As the ramifications of Bulgaria’s European Parliament elections continue to unfold, seemingly now bringing forward a realistic prospect of early national parliamentary elections in the light of the results, it is worth revisiting just why the results that began to emerge on the night of May 25 came took so many aback.
Leaving aside those who ignored many of the polling agencies, the Bulgarian public who were following the election campaign repeatedly were told by most agencies that it would be a close-run thing between centre-right GERB and the Bulgarian Socialist Party.
Any number of polls put the BSP slightly, and sometimes more than slightly, ahead of Boiko Borissov’s GERB. Time and again, pollsters said with firm confidence that just four parties would win shares of the 17 MEP seats allocated to Bulgaria in the next European Parliament.
The centre-right Reformist Bloc was written off as in a fifth place that gave it only the scantest chance of a seat.
In reality, according to the official results released by Bulgaria’s Central Election Commission, GERB came in with 30.4 per cent, the BSP trailed at 18.93 per cent and the Reformist Bloc came in fifth, winning one seat with 6.45 per cent.
By its own calculations, the Reformist Bloc – which has asked state institutions for help in challenging the election results in the Constitutional Court on a long list of grounds – fell just about 1000 votes short of winning two seats, and that is going by the official results which it rejects.
Just two agencies came the closest to getting it right, and were consistent in doing so in the final weeks before the May 25 vote: Alpha Research and Exacta, the latter largely made up of people from the now-defunct Centre for the Study of Public Opinion, which was shut down at the instance of the BSP after the May 2013 elections.
As noted in a report by Offnews, the only other agency that got something right was the BBSS “Gallup” agency, which was alone in predicting the vote for the MRF with great accuracy. It said that votes for the MRF would be in the 340 000 to 420 000 range. According to the CEC, the MRF got 386 725 votes.
In mid-May, the Mediana and Institute for Modern Politics agencies saw the BSP just ahead of GERB, Nikolai Barekov’s Bulgaria Without Censorship in third place ahead of the MRF, and Georgi Purvanov’s ABC ahead of the Reformist Bloc (the variance between that Mediana ahead of ABC). Again: GERB came first by a long chalk, the MRF was hard on the heels of the BSP, with BWC well behind the MRF at just 10.66 per cent and Purvanov’s ABC was below the threshold to win a seat.
The Sova Harris agency saw the BSP as continously ahead of GERB by three per cent from September onward. Three weeks before the election, Sova Harris saw the BSP vote as 19.2 per cent and GERB as 17 per cent, the MRF 6.2 per cent and BWC 5.5 per cent, with the Reformist Bloc below the threshold at 4.1 per cent.
According to the Offnews report, Sova Harris’s Vassil Tonchev, in appearances on TV7, had repeated three points – that the BSP was ahead of GERB, that BWC was either ahead of the MRF or running close to it, while the Reformist Bloc was on the edge and a vote for it might go nowhere.
In contrast, from late February onwards, Alpha saw GERB leading ahead of BSP, the MRF ahead of BWC, the Reformist Bloc winning a seat and the National Front for the Salvation of Bulgaria beating its elder ultra-nationalist rival Ataka.
Exacta did the best in forecasting a decisive gap between GERB and the BSP, although it said that it would be about 4.5 per cent. The agency also predicted a minimal lead of the MRF over BWC, saw the Reformist Bloc as certain to win a seat, and even chances of ABC either getting a seat or falling just below the threshold.
After the elections, the pollsters’ performance was the stuff of television talk shows – and even highly popular satirical show Gospodari Na Efira handing one of its Golden Skunk awards to Sova Harris’s Tonchev for the inaccuracy of his agency’s polls.
Such things happen, Tonchev said, reportedly saying that one time out of 10, things go wrong, and go really wrong.
Nikola Kolev of Mediana, also the recipient of a Golden Skunk, ascribed the error to low voter turnout, saying that the inaccuracy was the result of people who had told pollsters that they had made up their minds to go to vote and for which party they would be voting – but in the end did not head for the ballot booths.
Kolev said that he would keep his Golden Skunk in a prominently visible place so that he would see it whenever doing research.
After the elections, some politicians, notably Borissov, twisted the knife in scornful comments on the predictions that polling agencies has given.
This is not the first controversy about polling agencies in Bulgaria. Some frustrated politicians, as well as rival agencies, have made dark suggestions (all denied) about certain agencies being motivated by considerations other than pure calculation.
Meanwhile, polling agencies were said to be reconsidering their methodologies, taking account of the gross inaccuracies in several polls ahead of the European Parliament elections.
(Photo: Michal Zacharzewski/sxc.hu)