German election, 2013

As I type this, it is about 90 minutes from poll closing in Germany. BBC reports that German exit polls are “notoriously accurate” and that we should know the outcome of the election within two minutes of closing.

This is an open planting hole about the election.

47 thoughts on “German election, 2013

  1. Isn’t 6:00 p.m., local time, an unusually early poll-closing time? Or is it not so early for those sensible countries that vote on days when most people do not have to work?

  2. It was 8 pm in Australia for a while but the numbers voting between 6 and 8 were so small it was eventually reduced.

  3. I’m not sure if this is best placed in the this thread of the one of the MMP change, but here we go…

    What happens if, and I admit that this isn’t very likely, a party wins so many other direct mandates that one or more other parties cannot fill their seat entitlement in the expanded Bundestag?

  4. With those numbers, it would appear that the only reasonably plausible coalitions are the Big Two or the three parties on the left. Could the SPD swallow its pride and try to cut a deal with the Left? Could Merkel try to make a deal with the Greens? Would that even have a majority?

  5. Mark, prior to the election there was actually some concern that the parties would not have enough candidates to fill all the seats in an expanded Bundestag, and not only under the scenario you outlined: if one of the smaller parties somehow ends up over-represented in the initial, state-level allocation of seats, it would have a cascading effect, as the larger parties would have to be compensated as well in order to achieve a fully proportional seat distribution.

    At any rate, I seem to recall – but I’m by no means certain about this – that if a party doesn’t have enough list candidates to fill its corresponding seat allocation, it forfeits the unfilled mandates.

  6. Hadn’t thought about major parties suffering from shortages, but then again I don’t think many suspect minor parties winning a great deal of direct mandates.

    Allowing underhangs to stand makes the most sense, but somehow I feel, if one should ever happen, that someone will sue because of the disproportionality. Hopefully the ruling will be along the lines of “Oh well, run more candidates.”

  7. Mark, the second scenario I outlined does not involve minor parties winning direct mandates, but rather winning more list seats in the initial state-by-state PR seat allocation than they would be otherwise entitled to.

    Specifically, the simulation of Bundestag election results since 1994 under the new electoral system points out to at least one instance where this would have taken place: in 1998, PDS would have been entitled to a minimum of 40 of 661 seats (6.1%) with just 5.4% of the second votes cast for qualifying parties, and in turn this would have increased the size of the Bundestag to a whooping 729 seats.

    But back to today’s vote, both ARD and ZDF have made slight adjustments to their forecasts, which place AfD tantalizingly close to the five percent threshold. Meanwhile, FDP remains just behind AfD, and (most importantly) further away from the threshold.

  8. While exit poll numbers have been going back and forth on whether or not CDU/CSU will have an absolute Bundestag majority – provided both FDP and AfD fail to reach the five percent threshold – the Federal Returning Officer is now publishing live 2013 Bundestag election results in both German and English.

    So far it’s just constituency-level figures, mainly from Bavaria; from past experience results at higher levels won’t be shown until enough constituencies have reported in, but a cursory view suggests that both CDU/CSU and (to a lesser degree) SPD are up, while the established smaller parties are all down, with FDP taking a beating.

    AfD is for the most part just below five percent, but since it’s a new party it’s hard to discern a pattern, all the more so given that Bavaria is usually tough territory for smaller parties, given the CSU’s sheer dominance, and the additional competition on the center-right from the regionalist Free Voters.

  9. The German Constitutional Court ruled after the last election that the very minor disproportionality inherent in the mixed-member proportional system violated the constitution (as discussed in previous threads). The justices should be even more upset at this result, despite the reforms to the system that have made it even more proportional–that is, for those parties that clear the (sacrosanct?) 5% threshold.

    As of now it looks like both the FDP and the anti-euro AfD won more than 94% of the votes they would need to clear the threshold, but missed it. These two parties, adding up to nearly ten percent of the vote, will contribute a whole lot of disproportionality!

    Moreover, the total below-threshold vote might even reach 15%.

    • JD, I have no idea. I was simply referencing where this point came up in the other thread (to which you also contributed).

      Manuel, I have only sporadically checked the live results today, but it seemed to me as if the FDP had been drifting upward and the AfD downward as more votes were counted.

  10. I’m still not 100% convinced that AfD will end up below the five percent threshold – it appears to be doing quite well in the former GDR – but even if it makes to the Bundestag the share of “wasted” votes will in all likelihood remain in double-digit territory – a record for postwar Germany if I’m not mistaken.

    (My skepticism on AfD not making it to the Bundestag stems in part from memories of the 1983 Bundestag election, when the then-fledging Greens were initially forecast to fall just short of the five percent threshold, only to make it after all when all the votes were counted – and in the process deprive CDU/CSU of an absolute Bundestag majority.)

  11. Matthew, I noticed that earlier on as well. That said, with only 12 of 299 constituencies outstanding, FDP stands at 4.8% of the second vote, just slightly ahead of AfD’s 4.7%; unless either party performs exceptionally well in the remaining constituencies (seemingly unlikely from looking at the overall results of the states with pending constituencies), it appears now that neither will make it to the Bundestag, at least according to the preliminary results.

  12. I’ve always wondered why the Germans don’t allow joinder of lists in federal elections. (Joinder of allied lists in the same State, that is: of course the CDU and CSU, like the various branches of the SPD and other parties, join their State lists (maximum one per Land) nationwide to ensure national-level proportionality.
    Allowing joint lists would (1) encourage coalitions to be formed before, rather than after, election day (which I’d favour as long as it doesn’t mean you can’t vote for your preferred party in your electorate), and (2) end Germany’s quadrennial kabuki over whether the FDP will get 4.9 or 5.1%. Ironically, linking the FDP with the CDU-CSU (or, at times, the SPD) more closely would allow the larger party to compete more assertively against its smaller ally ( as the Libs and Nats do – occasionally – in Australia) rather than worrying about propping them up (at times even “lending” then voters to keep their heads above the electoral Plimsoll line).
    In the past I’d assumed the three-districts bypass was intended as a way to do this tacitly – since list joinder has been abolished in some PR-List countries after coming to be viewed as a bit “corrupt”. But the big German two don’t seem to practice the ceding of districts to their Free Democrat allies as a backup against their coalition narrowly missing out on as many as 32 seats. When John Key was perceived to have tried it in NZ, it was viewed as a bit improper. I do find that odd, since there is a consensus that, wherever possible, the likely or intended post-election legislative alliances should be made.known to the voters before they cast their ballots. (And when minor parties play coy before election say – yes, looking at YOU, Bob Brown Tas ’89 and Winston Peters NZ ’96 – they furnish opponents of PR with more ammunition.)

  13. JD, the court’s ruling isn’t specifically mentioned in the constitution, just like abortion rights are protected, but not specifically mentioned, by the American constitution.

  14. With the preliminary vote count completed a short while ago, my website’s Germany page now has federal- and state-level results of today’s vote.

    I can also confirm that the estimated distribution of mandates published on Wahlrecht would be the correct one for a 630-seat Bundestag, although I have yet to run the full distribution process.

  15. Glancing at the map, it appears unsurprisingly that the same results would have given the CDU a large majority when run using single member districts. The CSU seems to have won nearly as many if not more districts than the SPD. The other minor parties won four districts, with three PDS and one Green member being reelected in east Berlin.

    The interesting thing about this result is that the near miss of two parties generally being perceived as being on the right in clearing the 5% threshold will produce a narrow Bundestag majority for the left. Essentially, the German system of proportional representation has delivered a reversed result, essentially due to the threshold.

  16. Ed, SPD actually won more direct mandates than CSU, while The Left won four, all in Berlin; my website’s Germany page now has the official, preliminary seat totals.

    Incidentally, in the territory of the old, pre-reunification FRG (excluding West Berlin), CDU/CSU won a small absolute majority of seats, and FDP managed to win 5.2% of the vote; however, the party’s disastrous showing in the old GDR minus East Berlin – 2.7%, lower than even the far-right NPD – did them in. However, in the case of AfD it was the opposite: the party did better in the East (5.9%) than in the West (4.4%).

  17. Some of this modern European suspicion of apparentement of lists is reflected in the requirement in some countries’ election laws that, eg, the threshold is 5% for a single party but 8% for a coalition (Poland and/or Czech Republic?). Greece in the 1980s had a sharply rising series of thresholds – something like 10% for a single party to win compensatory seats, 16% for an alliance of two parties, and 25% for a coalition of three of more.
    I suppose these safeguard against the perceived danger that a Ricky Muir type candidate could win a seat on one-tenth of a quota just by heading the highest-polling list in a dozen-party alliance.
    A possible safeguard would be an absolute threshold of 0.5 quota for any list. This might mean, eg, that a 12-party coalition that po lls nearly six quotas in total, nonetheless wins no seats because none of its member parties exceeds 0.49 quota.

  18. Chris @20: I can understand that, but the question remains from what provision they interpreted that requirement. From the mixed reports I’m hearing, combined with my reading of the constitution, it’s very unclear if the Court even said anything binding on proportionality or not.

  19. I think the court’s ruling were along the lines of, “If you’re having a PR election, the results should be proportional.” I don’t know if my interpretation of the court’s ruling was accurate. I certainly don’t know if the court would have been okay if they went to non-proportional system, such as going to parallel voting or chucking the list seats and having FPTP.

    The specific case started from a delayed election in 2005 (or was it 2009?). At any rate, a candidate died and one district’s vote was delayed. Because everyone already knew what the results were from the 298 other districts, voters were being specifically told NOT to vote for the SPD because if SPD picked up too many votes in that district it would lose seats. That was the major problem the courts had with the electoral system and mandated changes to overcome it. It is possible that they saw the same problem with the modified system. In that case they approved the current system because a party can no longer loose influence when it gains votes.

  20. Like Britain’s complicated system of first-past-the-post voting, Germany’s MMP is vulnerable to reversal of the majority will through splitting the vote. However, the _way_ it’s split matters profoundly. Under FPTP, you lose if you divide too evenly (eg, UK 1983, South Korea 1988, US 1912) because then it’s not clear to voters whom they should swing behind and vote tactically for.
    Under MMP, by contrast, the three left-wing German parties divided their votes roughly evenly (in a ratio of 4:1:1) which means they all cleared the 5% threshold. The CDU-CSU, unfortunately for Merkel, achieved too lopsided a plurality – large enough to sweep the districts by a nearly 75% majority, but not big enough to reach 50%.
    Paradoxically, Merkel would have a majority – for her coalition, if not for her single party (well, 1.5 parties!) – had she implored all CDU-CSU supporters surnamed A to D to give their list votes to the FDP instead. And ironically, an electoral rule (the threshold) designed to favour larger parties over smaller has enabled an alliance of three quite separate parties (who were unable to coordinate their candidate nominations, or even their tactical voting, enough to come first in more than with 21% of the districts) to defeat a plurality party with a 16-point lead over its largest rival. Pure South African style PR _or_ FPTP in SMDs would have given Merkel a majority again in the Bundestag.

  21. “… The West German Constitutional Court has examined the question of whether the five percent threshold of representation contained in the West German electoral law is compatible with the principle of proportional representation. In its positive judgment, the Court argued, however, that a barrier of five per cent should form the upper limit of such a threshold, and it barred any further change that would strengthen the ‘constraining effect’ of the electoral system as incompatible with the principle of proportional representation. Consequently, if the West German legislature would like to strengthen those effects – for example, by raising the five per cent barrier – it would first have to opt for the other principle of representation (ie, the principle of majority representation); only after having done so would the West German Bundestag [sic] be allowed to introduce a considerably higher threshold of representation…”

    – Dieter Nohlen, “Two Incompatible Principles of Representation,” in Arend Lijphart and Bernard Grofman (eds) Choosing an Electoral System, Issues and Alternatives (NY, Praeger, 1984), pp 83-89.

    I didn’t record the year or title of the decision (if indeed Nohlen footnoted it), but you can tell from the “West German” reference that the judgment was given several decades ago.

  22. @27 Pure South African PR, without a threshold, would have given the CDU an absolute majority only in coalition with another party.

  23. I’m travelling in Canada, and the media coverage of the election constantly refers to “Merkel’s victory” or even “Merkel’s sweeping victory” when its not even clear the CDU won’t be displaced by a coalition of the left-wing parties.

    And I can’t blame the miscoverage on the American media (its worse when you consider that Canadians should be able to understand how a multi-party parliamentary system works), which probably isn’t covering the German election at all, though I’m not sure if no coverage isn’t better than miscoverage anyway.

  24. Assuming that the SPD doesn’t break its repeated vows against working with the Left, how exactly will Merkel not be reelected as chancellor?

  25. As The Economist aptly put it, Merkel’s victory was a bittersweet triumph. To be certain, the Union parties’ share of the vote rose significantly and it’s now back to where it stood in 1994, but one shouldn’t overlook the fact that with FDP out of the Bundestag, her parliamentary majority is now gone.

    Speaking of yesterday’s election, had the vote been carried out under the old system that the Federal Constitutional Court struck down back in 2008, the outcome would not have been substantially different, as the distribution of Bundestag seats would have been as follows (assuming voters had cast their ballots in the same manner) :

    CDU – 246
    CSU – 53
    SPD – 182
    Die Linke – 61
    Grüne – 60

    CDU would have won four overhang seats but the three left-wing parties would still have a 303-299 majority in a 602-seat Bundestag. Moreover, the overhang seats would have had little impact on the overall proportionality of the outcome, with votes-to-seats rations fluctuating between 60,625 for CDU and 61,798 for SPD.

    Incidentally, under the new system CDU also had four overhang seats in the initial state-by-state distribution of seats from which party seat minimums are obtained, but the extra seats were not what caused the increase in the size of the Bundestag to 630 mandates; instead, the increase was triggered by the proportional over-representation of CSU in Bavaria at the expense of the below-threshold parties in that state (which polled a combined 18.6% of the second vote there). As such, CSU ended up with three extra list seats, and the other qualifying parties had to be awarded twenty-five adjustment seats in order to restore full proportionality.

  26. Manuel, that is very interesting! Do you think you could explain it in a bit more detail, perhaps as part of your coverage of it on your website? Specifically, how does a party win too many list seats?

  27. I think the CSU’s extra list seats are rather easy to explain. So many parties fell below the threshold that the CSU picked up all the vacant seats that “should” have gone to them in Bavaria. Unfortunately getting those Bavarian seats left the CSU overrepresented nationally, necessitating that everyone else receive extra seats.

    I think this sort of problem is something that is very easy to have in a system with regional parties, regional seat allocations and a requirement for national balancing seats to achieve national proportionality.

  28. Alan @29: “for her coalition, if not for her single party (well, 1.5 parties!)” – ie, the right-wing bloc including the FDP and possibly AFD (albeit while Euroskeptics would normally grudgingly back the centre-right over the left, Merkel would be even less palatable to them than Cameron is to UKIP).
    Chris @31: refer @30 and @32.

  29. Mark, that is part of the explanation, but there are other factors at play.

    Under the new Bundestag electoral system, and for the purposes of carrying out the initial, state-by-state distribution of seats to determine party seat minimums, every state is allocated a fixed number of seats based on its proportion of German inhabitants, relative to the entire country.

    This is quite different from the old electoral system, under which states were not guaranteed a fixed number of Bundestag mandates beyond their assigned constituency seats, and a state’s representation was dependent upon three factors, namely: voter turnout; votes cast for below-threshold parties; and overhang seats.

    Now, Bavaria had both a slightly below-average voter turnout rate, relative to the entire FRG (70.2%, as opposed to 71.5% for the country as a whole); and 2) as noted in a previous comment, an above-average proportion of votes cast for below-threshold parties. As such, there were 5,358,402 second votes cast for qualifying parties in the state, which constituted 14.5% of the 36,847,430 second votes won by qualifying parties at the federal level, and the state would have ended up with 87 Bundestag seats – 53 CSU, 21 SPD, 9 Grüne and 4 Die Linke – which would also happen to be 14.5% of the legislative body’s 602 seats under the old system.

    However, under the new system Bavaria, with 11,353,264 (15.3%) of the Federal Republic’s 74,324,165 German inhabitants, was assigned 92 of the Bundestag’s 598 seats, (15.4%); in the initial state-level distribution, 56 of these went to CSU, for a gain of three.

    Once the seat minimums were established (242 CDU, 56 CSU, 183 SPD, 61 Grüne and 60 Die Linke), it was determined that CSU had the lowest vote-to-minimum seats ratio of the five parties (58,438), and the Bundestag had to be increased to 630 seats so that CSU could reach its minimum of 56 mandates under the Sainte-Lagüe/Schepers method.

  30. Although I’ve already noted as much on my blog, it’s worth repeating that beyond establishing nationwide seat minimums for qualifying parties, the initial state-by-state distribution of seats is otherwise non-binding, and its results are discarded after the allocation of seats at the federal level. In fact, once party seats were distributed among their corresponding state lists on a party-by-party basis, Bavaria ended up with 91 of 630 Bundestag seats, or 14.4% of the total – fairly in line with the state’s proportion of second votes for qualifying parties.

    On a personal note, I must say I was surprised by the complexity of the new system upon learning its details: I would have thought it simpler to take what used to be the final nationwide seat allocation under the old system as the starting point for introducing adjustment seats; apparently SPD proposed an electoral reform along those lines, but it was rejected by the ruling coalition parties, even though it would have resulted in a Bundestag size increase of just nine seats had it been adopted for yesterday’s election.

  31. While giving some thought to the previously described problem with the initial state-by-state allocation of Bundestag seats, I have come across a simple tweak that could at least ameliorate the excessive inflation of the Bundestag.

    Specifically, I re-ran the seat distribution process in the manner currently prescribed by law, except that I used second vote totals for qualifying parties (instead of population figures) to assign fixed seat totals to the states. This makes sense since both under the old and new electoral systems, at the end of the day the proportion of total Bundestag seats each state receives is usually very close to its proportion of total second votes for qualifying parties, relative to the corresponding federal total.

    This seemingly minor change – which does not affect the proportionality of the overall distribution of mandates – resulted in a 615-seat Bundestag for this year’s election – that is, fifteen seats fewer than under the system as it currently stands.

    However, I also ran the change for the notional seat distribution in the 1998 Bundestag election under the new system, and in that case the difference was quite dramatic: using German national population figures in the first stage caused the size of the Bundestag to balloon from 656 to 729, but switching to second vote totals for qualifying parties brought it back down to 684 – a reduction of 45 seats.

    In this particular case, the Bundestag size inflation was triggered by the fact that on the basis of population, the states of the former GDR (excluding East Berlin) would have received a combined fixed allocation of 134 seats, whereas on the basis of second votes for qualifying parties the figure would have come down to 114 (due to a combination of lower turnout and a higher-than-average share of the vote for below-threshold parties); the twenty-seat difference would have worked in favor of PDS – whose vote is precisely concentrated in that area – causing the party to be over-represented, and triggering a cascading effect as the larger parties would have had to be compensated to restore full proportionality to the distribution of parliamentary mandates. However, the switch to second votes for the allocation of seats among the states would have eliminated the party’s over-representation, and the addition of 28 seats would have been triggered instead by 13 SPD overhang seats in the initial stage.

  32. I have now re-run the notional seat distribution under the new electoral system of every Bundestag election since 1994, first using German population totals for distributing seats among states in the first stage, and then repeating the calculation with qualifying party second vote totals; the results would be as follows (with the initial number of Bundestag seats shown in parentheses):

    1994 (656): 692->685 (-7)
    1998 (656): 729->684 (-45)
    2002 (598): 612->616 (+4)
    2005 (598): 625->627 (+2)
    2009 (598): 671->671 (-)
    2013 (598): 630->615 (-15)

    The switch would have brought no change in 2009 because that year adjustment seats would have been triggered by CDU overhang seats. However in 2005, it would have resulted in an extra overhang seat for SPD, which would have required two extra adjustment seats for other parties.

    In 2002 the change would have led to the Greens obtaining a couple extra list seats due to rounding at the state level, which would have reduced the minimum vote-to seats minus 0.5 ratio (from 73,217 to 72,749), and in turn would have resulted in four extra adjustment seats. Finally, in 1994 PDS over-representation in the old GDR (not as severe as in 1998 – see above – but still a factor) would have been eliminated, but there would have been more CDU overhang seats; even so, the Bundestag would have had seven fewer seats.

    Incidentally, 1994 is probably the only recent Bundestag poll in which the new electoral system might have changed the outcome had it been in place at the time, as the ruling CDU/CSU-FDP coalition majority would have been reduced from a manageable ten seats to a precarious two mandates (the switch from population to second votes would have increased it slightly, to three seats), and it’s uncertain if then-Chancellor Kohl would have chosen to continue governing with such a fragile mandate.

  33. An e-mail exchange with Dr. Martin Fehndrich – a member of the team – quickly brought up a problem I had not considered when switching from naturalized population to qualifying second votes for the lower-tier, non-binding distribution of seats among the states: it could trigger negative voting weight.

    Dr. Fehndrich did not furnish any examples, but I was able to reproduce the phenomenon by distributing seats among the states using qualifying party second votes, then distributing the seats in each state among the parties, and finally bringing down the CDU second vote
    in Brandenburg – where the party had one overhang seat – from 482,593 to 441,445. That would have caused CDU to win an extra seat (going from 245 of 602 to 246 of 603) with fewer votes.

    While the subsequent upper-tier, nationwide seat distribution on the basis of the smallest votes-to-lower tier seats – 0.5 ratio would have restored full proportionality to the distribution of mandates, I could see the Federal Constitutional Court taking exception to a system that allows negative vote weight to arise even in an intermediary stage, in light of its 2008 ruling on the matter.

    That said, under the current system it’s still possible for a party to win fewer seats with more votes; however, all parties would receive fewer mandates, with proportionate seat losses across the board.

  34. Question: In the current system does losing seats without a corresponding loss of proportional representation in the Bundestag count as negative vote weight? If a party would be allotted 310 seats in a 620 seat chamber but ends up with 300 seats in a 600 seat chamber, has its voting weight actually diminished?

  35. Mark, since all the parties would be equally affected in proportion to their voting strength, I don’t think that would be considered negative vote weight.

    For example, in the notional distribution of seats for the 2009 Bundestag election, had Die Linke won an extra 8,000 votes in Hamburg, it would have captured an additional seat at the expense of CDU in the initial allocation, leaving CDU with a minimum seat allocation of 194. However, in turn this would have increased the minimum votes-to-seats – 0.5 ratio from 60,813 to 61,128, and reduced the final size of the Bundestag to 666, of which 84 would have gone to Die Linke, for a loss of one seat. Nevertheless, the other parties would have also incurred in small proportionate losses, as a fully proportional distribution of seats would have been achieved with a slightly smaller Bundestag.

  36. After some thought I think the best improvement to the German electoral system (and indeed to almost any version of MMP) would be to let people rank their preferences on both votes. Districts would then be filled using IRV (or Borda, Condorcet, or any preferential method), meaning more competitive contests, fewer wasted votes and a better chance for small-party candidates to get district seats. The list votes of parties that did not get 5% of first-preference votes will be redistributed among the parties that did pass the threshold, meaning a fairer deal for small-party voters, whose votes will no longer be completely wasted when their party doesn’t reach the threshold.

  37. What would the German Election result been if there had been no 5% threshold?

    What would the German Election result been if the 5% threshold was not nationwide, but Lander-wide as in the First German MMP Election?

  38. Had the 1949 Bundestag electoral system been in place for last month’s German federal election and had voters cast their ballots in the same way, the distribution of Bundestag mandates would have been as follows:

    CDU – 240
    CSU – 56
    SPD – 176
    Grüne – 49
    Die Linke – 48
    FDP – 22
    AfD – 10

    Both FDP and AfD would have been in the Bundestag, although significantly under-represented. More importantly, Chancellor Merkel’s CDU/CSU-FDP coalition government would have retained a comfortable majority of 318-to-283.

    While CDU would have been helped by the state-level application of the D’Hondt rule, the use of the Sainte-Laguë rule would not have changed the outcome that much: the ruling coalition would have still had a 310-to-291 Bundestag majority, and the main beneficiary would have been AfD, which would have had fourteen mandates.

    Finally, FDP would have won substantially more seats than AfD despite having just slightly more votes than the latter due to the fact that just under one-quarter of the Free Democrats’ votes were cast in states where the party fell below the five percent threshold, whereas more than four out of seven AfD votes came from states where it polled under five percent of the vote.

    As for last month’s outcome with no threshold, suffice it to say there would be sixteen parties in the Bundestag, including three far-right parties.

  39. Speaking of what-if scenarios, here’s an interesting one, and historically recent as well: the 1990-only, two-zone threshold provision, which allowed any party to secure Bundestag representation by winning at least five percent in either of two electoral zones, one comprising the former West Germany (including West Berlin), the other the former East Germany. PDS was the only party to take advantage of that provision, winning 17 seats with 2.4% of the vote by having polled 11.1% in the former GDR (in the former FRG the ex-Communists won only 0.3% of the vote, but even that was good enough for one list seat in North Rhine-Westphalia, since votes won in both zones were taken into account once the threshold requirement was satisfied).

    While the two-zone provision was disliked at the time by the established parties – not least because it allowed the despised PDS to enter the Bundestag – had it been in place last September, both FDP and AfD would have won Bundestag seats – the former by crossing the threshold in the western zone, the latter in the eastern zone – and the distribution of Bundestag mandates would have been as follows:

    CDU – 228
    CSU – 50
    SPD – 172
    Die Linke – 57
    Grüne – 56
    FDP – 32
    AfD – 31

    However, even with the Liberals in, the ruling CDU/CSU-FDP coalition would have fallen four seats short of an overall majority.

    Finally, the two-zone threshold provision would have had the same effect as lowering the threshold to the three percent level recently established for European Parliament elections.

  40. Final results of last September 22 Bundestag election – published today by Germany’s Federal Returning Officer – are now available on my website’s Germany page.

    Although the final numbers are largely in line with preliminary election night figures, the Social Democrats gained an additional adjustment mandate, which increased the size of the Bundestag to 631 seats.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.