Research Points to Election Fraud

By Karuti Kanyinga
Karuti Kanyinga, a senior research fellow at the Institute for Development Studies of the University of Nairobi, explains in these excerpts from an AllAfrica interview why analysts have concluded that December’s presidential election in Kenya was rigged.

We’ve been able to look at the multiparty elections and ask ourselves how the 2007 elections compare with other elections and ask ourselves what justifies the claims of irregularities, of vote-rigging and of rigging in favor of a particular candidate, including Mwai Kibaki, the PNU (Party of National Unity) presidential candidate, and Raila Odinga, who was ODM’s (the Orange Democratic Movement) presidential candidate.

Our beginning point was a quick comparison of three elections held on the same day, at the same time, on the same spot: a civic election for local government, a presidential election and a parliamentary election. We are different from many countries – a voter comes to a polling station and is given three ballots for a civic, presidential, and parliamentary election, and the only difference is the color of the ballots... There are very few people who choose not to vote for a councillor – for local government – or not to vote for a member of parliament, or not to vote for president. The only difference is there might be a voter who might spoil a ballot...

In sum, we are saying the difference [between the number of ballots cast in voting for different levels of government] must not exceed one percent. How do we arrive at this one percent? We have done a forensic investigation of the 1992, 1997, and 2002 elections to ask ourselves the minimum number of votes that we can allow as anomalous, caused by spoiled ballots or any other reason. We have come to the conclusion that it is one percent.

In 2007, discrepancies [emerged] that cannot be explained by any scientific method.

Scientifically it does not make sense that out of all the elections that Kenyans have conducted, it's only the 2007 election where you find a difference of five percent [between the number of votes cast in balloting for different levels of government]. In some cases, five percent is a lot of votes because it's comprised of thousands of votes, implying... thousands of people went to a polling station and made a decision: "I am not voting for a member of parliament," or "I am not voting for a councillor. I am only coming here to vote for a president." That's unheard of scientifically, given what we have done in our own country [in the past]...

Put together, we found that the anomalous amount of votes - in all the results - is over 365,000. This number comes from the records of the electoral commission, even after everything had been resolved – implying that a week after results had been announced, after they had cleared their data, we still cannot account for 365,000 votes... The question is: to whom do we attribute these 365,000 votes?

When you do a cross-section of where these 365,000 votes come from, you realize that we cannot attribute all of them to Raila [Odinga], the ODM presidential candidate, because we do not have serious cases of anomalies in the ODM-controlled area of Luo Nyanza... The majority of cases where we find anomalies are in President Kibaki's strongholds, and also on the coast... We are not saying that we can attribute this wholly to that, but scientific evidence is pointing fingers to stuffing ballots, to inflating votes on the coast in order to ensure that the president gets 25 percent of the total [number of] votes. There is a constitutional requirement that a winning presidential candidate must have acquired 25 percent of the total vote in at least five out of the country's eight provinces... And on the coast is where we find very serious cases of anomalies between presidential and parliamentary votes.

Politically, of course, someone may want to argue that this is not true. Politically someone may want to argue that in the past - in 1992 and 1997 elections... the parliamentary votes had a difference of over 600,000 votes [from the presidential elections]. They forget that in both elections, there were 10 constituencies where there were no members of parliament contesting [seats]... meaning, there was no parliamentary election. There was only a presidential election, therefore presidential votes in those two elections were higher than the parliamentary vote.

The bottom line is that this election was fraudulent... that this election appears to have been rigged in favor of a certain presidential candidate, and all evidence seems to point fingers to President Kibaki's strongholds.