Daily Archives: April 11, 2009

Europe 2009: Revised Predictions

I have found the first prediction for the European elections on the interwebs. It’s a model-based prediction by three psephologists, two from the London School of Economics and one from the Trinity College in Dublin, Ireland. The model integrates data from the EU elections in said country since 1979 (or whenever it first voted), as well as variables such as opinion polling and government/opposition status. In addition, it attempts to correct the overestimation of large parties and underestimation of small (notably Eurosceptic) parties. This prediction is done in every individual country and seats are calculated in said country using that country’s electoral law. I’ve always been sceptic of these model-based predictions that use too much past statistics in the analysis and too little variables such as electoral alliances and so forth.

Here is the overall prediction of the model for the EU Parliament

EPP 249 (vs 288)
PES 209 (vs 217)
ELDR 87 (vs 100)
UEN-MER 58 (vs 44)
EUL-NGL 48 (vs 41)
Greenies 39 (vs 43)
NI 29 (vs 30)
ID 17 (vs 22)
Total 736 (vs 786)

Analysis by member state is here, for reference.

Here is my critique, analysis, revised predictions, whatever thee wants to call it.

Austria: First mistake is not counting Hans-Peter Martin’s list. He’ll probably run. HPM’s exclusion screws this whole thing up. Their prediction has obviously been based on federal election polling, and not European polling. European polling now indicates a narrow edge to the ÖVP and much lower results for the far-right FPÖ-BZÖ, which suffers from low turnout amongst its anti-European supporters. As of now, I expect the ÖVP and SPÖ to poll roughly 32% each, maybe the ÖVP slightly ahead, with Martin at around 8-10%, like the FPÖ and the Greenies. The BZÖ will probably be right at the 4% threshold, and I probably agree that they’ll win one MEP. ÖVP (EPP)

Belgium: Belgian politics are fragmented, they’re right on that. The rest of their prediction isn’t that bad either. While I agree that the CDV will pull ahead in Flanders, I disagree on placing the Flemish Socialists second in Flanders (that is a fight between VLD and Ljist Dedecker) and the MR first in Wallonia (the PS will take first place rather easily). Current polling in Belgium (regional elections are being held the same day) indicates that the Ljist Dedecker in second is a very real possibility, as is a collapse of Vlaams Belang and a general fall in the votes of the main parties (CDV, VLD, Socialists, Greenies) to the benefit of Dedecker and the Flemish nationalist N-VA. In Wallonia, the two biggies (PS and MR) should both lose ground to the benefit firstly of the Greenies (who are nearly polling 20%) and the Christian democrats. Once again, the far-right (FN) is falling back. No gains for them. CDV (EPP) overall

Bulgaria: GERB is, I think, by far the favourite party in this election. Good call. But NDSV at 9.5%, gaining over 2007? Haha. I think not. They’re probably still polling 1% nationally. Hard to predict the Turkish DPS. Depends on how good their GOTV effort is and how low Bulgarian turnout gets. GERB (EPP)

Cyprus: I don’t know Christofias’ approval ratings, but assuming it’s still sky-high, AKEL wins and gets a third seat. AKEL (EUL-NGL)

Czech Republic: The effect of the recent political crisis and caretaker government will be interesting to see, but current polls seem to indicate a tightening race between the ODS and the opposition ČSSD. I still think that ČSSD, but I disagree with the prediction that it will be by such a big margin (39-33). I agree that the Greenies should pick up a seat. ČSSD (PES)

Denmark: Denmark got a new Prime Minister just recently, Lars Løkke Rasmussen to replace Anders Fogh Rasmussen (no relation), who will be heading to NATO. Polling indicate that Lars Løkke as Prime Minister absolutely destroys the governing Venstre to the benefit of the other parties, notably the Social Democratic, Socialist People’s, and Venstre’s Conservative coalition partner. In fact, Venstre would be behind SD and the Socialists! It is important to cite (I just learned this too) that Denmark uses surplus votes in EU elections. The Social Democrats, Socialists, and Social Liberals have formed a electoral alliance, meaning that surplus will get distributed inside those parties, allowing the very weak Social Liberals to keep an MEP. The two anti-EU outfits have also formed an alliance, practically guaranteeing one (maybe both, but that’s uncertain) of them an MEP. Probably the June Movement. SD (PES)

Estonia: A good analysis. Centre wins, governing Reform second, massive Social Democratic loses. Centre (ELDR)

Finland: Another good analysis. Expect the conservative KOK to get first, ahead of the opposition Social Democrats and the senior government party, the Centre. The True Finns (nationalists) should win their first MEP. And it will be another tough fight for the Swedes to get in. However, I have no clue where they’re getting the Christian Democrats at 5.5% from. KOK (EPP)

France: Gah! Correct on the UMP ahead of the PS this time around, with important loses for the Socialists. Their prediction for Besancenot’s NPA and the MoDem seem too high and too low respectively. MoDem at 11% and NPA at 9% would make more sense. Now it gets silly. DLR at 6.6%? What? Unless they got dibs on some ultra-secret poll, they pulled that number out of their ass. The DLR will get 1%, 2% if they’re lucky. The DLR will never get 6.6%. The MPF number isn’t too bad, though. I have no clue why the Greenies are down to 5.6% all of a sudden, when all polling places them at 7-9%. I feel that they’re slightly overestimating the FN (they forgot to include the FN dissident lists, which will play an important role in certain regions) and the LO, while underestimating the PCF. I won’t mention seat numbers, since heaven knows it’s hard to work with those retarded constituencies. UMP (EPP)

Germany: Not too bad, though separating CDU from CSU is a bit weird since they’re always included together. Their figures put CDU/CSU at around 34.5%, so ahead of the SPD. But it seems the pollsters tend to overestimate the SPD to the detriment of the CDU/CSU. Current polling also puts the Greenies quite high (13%), with the FDP (10%) and Linke (8%) improving on their 2004 results.  CDU/CSU (EPP)

Greece: No qualms with placing PASOK ahead of ND, and a 38-33 prediction seems reasonable. However, placing LAOS at 8% is clearly too much. They’re polling low and they tend to overpoll, it seems. On the other hand, the Communists and Radical Left seem a tad too low. It will be tight for the Greenies, nobody knows if they’ll make it or not. One can hope. PASOK (PES)

Hungary: I doubt anybody that is reasonably sane could disagree with the assessment that Fidesz will win a huge victory in Hungary. The debate is over whether or not they break 60%. While these guys say they don’t, I do, and polls seem to agree with me. Their problem seems to be overestimating the small parties, the MDF and SZDSZ. Fidesz (EPP)

Ireland: I don’t know where they’re getting this from. They’re expecting Fianna Fáil, massively unpopular, to move ahead of Fine Gael and make gains vote-wise? No, no. Fianna Fáil will lose votes, certainly, get second, certainly. Fianna Fáil’s seat count could fall from 4 to 2, even.Fine Gael could lose votes, but they’ll stay ahead of Fianna Fáil. On the other hand, they’re pegging Labor at a small +3% gain, while they’re pulling 17%, maybe even 20% in polls right now. Sinn Féin’s performance will be interesting to watch. Fine Gael (EPP)

Italy: Pegging PdL at 32.5% seems too modest for me. 34-36% is more the range for PdL this time. On the other hand, the PD will be lucky to get 29.8%, very lucky. The PD’s range seems to be 22-25% for now, 25% being generous. They’re being too mean to the smaller parties, notably Italia dei Valori and Lega Nord. Both could get around 8-10%. Polls disagree with their placing of the UDC at 5%, polls say that it’s nearly 8%. Furthermore, they don’t take into account two electoral coalitions being formed. The first, Sinistra e libertà, composed of the PS, Democratic Left (Communist dissidents) and the Greenies, is polling around 5-7% right now, which would save some of the PS and Greenie MEPs. The PRC+PdCI electoral alliance is very weak, at around 3%, and could potentially lose all seats. Lastly, I have no clue where they’re getting Alessandra Mussolini at 3.7%. That makes zero sense. Zero. On a very last notes, European elections in Italy defy common sense and established reality because of intricate eletoral deals and coalitions, which allow a ton of parties to get MEPs while polling way under the 4% “threshold”. PdL (EPP)

Latvia: The government is unpopular, resigned in February, and has yet to be replaced. Plus, opinion polling is hard to come by. Common sense says that the opposition, the New Era and Harmony Centre, should be the main benefactors. The Union of Greens and Farmers should also win seats, and the governing People’s Party should minimally improve on its 2004 result (but still lose the election). New Era (EPP)

Lithuania: If my memory serves me correctly, I think the governing Homeland Union (TS-LKD) is still leading, with like 15%. The far-right Order and Justice is second, the opposition Social Democrats are third. I agree that the main losers will be the Labour Party, which has suffered a lot since 2004. TS-LKD (EPP)

Luxembourg: Polls are very hard to find in Luxembourg, though I think it’s very likely the CSV will win. There should be little or no movement in terms of seats. CSV (EPP)

Malta: I agree with these guy’s predictions. Labour wins probably. Interesting to see how the greenies do. They came close to a seat in 2004, they’d be very lucky to get one this time. Labour (PES)

Netherlands: There was some speculation as to if Geert Wilder’s far-right PVV could top the poll, seeing as some (generally bad) pollsters have them leading all other parties. Thankfully, he probably won’t. The sane parties, the Christian Democrats (CDA) should get first and Labour (PvdA) will fight PVV for second, though I think Labour will end up second in the long run. Though these guys are underestimating Wilders, I think. Apart from PVV, D66 will be the other main winner in this election. The SP will manage a few percentage points higher, but will probably stay under 10%. ChristianUnion-SGP (electoral alliance between slightly sane Christian right [CU] and totally insane ultra-orthodox Calvinist Christian right that closes party website on Sundays [SGP]) should perform at its 2004 levels. The liberal-conservative VVD should suffer, as will the governing CDA and PvdA. CDA (EPP)

Poland: I think, like Hungary, it’s certain beyond all reasonable doubt that the governing PO will win massively. The question here is if they break 50%. On here, I agree with this estimate. They won’t. 34% for PiS is being a bit generous, but not that much. However, it seems as if these predictors are not taking into effect the split in LiD, which split up into three old parties. I think the biggest of those is the bunch of ex-Communists, Democrats of the Left or something. That gang should pull 8-10%. The far-right will get wiped out. But, as I said in December: Turnout gives weird results here, as seen in 2004. So take these with a grain of salt. Maybe a bag of salt. PO (EPP)

Portugal: These elections will be a dry-run for the September general elections in Portugal. I agree that the PS is likely to pull first place again, ahead of the conservative Social Democrats (PSD). However, the PSD will have a hard time breaking 30% as they seem to predict. However, based on recent polling, they seem to be underestimating the Left Bloc (the libertarian part of the far-left) slightly, while being too generous with the Communist Party. If the conservative CDS/PP allies itself with the PSD as in 2004, then that coalition would likely pull first. In lack of such coalition, expect the PS to shed some seats and votes but remain on top. The PSD should also shed some votes, to the benefit of the smaller parties. PS (PES)

Romania: It’s hard to get polling figures for Romania now, but I believe the Emil Boc PD-L/Social Democratic government is still enjoying a honeymoon period, so I’d expect the PD-L to win, and the PSD in second. The rest of their prediction is generally good, save for giving Independents (László Tőkés and Elena Băsescu) 13% of the votes. László Tőkés got 3.4% in 2007, and since he’s Hungarian, that’s unlikely to change much. I somehow doubt Elena Băsescu will do that well.

Slovakia: As with Hungary, it’s obvious Smer wins by a large, large margin over SDKÚ-DS. The main losers should be all the EPP parties, SDKÚ-DS, Mečiar’s ĽS-HZDS (which will be suffering massive loses), Christian Democrats, and the Hungarian MKP. Smer’s fascist friends, the SNS, should have little trouble getting past the 5% threshold. These guy’s guesses don’t seem too bad, except being too harsh on Smer and a little too generous with SDKÚ-DS. Smer (PES)

Slovenia: The Social Democratic government is running high in polling apparently, and it’s quite certain they’ll win. Their predictions seem pretty spot on here, I think. LDS might keep one seat, but it will be hard for them to survive due to only 7 seats in play and their imminent political death (as if they weren’t dead already). SD (PES)

Spain: On this, I must differ with these predictions. Despite the troubles in financeland being felt a lot in Spain, the Socialist government actually remains high in polls, ahead of the conservative PP. However, both the PSOE and PP should shed some votes and seats since 2004, to the benefit of the liberal-unionist UPyD primarily, and potentially the communists (IU) and maybe the nationalists. These predictions are also too nice on the nationalist Galeusca (which lost the Gal-Galician part with the BNG joining the more leftie more nationalist Europe of Peoples with the Catalonian ERC), which won’t get 3 seats but will rather keep its two seats (one CiU, one Basque). As for Europe of the Peoples, ERC’s coalition now including the Galicians and the leftie Basque nats, they won’t gain anything (let alone win 6%!) except holding ERC’s lone MEP. BNG won’t gain anything back judging on their poor performance in the Galician elections. 5.8% for IU is probably too nice, as is 5.2% for UPyD, despite everything. PSOE (PES)

Sweden: While it’s pretty certain the Social Democrats will get first place (it would be practically impossible for the governing Moderate Party to get first place) and improve on their disastrous 2004 result, the interesting thing to watch will be whether the combined governing right-wing parties (Moderate, Liberals, Centre, Christian Democrats) will run ahead of the Red-Green opposition (SD, Left, Greenies) as some current polls are suggesting. At any rate, the main losers in the circle of establishment parties will probably be the smaller Liberals, Left, and potentially Christian Democrats all at the benefit of the Moderates and SDs. As to the June List, these guy’s predictions that they’ll fall to 6.2% is as good as any guess. It’s hard to predict these uniquely European anti-European lists that may do atrociously in nationwide elections but have a much larger base in European elections. However, I don’t expect the June List to grow beyond its 2004 level, 14.5%. It will be interesting to see if the xenophobic far-right Swedish Democrats break the 4% threshold to get an MEP as some polling suggests. Or maybe the June List will keep them low. SD (PES)

United Kingdom: A bad prediction on their part. UKIP’s total collapse will help the Conservatives, and they should certainly make further gains in votes and seats. Labour won’t collapse any further, and should win back seats, but +3 is a bit too much. The LibDems will stay at their 2004 levels, and UKIP will be lucky to get 4 MEPs. 2 is probably what they’ll end up with. The Greenies shouldn’t be under any pressure and their 2 MEPs are safe. The Nazis BNP’s exclusion from their calculations is very frustrating and reflects on the poor nature of their prediction, especially since the BNP is very close to getting an MEP. 1 DUP, 1 UUP, 1 Sinn Féin is a sane guess in Northern Ireland, though it depends on preferences here. Conservatives (ED)

My revised prediction in terms of member states:


Government 15
Opposition 12