Welcome to European Tribune. It's gone a bit quiet around here these days, but it's still going.
As of Saturday afternoon, there were some very interesting bits of data to chew on.

CSA's final polls on Friday clearly showed Bayrou's support collapsing and breaking towards Royal, as she had -- in both their polls (presumably for different clients?) -- closed the gap in both the first and second rounds with Sarkozy. They also showed a gain for LePen, coming apparently at the expense of Bayrou and Sarkozy.

On the other hand, a poll released Saturday by the the Swiss paper Tribune de Geneve -- though without reporting who had done it, or what method used -- also showed Royal closing the gap on Sarkozy in the first round, but reported a sharp spike in the firmness of Bayrou's support -- and the reporter, again without attribution, suggests that the raw data had Sarkozy, Royal and Bayrou within a single point of each other.

Finally, there's the poll out of a previously unheard of "institute" in Tunisia, released what it claimed to be a tracking poll of rolling, 3-day averages, showing Sarkozy and Royal more or less even (27 - 26) and well ahead of Bayrou -- but Royal well ahead in raw data (27-20 over Sarkozy) and Royal slightly ahead after attempting to adjust (based on what appears to be simply guess work) for a LePen bias, Royal 25, Sarko 22, LePen 21.

Who knows if these are simply made up?

On somewhat firmer ground, we do know (from a very interesting interview on France-Inter earlier this week with the directors of two of the largest polling institutes) that the "redressement" of the raw data which has placed Sarkozy in first for months, is based on presumptions that certain groups (notably Chirac and LePen 02 voters) are under-represented in the raw data (leading to a weighting in favor of LePen and Sakrozy) and that 02 Jospin voters are over-represented (leading to a weighting against Royal). But these presuppositions seem to me not to allow for the obvious response bias against telling a pollster that one actually abstained in 02 and instead telling a pollster one voted for Jospin. So these adjustments might actually over-compensate.

We also know the poll data under-represents younger, urban votes (less likely to have fixed land lines) and under-represents the 3million newly registered voters (nearly 10% of the eligible electorate).

All of this could, could add up to good news for Royal on Sunday night.

Also, of interest, for those who followed the controversy on reporting of exit polls by blogs in the 2004 US presidential election, check out the wry mockery of attempts to prevent this in France on the French blog "NRV".  

Away from polls (or pseudo-polls) I think Royal closed with one of her best weeks in months while Sarkozy literally tried to ride off into the sunset (on a white horse, no less!). For progressive politics in general, I think its important that she make the case strongly for some of the proposals that really could make a difference and which separate her out from Sarkozy -- notably increasing minimum wage, increasing small pensions, creating employment for entry-level workers and not merely cutting taxes wantonly. (The New Yorker story is the first one I've seen in print note that if Sarkozy were actually to cut taxes by 4% of GDP, he'd have to make massive layoffs in the public sector -- not just a hiring freeze.

The reality is that Sarkozy would not be able to make such cuts and would drive up the French debt even higher, with negative consequences for the rest of the Eurozone (and for his friends among US businessmen who are in favor of the weak dollar).

by desmoulins (gsb6@lycos.com) on Sat Apr 21st, 2007 at 06:15:16 PM EST
I have a question for those in France, especially outside Paris. Did anyone see any evidence on Saturday of the UMP's "72 heures pour decider"? Its clear they at least think they have in mind the once-vaunted Republican turnout operation of 04, but this kind of gotv has never been done before in French elections (because no one's ever had the time or resources to compile adequate voter files).

If the UMP has been doing this under the radar, it would mean Sarkozy might perform a few points better than expected.

If its just upper-class kids in blue t-shirts handing out brochures at the market, its not more than a slogan.

by desmoulins (gsb6@lycos.com) on Sat Apr 21st, 2007 at 06:19:28 PM EST
[ Parent ]


Occasional Series