Monday, July 11, 2011

Do Bayrou and Le Pen Draw on the Same Pool of Voters?

Frédéric Lefebvre-Naré asks the question and uses his statistical skills to answer it in the negative:

J'ai vu ressurgir sur Facebook aujourd'hui une vieille idée typiquement colombano-minco-UMPS, selon laquelle le vote Bayrou et le vote Le Pen puiseraient à un même réservoir électoral, celui du "ni ni" (populiste, démago-râleur, etc.).
J'attends depuis de longues années de voir cette assertion étayée...
...Mais en tant que statisticien, je peux verser quelques éléments publics au débat.
[for graphics, see original post]
Le vote Bayrou a évolué de façon tout à fait indépendante du vote Le Pen ; le vote Bayrou a concurrencé le vote Royal en début et fin de période ; le plus fort transfert semble avoir été, fin février-début mars[1], du vote Sarkozy vers le vote Bayrou.
Si, bien sûr, les chiffres des sondeurs méritent la moindre crédibilité. Mais sur quelles autres sources se fondent ceux qui parlent de perméabilité entre les électorats nationaliste et démocrate ?


Mitch Guthman said...

All these graphs and charts are beyond me but what Frédéric Lefebvre-Naré says is interesting because, if I understand him correctly, it seems to confirm much of what Bayrou has been saying. It suggests that the smart move by the UMP would indeed be shifting slightly "leftish" (and not further right) to compete with the PS for the much larger, arguably more fluid pool of disaffected center-left voters who are not enthusiastic supporters of multiculturalism and wish to crackdown on things like binationalité and large-scale economic migration on the grounds that such things are incompatible with secularism and the social welfare state. (A position I actually do hold).

(Another implication might be that Sarkozy has erred by shifting further right in an effort to compete with the FN for what just might a much smaller, sticker and less accessible pool of FN leaning voters)

Unless, of course, I have seriously misunderstood this quote from Frédéric and perhaps also Bayrou’s political approach as a whole in which case this entire comment should be disregarded(something which I concede is a very distinct possibility).

FrédéricLN said...

For sure, it should not be disregarded!

I wonder whether these many voters who "wish to crackdown on things like binationalité and large-scale economic migration" would be considered as "center-left" in French political terms, but you may be right. On such topics: it's complicated. What people may say to express frustration or anger, may be very different of what they really expect from political leaders.

My analysis of the 2007 campaign is not far from yours, after all.

Bayrou seemed to win the contest for the "center" voters because his agenda appeared as much more serious and credible than Royal's (as usual for the left) and Sarkozy's (while right-wing candidates make usually, in France, much more serious proposals than left-wing ones, for which "le projet" is "un os à ronger pour les militants" - remember Mitterrand did not even *read* his own "110 propositions" for the 1981 campaign).

So Bayrou earned some percentage points at Sarkozy's expense from mid-February 2007 to the beginning of March. This remained unnoticed from the media because they focused on the fight for the 2nd (qualifying) place between Royal and Bayrou. But I guess Sarkozy's attendants did notice, and pushed him to reply on the same ground, not to loose the center. Sarkozy made the opposite move, at his MP's or his new supporter Simone Veil's surprise, in the direction of the nationalist voters — announcing the institution of a Ministère de l'Identité Nationale. He defined the terms of the debate of the next three weeks. We in the Bayrou campaign remained unable to make a counter-attack at the same level of power. The PS was basically running for the 2nd place ("not to repeat the 21 avril") and, Royal herself excepted, did not fight for victory.

Usually, older voters in France choose the candidate in the center, the "moderate" one. We discovered, or admitted, very late in the campaign (around mid-March, I guess), they were to vote massively Sarkozy. The "national identity" (anti-immigration) stance appealed much more to them than our reasonable / humanist speech. It's still so since (in the 2009 and 2010 elections at last).

You can understand that by the assumption that the "national identity" (anti-immigration) stance IS, presently, considered in the opinion as a moderate, reasonable, "center" orientation, and that would be where older voters remained after all. "Center" or, according to your comment, "center-left"!

(and thank you Art for recommending these curious graphs!).

FrédéricLN said...

There is new stuff on the topic. The best source to know if there are transfers between Bayrou and Le Pen supporters is OpinionWay, because they run a large panel (repeated interviews of the same people), allowing to estimate switch rates.

But their findings are usually not in the media — only for their subscribers (in 2007: Sarkozy's campaign).

In tomorrow's daily newspaper La Croix, OpinionWay's Julien Goarant says that "there is <> (to assess) this alleged transfer, as <>.

"cet éventuel transfert ne correspond ... à « aucune réalité statistique, la porosité entre les deux électorats étant extrêmement faible »"

FrédéricLN said...

Oops, I forgot the link :