It’s a truth universally acknowledged that the ur-podcast must have been two blokes in the pub going on about politics for hours. Thankfully, the format has evolved somewhat. So if you are interested in German Politics, why not listen to Cas Mudde and me discussing the 2021 elections in just over 30 minutes?
A week ago, Germany’s alliance of Christian Democratic parties posted its worst national election result ever: they lost nine points on their (already pretty disappointing) 2017 result. For what it’s worth, these results are in line with a long-term downward trend that was briefly reversed in the 2005 and 2009 elections.
The Social Democrats emerged as the election’s big winner, partly because they improved on their abysmal 2017 result and emerged as the biggest party, but also because for a long time, the polls had suggested that they would fall well below 20 per cent. But from a historical perspective, their result is pretty weak. All this is hardly surprising, because party identification, or at least identification with the former main parties, has been in decline for a long, long time.
Because no-one wants to form a coalition with the AfD, and because the Left is almost as isolated and moreover to weak to be a necessary part of a coalition, that leaves us with three options: the “traffic light” (SPD, Greens, FDP) and “Jamaica” (CDU, Greens, FDP) coalitions, and a reversed “Grand Coalition” (SPD/CDU). The latter has some limited use as leverage for the bigger parties and insurance against minority government and fresh elections, but is deeply unpopular with voters and also the respective parties.
Jamaica looked like a non-starter on election night and has become even less likely: the CDU (and to a much lesser degree the CSU) is in disarray, fighting over its leadership and future course. This would make them a very awkward partner for the smaller parties. Moreover, in the days after the election, the discourse has focused on the SPD winning and the CDU losing the election. Laschet, the CDU/CSU’s candidate for the chancellorship, always had terrible ratings and is widely seen as one important factor contributing to the Christian Democrats’ disaster. Installing him in the chancellery would hardly chime with the Greens’ and FDP’s message of reform and modernisation.
The most likely result of this is the “traffic light”, which would be Germany’s most secular coalition since (probably) the 1970s. Why is this? While this may change in the future, Germany is currently a typical “religious world” country, where conflict over the role of religion in public life is reflected in the structure of the party system. The name of Germany’s dominant party gives a strong hint: the religious/secular cleavage is embodied in the CDU/CSU. And this is not just tradition and the proverbial “C” in their name: At least in the 2009-13 Bundestag, MPs for the Christian Democrats had a surprisingly large number of affiliations with religious groups and institutions. However, the Greens scored almost as high on this measure.
|Party||mean number of affiliations|
An analysis of the crucial free vote that legalised Preimplantation Genetic Testing in Germany shows that even after controlling for these remarkable differences, MPs for the SPD and particularly the FDP were much more likely to support legalisation. This is in line with secular traditions in both parties. The FDP in particular has consistently voted for more liberal (again, the hint is in the name) bioethical rules.
On the other hand, the voting behaviour of Green MPs did not differ significantly from that of their CDU counterparts.1 Again, this is not terribly surprising: while the Greens are by and large a left-libertarian outfit, Christian movements for peace and the protection of the environment are part of their heritage. Winfried Kretschmann, their only Minister President, is a devout Catholic who runs a coalition with the CDU and likes to stress that they operate on a base of common values. The long-standing leader of their parliamentary party, Katrin Göring-Eckardt, studied theology (although she never graduated) and was the top lay person involved in the leadership of Germany’s Protestant Church as president of its synod from 2009-13. She also is in favour of compulsory religious education in schools and supports the (remaining) exemptions from workers’ rights for church-run hospitals, care-homes and kindergartens.
Of course, the numbers above are from 2011. Many new MPs for the Greens are very young, and I have no idea how important that Christian strain is for the new parliamentary party. Either way, with the Christian Democrats out of government, the new government will by definition be more secular than the last four, and bringing in the FDP might make them more secular than the Schröder governments. If and how this will affect actual policies is, of course, a different question that is way too nuanced for a blog post.
Conversely, the CSU MPs displayed more restrictive preferences than anyone else.
Over on twitter, people have been quite disparaging, pointing out that Allensbach has been perceived as close to the Christian Democrats for decades, uses methods that are now seen as outmoded, and considerably overestimated support for the CDU/CSU in 2017. While I tend to agree, I thought I might as well pass the time by having a look at how well the major pollsters did in 2017 using, you know, actual data.
For multi-party elections, the sum of squared differences between a pre-election poll and the final vote shares is a crude but intuitively plausible measure of accuracy. Thankfully, the good folks over at wahlrecht.de collect headline findings from the major houses going back all the way to the late 1990s. So I plugged the last surveys published immediately before the 2017 election into this shiny table and calculated the differences.
|CDU/CSU||SPD||Greens||FDP||Left||AfD||Others||Sum of squared differences|
The numbers are not quite what I expected. Kantar/Emnid and FGW, who have been in the business for ages, are placed second and third with very similar deviations from the result. Fellow household names Allensbach, Forsa, and Infratest form a second, slightly worse performing but very homogeneous cluster. GMS and YouGov, on the other hand, were most (and similarly) off.
The best performer, and this is the surprising part, was INSA, who are, let’s say, are slightly less respected for both political and methodological reasons. According to wahlrecht.de, their final poll in 2017 was published on September 22, just two days before the election, with the data being collected on September 21/22. So it could be that they were simply interviewing closer to the event than the others and picked up some last minute swing away from the Christian Democrats. Or perhaps they were just lucky.
Coming back to Allensbach, the table shows that everyone overestimated the CDU/CSU (herding, anyone?), and that Allensbach was by no means an outlier. So if past (squared) performance can serve as a guide for the present, there is no particular reason to rubbish this latest Allensbach poll.
Does radical right success lead to mainstream re-positioning?
Radical right parties have existed for decades now, but most of them are still seen as challengers, because they aim to disrupt the (liberal democratic) consensus in their respective societies. Existing parties can react by digging their heels in, or by accommodation. As I have argued elsewhere, their best bet might even be to ignore the challenge. But if they chose accommodation, it is by no means clear whether this position shift was caused by the emergence of a radical right party: mainstream parties could simply react to the perceived shift in the preferences of their electorate. They might even try to preempt the rise of the radical right.
One new and very cool paper that tries to shed some light on these questions of causality is Abou-Chadi & Krause (2020), which we read this summer.
Abou-Chadi, T., & Krause, W. (2020). The causal effect of radical right success on mainstream parties” policy positions: a regression discontinuity approach. British Journal of Political Science, 50(3), 829–847. http://dx.doi.org/10.1017/S0007123418000029
What we liked
My students had rarely seen difference-in-difference and discontinuity designs in the wild. They were hugely impressed by this elegant (and quite conservative) setup. Obviously, they thought this was interesting and politically relevant research. More generally, they liked the idea (new to many) of investigating inter-party strategic behaviour.
What we did not like so much
There was not much we did not like much. Students wondered whether the assumption of continuity around the threshold is defensible, and whether the various parties are comparable. More importantly, they suggested that in the real world, parties might chose mixed strategies (e.g. introduce tougher rules on immigration to drive down numbers whilst also trying to reduce the salience of immigration and going all-out liberal on other “cultural” issues). Finally, they would have liked the authors to talk even more about the political implications of their findings.
Something good in everything?
Could radical right-wing populism be a (whispers) good thing? Of course it all depends on what we mean by “good”. Backlund and Jungar have a modest proposal: they suggest that radical right success could improve the representation of policy preferences in parliament. Using data from both expert and voter surveys in ten (West) European countries, they find that radical right parties occupy an almost unique position (both against immigration and against the European Union). They provide a good fit for their voters in this respect. On the other hand, most radical right parties are socially conservative (and more specifically homophobic), which many of their voters are not.
Backlund, A., & Jungar, A. (2019). Populist radical right party-voter policy representation in western europe. Representation, 55(4), 393–413. http://dx.doi.org/10.1080/00344893.2019.1674911
What we liked
Students liked that Backlund und Jungar disaggregate “left” and “right” into policy positions. Most of them had also never seen work based on expert surveys (!) and were duly impressed. They found the derivation of the hypotheses convincing and praised the effective tables and graphs. Students also said that a lot of useful information was found in the appendix, then realised that this was a double-edged compliment.
What we did not like so much
Students were a bit disappointed that populism did not really feature while nativism was really prominent in the analysis. Given the Muddean concept of populism and the Chapel Hill-based measurement effort, this is hardly surprising, but they were still somewhat disappointed. They also said that they would have liked to see data for more countries as well as a clearer rationale for country selection, and harboured doubts about the validity/comparability of the gay rights-items in the EES/CHES. Finally (and this may well be a Mainz thing), they said that representation/representative and responsive(ness) were used more or less interchangeably by the authors. While I’m not sure whether this is really true, I’m happy to see that my students strife for conceptual clarity (at least as far as other people’s work is concerned). Having said that, we thought that this is a fresh, almost unique take.
Not a huge fan of Google/Alphabet (the company), but I really like Google Scholar. It brings a certain kind of democratisation to science. Unlike commercial giants Web of Science/Clarivate and Scopus/Elsevier (vade retro satanas!), access is free (yes I know, we pay with our data). Moreover, it covers a much wider range of sources than just elite journals, and it happily points one towards pre-prints, institutional repositories and other ungated sources of research.
And Google Scholar (being an aspect of the all-knowing if not all-loving Google Deity) has a knack for uncovering stuff that could (or should) interest me. Occasionally, I get an email suggesting that I should “follow” colleagues X or Y, i.e. get updates about their new publications. As one would expect from Google, these suggestions are uncannily accurate.
A couple of days ago, I received one such recommendation that was also accurate, but uncanny in a completely different way. In this specific case, I’m sure that there will be no stream of new publications that I could follow, because the colleague in question died a couple of months ago, quite young and unexpectedly.
Which, for one morbid moment, made me think of the digital clutter/legacies that we are accumulating every day: the preprints, working papers, published work in journals and ebooks, not to mention the podcasts, video footage of lectures, blogposts (sic!) and everything we are posting on social media. That much of this will be around after our demise, at least for a time, is of course normal. Science must be cumulative, what with the standing on shoulders of giants and all that. What I had not fully appreciated up to that moment is that algorithms will happily continue to harvest our surviving digital twins, whether they are twitter handles or Google Scholar profiles, sending updates about us to other persons (who might be dead as well) until their parent companies go bust or lose interest in that specific service. Goose over my grave etc.
Just for the fun of it, I have turned my recent article on Germans’ support for pre-implantation genetic diagnostics (PGD) into a short video. If you want to know why Germans would rather have laws along the lines of Belgium or the UK, and why they don’t get them, but do not want to read the full 8,000 piece, look no further – just click on the video below.
I love what 60 years of economics envy have done to the social sciences.
Something new about protest voting and the radical right?
Their story is roughly this: yes, radical right voters are dissatisfied, but their unhappiness is ideological. They crave even tougher immigration policies (and possibly a more generally illiberal setup of politics and society).
While Wouter and friends were writing about West European countries of the 1990s, their core findings have been confirmed time and again with newer data. End of story.
So I was quite intrigued when I saw this new paper:
Cohen, D. (2020). Between Strategy and Protest. How Policy Demand, Political Dissatisfaction and Strategic Incentives Matter for Far-Right Voting. Political Science Research and Methods, 8(4), 662–676. http://dx.doi.org/10.1017/psrm.2019.21
Cohen’s argument is that both policy demands and discontent are relevant motives, whose relative importance depends on the circumstances, i.e. radical right representation in parliament and government participation. What’s also novel is that the students tasked with introducing the text got in touch with Cohen, who sent us a video showcasing the article’s highlights and some of his other research. That’s pandemic political science for you.
What we liked
This is something which is often talked about but that is rarely implemented in practice. Perhaps the most obvious distinction is between situations where the radical right is in office/opposition (the latter still very much the rule).
What we did not like so much
Our main criticism was that legislative strength is measured by seat share in the last national (first order) election, whereas the dependent variable is voting behaviour in European (second order) elections. This seems conceptually dubious and also transforms parties operating under non-proportional national electoral systems (think UKIP and Rassemblement National) into outliers.
But even so, we liked the focus on the political context and opposition/government party status.
… but can’t be arsed to insert its name into the email template