If Life Gives you Lemons, Make "Le Monde" (Why there's Been No Rightward Shift in France AT ALL)

The Volokh Brothers Weblog ^ | April 21, 2002 | Sasha Volokh

[Timesink]

IF LIFE GIVES YOU LEMONS, MAKE LE MONDE: In a somewhat surprising turn of events and a major embarrassment for the Socialist Party, far-right candidate Jean-Marie Le Pen, and not Socialist Prime Minister Lionel Jospin, has advanced to the French presidential runoff against center-right President Jacques Chirac. (French readers, see the Le Monde article here.) (Plug for my favorite news source: The Economist noted this possibility three days ago ('Mr Jospin will be lucky to get 20% in the first round. . . . There could yet be a surge for one of the secondary candidates of the left or the right which nudges them past Mr Chirac or Mr Jospin.'). Maybe that wasn't deep or surprising, but hey, it's a plug for my favorite news source.)

I'm observing this from a distance just like the rest of you, but because I studied with the French from second grade through high school, I get to pontificate on French politics. (These were hard-core French -- lots of cultural exceptionalism, anti-Americanism, borderline Communist apologetics, and 'Third Way' economics -- though perhaps no worse than their American educational counterparts. I had to spend years unlearning that stuff! Nonetheless, it was an excellent high school; I'd send my kids there.) My conclusion in brief: There is less to these election results than meets the eye. This election is not a crushing defeat for the left, nor does it show any significant resurgence of the far right or of extremist parties. If there's an overall rightward shift at all, we didn't see it in this election. Everything is about the same as it used to be!

First, the results (percentages are approximate, and I've often taken midpoints of ranges):

• Jacques Chirac (center-right, current president) (19.8%),
  
• Jean-Marie Le Pen (far right) (17.2%),
  
• Lionel Jospin (socialist, current prime minister) (16.2%),
  
• Francois Bayrou (center-right) (6.7%),
  
• Arlette Laguiller (Trotskyist) (6.1%),
  
• Noel Mamere (Green) (5.4%),
  
• Jean-Pierre Chevenement ('sovereignist' = old-style socialist with right-wing appeal) (5.1%),
  
• Olivier Besancenot (Trotskyist) (4.4%),
  
• Jean Saint-Josse (hunters (!), countryside, right) (4.0%),
  
• Alain Madelin (libertarian right, 'Thatcherite') (3.8%),
  
• Robert Hue (Communist) (3.5%),
  
• Bruno Megret (far right [now has endorsed Le Pen]) (2.4%),
  
• Christiane Taubira (left) (1.8%),
  
• Corinne Lepage (right-wing Green (!)) (1.7%),
  
• Christine Boutin (right, 'family values') (1.2%),
  
• Daniel Gluckstein (Trotskyist) (0.5%).
  

Do some math, and you get roughly 57% for the right and 43% for the left, assuming all these people vote the way they're caricatured in the list above and not splitting up Chevenement's vote.

What does all this tell us?

• First, this doesn't show a far-right realignment in France -- barely even a mild rightward shift. (Remember this well, because it's the opposite of what everyone else will be saying for the next few weeks.) There may be a a rightward shift in Europe generally, but we don't see it in these election results. At most, we can see a slight far-rightward shift within the right; see my point on extremists below. This is Le Pen's fourth presidential election, and the last two times he got around 15%. So, no huge change. This is already being called this a spectacular failure of the left (Le Monde calls it a triple failure), but it's not really -- in fact, the left did better this year than in the last election in 1995! It's a political embarrassment to not be part of the debate for a few weeks (and this may affect how the debate gets framed in the coming weeks, though this may not help Le Pen at all), and the strong showing of minority left-wing parties certainly indicates at least a personal rejection of the accountant-like, visionless, and humorless Jospin, but it's hard to make this out into an amazing success of the far right, a backlash against the European Union, the French chattering classes, or Arab appeasement, or a 'swing from the brink of communism to the brink of nazism.' Nor does finishing within two points of Chirac mean Le Pen 'could actually win the runoff.' (Thanks to InstaPundit for good commentary to bounce off of.) Of course this election may lead to a rightward shift insofar as Chirac, who now gets all the honor-of-France left-wing votes, gets to campaign as more of a rightward candidate within the center-right (but see the concern below about the upcoming legislative elections). But if Le Pen does well in the general election, it'll be because of changes in the mood of the electorate between now and then, but again not because of the current numbers.
  

• What this really shows is the wacky math of elections. In elections for a single seat with more than two candidates, who wins depends on the rules of the game. The way we do things around here, a Nader, Perot, or Teddy Roosevelt can split the vote and make the 'best' candidate lose (where 'best' is defined as 'preferred over any other candidate by an absolute majority'). In France, Chirac is pretty sure to win handily (all the left-wing parties have already endorsed him and a very early poll has him beating Le Pen 78% to 22%), but if the runoff had been between the top three vote-getters instead of the top two, Chirac and Le Pen might have split the conservative vote and lost to Jospin. At their least loopy, these elections might make the 'right guy' win but generate a wacky runoff challenger; move a couple percent of voters one way or the other, and Jospin makes the runoff but loses anyway. The moral here: the #1 guy may be meaningful (not because he's the 'best,' but because he may win), and to the extent #1 may win, #2 isn't very meaningful as a guide to whom people like.
  

• Interested in the wacky math of elections? See this web site for a nice (though all-caps) basic description of Condorcet cycles, an even loopier scenario, easy to imagine in a Bush/Clinton/Perot '92 context, where there's no stable election winner at all! See this article, The Perplexing Mathematics of Presidential Elections, for a good general discussion of 'choice of math,' from single transferable votes to Borda counts. Here's a slightly more complete (but still nontechnical) discussion. Nobel Prize winner Kenneth Arrow proves that (simplistically speaking) in a democracy, there's no fair way to count the votes. (Here's a description and proof of Arrow's Impossibility Theorem -- technical, but not fatally so.) An exception: if all the candidates are arranged along a one-dimensional spectrum, say left to right, everyone agrees where the candidates stand, and everyone votes for whichever candidate was closest to their single preferred point on the spectrum -- for instance, if everyone were doing a one-time vote on their favorite tax rate. Then the median voter's preference wins, and that preference would beat any other choice, regardless of the order of voting. But this doesn't look like the world of presidential elections, where there are many separable policy dimensions: economic liberty, personal liberty, foreign policy, and so on.
  

• Back to the French elections: note that the far left and far right together add up to more than a third of the electorate. Does Le Monde give us the tools to tell whether this is an increase over the extremist vote in the last presidential elections in 1995 (as they imply)? No! Le Monde compares the total far-right score (just under 20%) to Le Pen's 15% last time, without taking other far-right candidates (if any) into account; there was a Philippe de Villiers who got 4.8% in 1995, and he seems to be a traditional-values, anti-Europe conservative, but I can't tell whether he's far-right. (If he is, no change in the far-right vote! Note that this goes to show what happens when you deal in slippery categories like "far right" and "far left" -- this all involves drawing rather arbitrary lines between people who are probably ideologically pretty close.) Then it compares the far left's 'over 10%' (14%, really, not counting the Greens) to Laguiller's 5% from last time, not counting any other far-left candidates; in fact, in 1995, Laguiller and Communist Robert Hue added up to 14%, not counting the Greens, so... no change in the far-left vote! (Small note: Hue's 3.5% is the lowest ever gotten by a Communist Party candidate -- I assume they mean since World War II or in the Fifth Republic or something similar -- so even if the far left vote hasn't changed, the CP itself is in trouble.) Ah well, misuse of statistics exists even among the French!
  

• But, bigger than in 1995 or not, what do we make of this extremist third? Hard to say. Our system is two-party, first, because we have no runoffs, and second, for deep structural reasons having to do with geographical districts; and the two-party system moderates political outcomes (though not necessarily political views). The French system is multi-party, first, because they have runoffs, and second, for deep structural reasons having to do with proportional representation. The French regularly vote their consciences in the first round and vote strategically in the second round; Americans don't vote their consciences because we have no runoffs. How would the French vote if they had our system, or we if we voted our consciences? Hard to compare.
  

• Just to be contrarian -- this could spell a loss for the right, to the extent that frightened moderates vote in a left-wing Parliament. The excellent book Partisan Politics, Divided Government, and the Economy, by Alberto Alesina (from my own Harvard economics department) and Howard Rosenthal, explains the midterm effect (where the party in power loses in midterm elections) by voter uncertainty: when voters voted for members of Congress in 2000, they didn't know who was going to be President; by the time the 2002 elections roll around, that uncertainty has been resolved, so the median voter (who's between the two parties) moderates the government by voting for Congressmen of the opposite party from the President. In France, though, legislative elections are June 9 and June 16, after the presidential runoff (May 5), so all uncertainty will have been resolved by then. (You may not even need uncertainty in this scenario, since it's pretty clear that Chirac will win.) So... where's the median voter? Assuming a nice left-right spectrum (contrary to my admonitions above), with 43% voting for the left and 20% voting for Chirac (and assume half of Chirac's voters are to his left), that makes at least 53% of voters to the left of Chirac, and those 53% of course include the median voter. Hence: the left keeps the legislature. This isn't a done deal, but it's a tendency, so Chirac will have to work against it. Possibly the need to get centrists voting for his party might limit how far right he can afford to go as a presidential candidate, even within his centrist parameters.
  

• Regardless, now that Le Pen has made it to the runoff, might as well talk about him. Let's let him speak for himself: this is from an interview with Haaretz, the Israeli newspaper. 'Once, a woman politician accused me of looking at her harshly. Of course, Madam, I answered, you're looking at my glass eye.' 'There is no antisemitism in France.' 'The Dreyfus affair is an exceptional case.' Vichy is 'a case to itself,' a period of history when France 'wasn't responsible for its crimes.' Le Pen's own movement, he says, has no fascists, national socialists, or antisemites: 'I don't know a single person in the National Front [his party] who has committed an action hostile to a Jew or to Jewish property.' 'The French have a natural sympathy for Israel,' but 'the French media is pro-Arab for two reasons: the large presence of the Arab community in France and the fact that Sharon is right-wing.' Le Pen supports Sharon: 'As long as the Israelis do not support the army, the battle is lost.' On racism: 'I don't support a theory of racial superiority, but there are differences between the races.' The Muslim veil: 'It protects us from ugly women.'
  

• Lots more interesting to say about Le Pen and France, mainly about the connection between generous welfare states and racial homogeneity. But that will wait for another time -- I have to get to sleep.
  

[Timesink]

From the excellent warblog called "The Volokh Brothers" (for obvious reasons, once you get there). Add it to your bookmarks today!

[Timesink]

evening bump

[Timesink]

To find all articles tagged or indexed using Europe_List, click below:
	click here >>>	Europe_List	<<< click here
(To view all FR Bump Lists, click here)

[weikel]

I'd want the Thatcherist to win.

[Timesink]

late night bump (I can't believe nobody's reading this! It's important!)

[general_re]

Interesting. "Volokh brothers" caught my eye, and it is indeed the same Eugene Volokh who teaches law at UCLA.

I thought I'd forgotten all this stuff, but I'm mildly pleased to see that it comes back to me pretty quickly. Bump for Condorcet cycles ;)

[AmericaUnited]

I think this guy is the only one who has really hit the nail on the head.

[Timesink]

afternoon bump

[Timesink]

one last afternoon bump