Sep 252013

What’s the Matter with Germany?

At least in Germany, people begin to realise that Merkel may have painted herself into a corner by winning so gloriously (told you so first thing on Monday). While her Christian Democrats are by far the largest party group in the new Bundestag, she needs a coalition partner, but nobody wants to play.

kanzleramt in berlin What Happens If No One Wants to Govern with Merkel? Are We All Belgians Now?
Werner Kunz / Foter / CC BY-NC-SA

The Social Democrats are not exactly keen to form a coalition with Merkel. They came out of the last Grand Coalition (2005-2009) very badly damaged and have hardly recovered from that electoral blow. And a new CDU/SPD coalition would be not so ‘grand’ any more. In 2005, the SPD and the Christian Democrats were not so far apart in terms of votes won. This time, there is a 16 point gap between the two.

The important NRW state party, which represents roughly a quarter of the party’s total membership,  is positioning itself against a Not-So-Grand Coalition. The party’s left-wing opposes on principal grounds, and because they feel that this would further strengthen the Left party. The party leadership is officially stalling just a bit, saying the situation is open. Talks will be ‘ergebnisoffen’ – non-directive. And they are seriously telling the Greens that it’s a dirty job, so perhaps they should do it.

A CDU/Green coalition on the other hand, while not impossible, is unlikely. The idea was all the rage a couple of years ago, but it did not work well in the Länder. Moreover, the party has moved to the left during the campaign (something they are regretting now) and returned to a more polarised view of the political world. The party leadership has resigned over the slightly disappointing result. Some of them will return, but it is not yet clear what the balance of power within the party will look like, and any new leadership will find it extremely difficult to sell a coalition with Merkel to the rank-and-file, who have a de facto veto.

This may very well be a thinly disguised attempt to drive the price of a black-red (or black-green) coalition upwards. In 1998 and 2002, it took roughly a month to form a government. In 2005, SPD and CDU needed 55 days. In 2009, the FDP/CDU/CSU government was sworn in about 40 days after the election. But what will happen if both the Greens and the SPD refuse to play (it is still silly season, but nobody’s talking CDU/Left. Yet)?

Is a Minority Government Possible? Will There Be New Elections?

In Germany, there is always a procedure, and in this case, it is spelled out in the constitution, whose framers were obsessed with stability (for very good reasons). The new parliament will be convened on October 22, 30 days after the election. That is the constitutional maximum. With this first plenary meeting (parliamentary parties haven been holding business meetings since Monday), Merkel’s second term as Chancellor will end, and so will the tenure of her ministers. The Federal President will however ask her ‘to continue to manage Germany’s affairs’ until a successor is appointed, and she is obliged to heed that request. So are the ministers. There is no ‘gap’: We’ll always have someone who tells us what to do.

The president will then go to parliament and propose a successor. But ‘then’ is relative. Interestingly, the constitution which is usually very precise, does not stipulate a time-frame. Leading commentators say the time-frame must be ‘appropriate’. Four weeks are ok. Six weeks would be ok, too, I think. But how much longer?

In theory, the president could come up with any proposal, but in practice, his suggestion has always been based on a viable coalition agreement between the parties, since his proposal must be confirmed by more than one half of its members (as opposed to more than half the members happening to be in the chamber at that point). The constitution is extremely wary of unstable majorities, let alone minority government.

Should the president’s candidate not be elected (this has never happened in the past), parliament has 14 days to make up their minds. During this period, they can elect anyone who manages to get the votes of more than half of the members without the president having any say.

Failing this, parliament will have one last vote on the Chancellor. Under this rule, the person receiving the most votes wins. If, by some happy coincidence, the number of these votes exceeds the number of half the members of parliament, the new Chancellor is sworn in and appointed. If it is less, the president has a choice: Within seven days, he can either appoint the Chancellor to lead a minority government or trigger new elections. Again, the choice is his in theory, but in practice, he would consult with the parties.

So where does that leave us?

Should both the Greens and the SPD refuse to join a Merkel government, they could still form a coalition with the Left or negotiate a toleration arrangement. But that seems unlikely, since the political costs would be very high, and it is not clear that all leftist MPs would vote for him. A small group of SPD and/or Green MPs could vote for Merkel to get on with it, without the parties entering a formal coalition, or one of the two parties could officially decide to tolerate her. As long as that would give her a majority in the inaugural vote, the president would have to appoint her. Or the parties could agree on having another election, with unknown consequences (AfD and/or FDP entering parliament, losses for all but the Left?).

Given these prospects, a CDU/SPD coalition will probably look like the lesser evil once everybody has calmed down a bit. But that might take some time.

Sep 192013
thank you for more than 28 000 clicks on that photo dome of the reichstag building la cupula del reichstag reichstagskuppel berlin How to Pool the Polls
alles-schlumpf / Foter / CC BY-NC-SA

Why Would I Want to Pool the Polls?

Pre-election polls are noisy for a number of reasons. First, there is sampling error. For n=1000, the confidence interval for a party whose true support is 40 per cent ranges from 37 to 43 per cent, which is more than most people would think. And this assuming simple random sampling. For multi-stage sampling, you could end up with one to two extra points at each end. Then there are house effects: Pollsters dress up their raw figures and different ways, use different sampling frames and slightly different modes and questions. And finally, political events and media coverage on the day of the poll will have effects, especially early on when many voters are undecided.

Combining results from different polls is one obvious strategy to deal with these problems: The combined sample size is bigger, and there is hope that the various sources of bias might offset each other. Hopefully.

Where Do the Data Come From?

The very useful site publishes margins from seven large German pollsters. Excluding INSA (they use an online-access panel), I check this site regularly and generate a data set from it that you can download here. To companies post (relatively) raw data, which I find preferable. What the others do to their figures, we cannot know.

How Does It Work?

The most straightforward idea in poll-pooling is calculating a moving (and possibly weighted). A more principled approach is model-based. My model borrows heavily from Simon Jackman’s (2005) paper and from Chris Hanretty’s application of a similar model to Italy, but differs in some respects. First, I treat the polls as draws from a multinomial distribution to account for Germany’s moderate multi-partyism. The parameters of this distribution depend on the relative strength of latent support for each party. Modelling the results as multinomial implies the constraint that the estimated shares must some to unity, which is useful. Second, like Jackman and Hanretty, I assume that latent support for each party follows a random walk (today’s support is yesterdays support plus a random quantity), but I allow for a drift: a linear trend in latent support over the course of the campaign. Third, I assign each poll to a week, because there are relatively few polls, and field-times are relatively long. Put differently, I assume that public opinion moves from week to week (but not from day to day).

The model estimates latent party support since January 2013 and makes predictions for the outcome of the election. The code (R & Bugs) is here.

Does It Work?

Honestly, I have no idea. This is work in progress, so take the findings with a pinch of salt.

What Could Possibly Go Wrong?

Everything? Anything? The most obviously dubious assumption of the model is that the polls are unbiased on average. Latent linear trends are a close second.

Sep 142013

I finally got around to clean/update the code for my model that pools the 2013 German pre-election polls, estimates latent party support and comes up with the most marvellous predictions (TM). You may download it (complete with the most recent version of the polling data) from my dataverse. If you spot a bug or some other problem, please drop me a line.

I will be back soon with the latest polls.

Sep 032013
pirates ahoy 187x300 Do Minor Parties Employ Models?

Is this serious?

Many of the activities parties would have done in-house during a campaign in bygone times have now been outsourced to agencies. This was nicely illustrated last week by a mildly embarrassing incident: The liberal FDP and the right-wing extremist NPD were using the same stock video footage of a happy family. So was a Finnish dairy company.

But some of the smaller parties still put people on their posters who are most probably not professional models. Today’s exhibit is widely deployed by the Pirate party. The caption reads (in my clunky translation) “Not for sale. Just eligible”. The latter does clearly not apply to the two youngsters. But if their elderly companion was a real candidate, shouldn’t they put her name on the poster? So: Are these real rank-and-file party members, or is this another stock photo otherwise used for … whatever?

Sep 012013

The Polls

majorparties week 35 300x225 German Bundestag Election: Six New Polls, Little Change

Support for the Major German Parties, Estimates and Predictions (Week 35, 2013)

Exactly three weeks before the 18th Bundestag election, it’s time for another look at the polls. This weekend brings six new entries: One late result from week 33 that was only published a week ago, three polls from week 34, and two that were conducted this week, with fieldwork done from Monday/Tuesday to Wednesday. For all purposes and intents, that means that any possible fallout from the Western (non-)intervention in Syria will not be reflected in the polls.

Raw Figures, Estimates and Predictions

As always, there is a good deal of variation in the published figures. The range for Merkel’s Christian Democratcs, for example, is 41 to 46 per cent. But for what it is worth, the model is ever more confident about the outcome of the election: The estimated probability of victory for the governing coalition is now 85 per cent (up from 78 per cent) even if one ignores tactical voting by CDU supporters. If this “loan vote” is factored in, the probability of a coalition victory is 94 per cent (up from 90). Unsurprisingly, the probability of a Red-Green majority is still estimated as zero.

minorparties week 33 300x225 German Bundestag Election: Six New Polls, Little Change

Support for the Minor German Parties, Estimates and Predictions (Week 35, 2013)

The  one remarkable change is the modest slump in support for the Greens, which have lost about two points over the last four weeks and are now well below their peak support of about 15 per cent in March. The slow upward trend of the Liberals is unbroken, and the Left is safely above the electoral threshold. Support for the two major parties is perfectly stable.

Since my interest here is (mostly) academic, I also began comparing past predictions (from week 33) with current estimates. The differences are small, but there is one interesting exception: Support for the Greens is now estimated to be 0.8 points lower than it should have been, given the information that was available two weeks ago. So it would seem that their support is indeed suffering from some random shock.

The Outlook

Today is the day of the televised debate between Steinbrück and Merkel (in Germany, known as “the Duel”). While we are professionally obliged to watch it, I don’t think that it will make much of a difference. Both candidates are extremely well known knowns. I also don’t think that Syria will matter for this campaign.

Have I just shot myself in the foot? Probably. Come back next week for the latest batch of surveys.

Aug 242013

The State of Play, Four Weeks Before the Election

Last week’s post on Merkel’s very good chances to win a third term created a bit of a stir. This week, I’m back with nine new polls (conducted between August 6 and August 19 by six different companies), which all point into the same direction.

What the Pollster Saw

On average, polls are in the field for five days (with a standard deviation of three days), so I continue to anchor each poll to a specific week in the calendar. Along with the raw data, the graphs show estimates for the true support for each party over 32 weeks, starting from Monday, the 31st of December. Eight of the new polls cover week 31 and week 32, while one is a late addition to estimate for week 30.

majorparties week 33 300x225 Nine New Polls Say Merkel Will Still Be Chancellor in October

Estimated/predicted Support for Major German Parties (2013 election). Click for Larger Image.


Support for Merkel’s Christian Democrats is between 39 and 47 per cent. The model, which accounts for previous levels of party support and variation across pollsters, puts them at 41 per cent. Findings for the major opposition party, the Social Democrats, are less variable at 22 to 25 per cent. The model places them at the upper limit of these current polls.

Results for the Greens are even more unanimous (12-13.5 per cent). The model agrees, confirming that their support has come down a tick or two over the last weeks.

The same cannot be said for the Left, which is almost static at seven per cent (polls: 6-8.1). That is well below their 2009 result, but also well above the electoral threshold of five per cent.

Finally, for the Liberals, Merkel’s coalition partner, things have improved ever so slightly. While the polls vary from three to seven per cent, the Liberals’ true level of support is currently estimated at 5.2 per cent. More importantly, after months of continuous near-death experiences, there seems to be an upward trend.

minorparties week 33 300x225 Nine New Polls Say Merkel Will Still Be Chancellor in October

Estimated/predicted Support for Smaller German Parties (2013 election). Click for Larger Image.


What Does That Mean for September 22 and Beyond?

This is my first shot at pooling the pre-election polls, so all predictions should be taken with a very generous pinch of salt. The model is possibly misspecified and rests on an number of questionable assumptions. The most obviously problematic amongst these is that polls are, on average, unbiased over the whole January-September timeframe. But hey, this is a blog, so let’s ignore this (and all other) problems for a second and believe that the trend-lines and credible intervals for the next four weeks are indeed credible.

Once we make this leap of faith, the probability of a return to a Red-Green coalition is approximately zero. Amongst 10000 simulations of week 38/39 (the election is on a Sunday), there is not a single one that gives a parliamentary majority to this prospective coalition.

The FDP, on the other hand, makes it past the electoral threshold in 83 per cent of my simulations, and in 78 per cent, there is a parliamentary majority for the present coalition. The true probability will be higher, as some CDU supporters will vote strategically for the FDP to help them across the threshold. If we assume that this behaviour is virtually guaranteed to succeed (it would be enough if about one in 40 CDU would cast a “loan vote”), the probability of a majority for the present coalition goes up to 90 per cent.

Put differently, the probability of a Red-Red-Green coalition (SPD, Left, Greens) is between 22 per cent (no loan votes for FDP) and 10 per cent (loan vote strategy works perfectly). But even if there was a majority for the three opposition parties, a coalition (or rather a toleration arrangement with the Left) would be highly unlikely (say p=0.1), making a Grand Coalition led by the CDU the default option. That again means that the probability of any government not being headed by the present chancellor is between one and two per cent (down from four per cent last week).

What About the Smaller Parties (AfD, Pirates, etc.)

For several months, most pollsters did not publish separate results for smaller parties such as the eurosceptic AfD or the internet-centric Pirates. Some have resumed giving itemized counts for “other” parties, and it currently seems safe to assume that neither will enter parliament. If they did, the Pirates would probably take away votes from the left parties, whereas the AfD would most likely weaken the two major parties. In either case, a Grand Coalition would become more likely.

What Everyone Else Thinks

The July issue of PS has two short pieces on forecasting models for the September election. Both pick Merkel as winner. So do Bundewahltrend (average over the six most recent polls), pollytix (weighted average of 15 most recent polls), and wahlistik (a poll aggregator run for the Zeit weekly). Las weekend, politicians in both major parties began floating the idea of a Grand Coalition, but given the latest polls, four more years of centre-right government seems to be the most likely option by far.

Stay Tuned

This post comes with lots of health warnings attached. In the past, forecasts have failed, faces have turned red, majorities have collapsed well before election day. I’ll be back once I have collected the next batch of polls.

Jun 112012

Colleague Rainbow Murray is in Paris to do a little bit of observing. Her personal account of the count in one Parisian ward is quite intriguing. I had no idea that each candidate has to provides their own ballot paper. Not a very green thing, it would seem. Art Goldhammer puts Laurent Joffrin’s comment on Mélenchon’s failture/LePen’s success in perspective. ‘We all tend to overinterpret the results of elections’. Can’t argue with that. Meanwhile, Matt Goodwin ponders the question if Marine LePen’s ‘detoxified’ version of her father’s Front National is serving once more the blueprint for the (West) European Extreme Right, with Greece providing the counterpoint.And my own thoughts? Looking back, perhaps the most remarkable fact is how much our collective excitement has waned since the presidential election, although legislative elections really arereally important. Is this just because hundreds of multi-person races can simply not compete with the drama of the shoot-out between Sarkozy and Hollande, or just another piece of evidence of the internet’s detrimental effects on our attention spans? Continue reading »

May 112012
Local deviations from regional voting patterns

A mere 2.75 years after the fact, the Definitive Volume (TM) on the German Federal Election of 2009 is almost (almost!) ready to go to the printers’. And so is our chapter on East-West differences in German voting behaviour, which is vintage before it is even out (Pirate party, anyone?). Obviously, the details are becoming more and more blurry, so going through the proofs actually made for a pleasant read.

Political Science is the magpie amongst the social sciences, which borrows heavily from other disciplines. These days, many political scientists are actually failed economists (even more failed economists are actually economists, however). I used to think of myself as a failed sociologist, but reading the proofs it dawned on me that I might actually aspire to become a failed geographer.

regional voting 2009 221x300 Mapping local deviations from regional voting patterns in Germany

Local deviations from regional voting patterns

On particular nice map that should have been discussed more thoroughly in the paper shows the local deviation from regional voting patterns. Yes, you read that right: I calculate an index (basically Pedersen’s) that summarises local (i.e. district level) deviations from the regional (East vs West) result and roll that into a choropleth.  This way, it is easy to see how heterogeneous the two regions really are. Most striking (in my view) is the difference between Bavaria and the other Western Länder, which is of course a result of the CSU’s still relatively strong position. The PDS/Left party’s stronghold over the eastern districts of Berlin is clearly visible, too.