It’s a tall order…but surely writing about German elections with statistics must beat it?
But there’s a German election coming up, although, as Der Spiegel points out, you might not have noticed, as both major parties are secretly quite pleased with the current situation. Polling data is here. Angela Merkel has spent the period since her triumph of 2005 governing well to the left of her party and being a quietly effective foreign-policy chancellor, just as we predicted; the Social Democrats have been struggling, as a result, to retain an independent profile, but (from their point of view) at least they’re in government, and paradoxically the main gainers from the economic crisis have been the FDP, the spokesmen for classical liberalism.
Their leader – still Guido Westerwelle after all these years – is behaving a little strangely in public, saying very frequently that he doesn’t believe there is any chance of the so-called traffic light coalition with the Social Democrats and Greens, but not saying that he rules it out. If the polls stay as they are, this would be the only chance of the Left taking power; but, of course, this is a huge assumption, especially in the light of their surge during the 2005 campaign. With the CDU on 37%, it’s essentially assumed that they are running up against demographic limits – a typically AFOE point, but a good one.
The all-time record conservative share of the vote is 39.7%, achieved in 1957, but more to the point, even another point-and-a-half would be more than one standard deviation from the long-run average, that is to say about a 3 in 10 chance. Theoretically, there is a 5% chance of getting to 42%, but if Konrad Adenauer couldn’t get over 40% in booming 1957 it’s probably even more unlikely that Merkel will in 2009. In fact, one thing that this little statistical exercise shows is that German party vote shares are very stable indeed – the SPD’s share of the vote has greater variance, but not that much.
So there is not much space for the rightwing vote to grow; and the Left Party is apparently stuck just under 10%. The strong Liberal showing – 15% in the current polls – suggests that the right could hope to form a new coalition without the Social Democrats, which would hold 50% of the vote. At the moment the only way the Social Democrats could checkmate this would be to get the Liberals and Greens into a coalition – the Left Party and the Greens wouldn’t be enough. This all assumes that nothing else changes, however; if the Left-Left-Green option was possible, all the coalition calculations would be altered, as the Liberals would face a serious risk of being left out in the cold. So what would it take to make it happen?
At the moment, the LLG coalition adds up to 46%, the “bourgeois” (i.e. CDU/FDP) option to 50%; so they need four percentage points to cross this strategic threshold. In fact, in so far as they are fighting a zero sum game, they might need fewer. The SPD’s share of vote in the current polls is on 23% – a shockingly low figure. In fact, based on the SPD’s historical vote shares, this would in itself be approaching a 1 in 100 event. Even taking account of the Left Party breakaway, the party polled just under the historical average last time out; and the 95% probability level corresponds to a vote share of 27.3%, which would put them back in the game. Actually, there doesn’t appear to be much covariance at all between the Left Party and SPD shares; this fits the explanation that the Left is still mostly the ex-PDS.
So I’m going to forecast that, even if the SPD looks down and out now, there’s an excellent chance of them being in with a chance on the night.