or Metz. The addition of Valls’ very urban first round 6% perhaps played a role here, as did increased turnout. The map in urban areas such as Paris, Lyon and Marseille retain a bourgeois-for-Hollande and poorer people-for-Aubry pattern, though here again it is less pronounced – Hollande won the very left-wing, diverse and low-income 13th in Paris.

Turnout increased the most in the bases of both candidates: both Corrèze and the Nord had turnout up by 15%. Aubryst Pas-de-Calais and Seine-Maritime also had similarly large increases in turnout. As aforementioned, turnout dropped significantly only in Montebourg’s Saône-et-Loire but also, interestingly, by a tiny amount in Baylet’s Tarn-et-Garonne. Probably a handful of PRG voters with little interest in the PS.

Conclusions: Past and Future

4%. That is what Hollande was polling one year ago. Though he rarely trailed in the polls following DSK’s arrest, Hollande did indeed come back from far. When Hollande left the party’s leadership in 2008, he was widely judged as a poor leader, the memories of the 2002 and 2007 presidential defeats fresh in mind. His leadership had been criticized for being too conciliatory and soft: making no enemies, but making no friends. Outside a close circle of supporters (Le Foll, Sapin, Vallini, Le Roux etc), Hollande went through a traversée du désert (a trough) which lasted pretty much until late 2010 or even early 2011. But slowly, Hollande fought back and was determined to win. His reelection as president of the general council of Corrèze in March boosted his profile and he rode on a little wave of momentum to announce his candidacy right after the cantonal elections. Unlike Aubry’s campaign, his was well managed and always at the forefront of events. Aubry seemed to have gone on vacation right after announcing her candidacy in the summer. At first, when DSK was still the prospective favourite, he was slowly rising on a profile of “normal” (and ‘provincial’) president – as opposed to the flashy international IMF-lifestyle of the former IMF director. When DSK was arrested, he surged into the lead and would (by large) only grow his lead as the weeks went. He took a lot of DSK’s potential support – moderate voters in the party’s reformist social democratic wing. His “normal” president stature and image was a serious boost, especially in presenting him as the candidate who was most able to beat Sarkozy (a vote-determinant not to be downplayed), but also against the more ‘elitist’/’system’ Aubry. His ability to stick his ground in the trick runoff campaign pitted against a powerful left flank (Montebourg) raised his credibility in the eyes of the wider public. Some might have thought that his ‘general election’ type of campaign wasn’t appropriate to a primary, but it actually worked out well for him. Set against Aubry’s boring, uninspiring and mismanaged campaign it isn’t a wonder, in the end, that he won like he finally did. To draw a link with American politics, Aubry was a bit the “Clinton” of the contest – longtime politico, apparatus insider support, experienced rough-and-tumble politician and in a position to win easily through their control of the party’s insider network and structures (but who finally led a poor campaign). Hollande, though most certainly not a political newbie or with low name recognition, was a bit the “Obama” of the contest – a bit on the outsides of the current structures, relatively inexperienced (Hollande never served in cabinet), starting out very low in polls but leading an innovative and organized campaign and playing on his strength with the ‘grassroots’ and the broader electorate.

The top echelon of the 2012 field is thus set: Nicolas Sarkozy, François Hollande and Marine Le Pen. Hollande is a strong candidate and a threat to Sarkozy. Of the two candidates, he was the strongest against Sarkozy. The blog Sondages 2012, run by a friend of mine, which has a wonderful rolling average of all polling pegged Hollande at 30.4% on October 10 and Aubry at 26.8%. Sarkozy was at 22.7% against Hollande and 23.3% against Aubry. One of Hollande’s main strength, which helped him in this campaign, is his ability to attract centrist and even centre-right voters to his fold. His conciliatory, moderate, reformist social democratic image gives him a stronger base with those voters. The main losers of Hollande’s nomination are the ‘centrist’ candidates: Bayrou, Villepin and Borloo before he dropped out. Bayrou, Villepin and Borloo all performed better in the Aubry scenarios than in the Hollande scenarios. Conversely, those who might stand to gain something out of Hollande’s nomination are those further to the left: Eva Joly (EELV) and especially Jean-Luc Mélenchon (FG). Hollande’s left flank is weaker than Aubry’s, but if he can position Montebourg well, he could cover his left flank and prevent losing some of the more left-wing PS voters to the Greens or Left Front. Hollande, however, does have his weaknesses. His relative inexperience in governance could be used against him, and Sarkozy will probably do so by running on his experience in foreign affairs and on the debt crisis issue. He could use a theme similar to Stephen Harper’s successful “in times of economic trouble, it’s me or chaos” line. His support seems to be based quite a lot on his “normal” president image or similar ‘image’ things rather than deep ideological affiliation, and as such is probably less solid. Sarkozy is a very strong candidate too, and should certainly not be taken for dead. But, with approvals at 35% and an anemic 22% in first round polls and a terrible 40-42% in the runoff, he is certainly in worse shape than any incumbent president of the Fifth Republic were one year out from reelection.

Posted on October 18, 2011, in France, Primaries, leadership contests or internal party votes. Bookmark the permalink. 2 Comments.

  1. Excellent analysis. This primary has been quit interesting to follow (even though I was, of course, disppointed in the end) and generally speaking has been a major boost for the PS.

    Your preview of things to come is mostly right (I’ve myself made a little list of the “bright sides” of this Hollande win, and think he has proven a better campaigner than expected). Now, you’re also right to say everything can happen in 6 months and that Sarkozy isn’t dead.

    Anyways it will be an interesting campaign, and unless Hollande makes some major mistake (in which case left-wing sympathizers should hang him by feet), he should be able to win. Let’s see what the next months give us !

  1. Pingback: Election Preview: France 2012 « World Elections

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: