Martin Fayulu has reason to thank Congo voting machines he once feared

The people of the Democratic Republic of Congo may have just experienced one of the biggest electoral frauds in recent history.

The country’s electoral commission last week announced opposition leader Felix Tshisekedi as the winner of a historic election marking the end of President Joseph Kabila’s 18 years in power. But a Financial Times analysis of two separate collections of voting data shows that he didn’t win. 

At 3am Kinshasa time on January 10, the commission told bleary-eyed Congolese that the tallies of the main candidates were as follows: Mr Tshisekedi had 38.6 per cent, rival opposition leader Martin Fayulu 34.8 per cent and ruling party pick Emmanuel Shadary 23.8 per cent. 

According to the election data seen by the FT, representing 86 per cent of total votes cast across the country, the outcome was very different. Mr Fayulu won 59.4 per cent of the vote, the data shows, while Mr Tshisekedi should have finished a distant second with 19 per cent and Mr Shadary polling at 18 per cent.

The FT compared the figures to a separate set of voting results collected manually by the country’s respected Catholic Church, which ran the biggest election observation mission. The two sets correlated almost exactly and would have been near impossible to fake, experts said. 

The group of Catholic bishops, known as the CENCO, has been outspoken since the vote. Five days after the election, it said voters has expressed “a clear choice at the ballot box” and called on the electoral commission to accurately report the result. Privately it told diplomats that Mr Fayulu was in an unassailable lead and after Mr Tshisekedi was declared the victor, it told the UN Security Council that the commission’s result did not match its own tallies.

Those tallies, representing 43 per cent of turnout and collected by hand from 28,733 polling points, were seen by the FT and show that Mr Fayulu secured 62.8 per cent of this sample of votes. 

By contrast, the figures in the leaked data are electronic tallies from 62,716 voting machines across the country and were said to have been obtained from the electoral commission’s central database before the results were announced. 

The electoral commission introduced the untested electronic voting machines in Congo for the first time in December, claiming the devices would help to reduce costs. The machines printed ballot papers that were counted by officials on election night — but they also stored electronic counts. 

During the election campaign, Mr Fayulu warned that the machines would be used to rig the vote. In the end the devices may have given the opposition leader the transparency he has demanded. 

An FT analysis of the electronic tallies in the leaked document shows a near perfect match with the Church’s partial results — with a correlation coefficient ranging from 0.976 to 0.991 for each of the three leading candidates (1 representing a perfect match). 

The electoral commission denied that its results were fraudulent. Barnabé Kikaya Bin Karubi, chief diplomatic adviser to Mr Kabila, said it would be up to the constitutional court to decide on the validity of the election and declined to comment on any potential fraud. Gilbert Kankonde Nkashama, a spokesman for Mr Tshisekedi, said it was impossible for Mr Fayulu to have won the election and questioned the independence of the Catholic Church. 

Mr Fayulu is seeking to overturn the result in the constitutional court, although the court’s impartiality has also been questioned. 

Congo, a former Belgian colony of 80m people, has held only four elections since independence in 1960 and has never before had a transfer of power through the ballot box. Mr Kabila had been due to step down in 2016 but elections were delayed until street protests and regional pressure forced him to hold the vote. 

Mr Kabila’s ruling coalition had sought to retain control of the presidency through Emmanuel Shadary, his chosen successor. Mr Fayulu’s supporters have alleged that when voters failed to come out for Mr Shadary in sufficient numbers — he finished third — the election commission was told to install Mr Tshisekedi instead. 

According to the results seen by the FT, Mr Fayulu received more than 9.3m votes, 3m more than the electoral commission announced, and won in 19 of Congo’s 26 provinces, including the capital Kinshasa and the heavily populated eastern provinces of North Kivu and South Kivu.

In contrast, Mr Tshisekedi took 3m votes, the data indicate, 4.1m fewer than the electoral commission showed, while Mr Shadary secured 2.9m votes, or 1.5m votes fewer than the published tally. 

The data show 15.7m out of the 18.3m votes cast on election day, but the missing votes could not have resulted in a different winner. Malfunctions in voting machines meant that not all vote tallies were transmitted to the central database, the person with knowledge of the database said. 

The file takes the form of a spreadsheet of comma-separated values, a format used by many software packages to store tabular data, and runs to more than 49,000 rows. Each polling station is identified by a unique six-digit number. 

The FT analysis also found no significant evidence that the data deviated from Benford’s law, a statistical test commonly used in fields such as forensic accounting to identify fraudulent data.