I will admit that I stole (no, borrowed) this title from Josep Colomer. I have been meaning to post for some time on the French presidential election campaign. The first round of this two-round majority contest is on 22 April.
The main contenders are Nicolas Sarkozy of the incumbent conservative bloc, SÃ©golÃ¨ne Royal of the Socialist party, and FranÃ§ois Bayrou of the Union for French Democracy. And, of course, there is Jean-Marie Le Pen of the Front National and a series of other candidates–twelve in total.
In 2002, France experienced something of a debacle with its majority-runoff system and the failure of the left to consolidate sufficiently as to place its leading candidate, Lionel Jospin, into the runoff. With the left badly split, Le Pen passed Jospin for second place, setting up a runoff between right and more right–and a national embarrassment in the process.
In 2007, there is little risk that Le Pen will play in the second round, but there is a different dilemma on the left. Polls suggest that Royal will not defeat Sarkozy, but Bayrou would. However, he may not make it to the runoff, as he is running a close third behind Sarkozy (who has been fairly consistent in the 27-31% range recently) and Royal (slipping recently to below 25%). Bayrou is essentially tied with Royal in the low-to-mid twenties.
For instance, this poll seems fairly typical of recent samplings that have asked about the possible runoff pairings:
Sarkozy beats Royal, 54-46
Bayrou beats Sarkozy, 54-46
Bayrou beats Royal, 60-40
In each pairing, 17-18% say that they would not vote in the runoff.
(Le Pen is polling around 12% for the first round and 46% of his supporters favor Sarkozy in the runoff.)
So, Bayrou looks like a clear Condorcet winner (the candidate who would beat any other in a head-to-head race). But if he can’t place first or second in the initial round of voting, he will never get to prove it. And Sarkozy, apparently, will be France’s next president.
Should leftist voters vote strategically for Bayrou, in order to stop Sarkozy? Or vote for their candidate and hope for the best? This will be one to keep an eye on.
As I have noted in the past, I am not of the view that a presidential-selection method necessarily must ensure the election of a Condorcet winner (when one exists). However, that view stems from the recognition that the potential Condorcet winner is often one with limited first-preference support. As long as a method prevents the election of a Condorcet loser (as two-round majority does), I prefer a method that ensures that the winner has significant first preferences over one that ensures a Condorcet winner.
This French race is different: The Condorcet winner, Bayrou, has support not much less than the other two (assuming he does, in fact, come in third place). This is a race in which it is possible that a different method of determining the majority could produce the more broadly supported candidate from the three leaders. A one-round ranked-choice ballot with sequential elimination (IRV or Alternative Vote) might allow Bayrou to pick up support from other defeated contenders and survive in the counts long enough to emerge as the winner. There is, of course, no guarantee that this would happen: the transferred votes of the more extreme parties might simply boost both Royal and Sarkozy in the later counts, and leave Sarkozy as the winner, with Bayrou falling further behind. The outcome would depend on the fuller preference profiles of voters, which I doubt any polling sample has recorded.
If the French left suffers from the vagaries of majority runoff twice in as many elections, I wonder if we will see an IRV movement in France. (I wonder if one exists now. Anyone?)