Discrepancies found in vote counts across EU polling districts, sparking concerns over electoral integrity

In every election, each voter has one vote. To ensure that each vote is counted only once, the number of votes cast can be compared with the number of envelopes and ballots. If the numbers match, everything is deemed correct. However, an examination of the vote counting in the recent EU election shows that this is far from always the case. In almost half of the examined polling districts, 196 out of 410, the numbers do not match. In more than one-third of districts, for example, the number of votes is either too many or too few compared to the number of envelopes. This equates to a total of 213 “errors”, which represents more than one in a thousand votes.

More votes than envelopes is the most common discrepancy, and in several protocols from polling stations, the difference is explained by likely missing double ballots in envelopes. This means it is too late to correct – the County Administrative Board, which conducts the final vote count, cannot possibly know which are the extra ballots if they were sorted incorrectly at the polling station.

Double ballots should be detected before the vote ends up in the ballot box or when the ballots are taken out of the envelope, says political scientist John Högström. In 100 of 410 polling districts, the number of votes cast does not match the number of envelopes either.

Even though the errors have not affected the election results, John Högström believes the extent is alarming and should serve as a wake-up call for the Election Authority. If voters feel that the risk of errors is high, it could lead them to refrain from voting. Therefore, it’s important to act before confidence in the election system is damaged. The Election Authority has been contacted but does not want to give an interview.

An examination was conducted in 14 municipalities and the vote counting in the municipalities’ polling districts – 410 in total. What has been looked at is the number of votes cast in the voter list, the number of counted envelopes and the number of counted ballots – both on election night and the county administrative board’s recounts. When differences have arisen in any or some of these variables, it has been noted. For the ballots, the final vote count of the County Administrative Board has been used. The municipalities that have been examined are: Borås, Falköping, Falun, Härnösand, Jokkmokk, Klippan, Landskrona, Mölndal, Ronneby, Sala, Skellefteå, Sölvesborg, Täby and Varberg.

In Sweden there are a total of 290 municipalities and about 6,000 polling districts – the examination is thus not comprehensive.