Liam Taylor

Uganda-based journalism

Winner takes all

Yoweri Museveni and Kizza Besigye have both claimed victory in last month’s presidential elections in Uganda. Who’s right? 


The Besigye procession through Kampala on the last day of campaigning

Kampalans say that they know who their president is – and it’s not Yoweri Museveni. The former guerilla leader, who has ruled Uganda for 30 years, was declared the winner of last month’s presidential elections, with 61% of the vote. But loyalties in the capital lie with Kizza Besigye, an opposition candidate and fierce critic of Museveni. Besigye says he won, and has described the electoral process as ‘a creeping military coup’. His supporters call him ‘the people’s president’, and describe his home, where he is held under house arrest, as ‘State House Kasangati’.

Even on official figures, Besigye won two-thirds of the vote in Kampala. He held massive pre-election rallies here, some blocked violently by police. The last day of campaigning felt like a revolution, as Besigye supporters beat down images of Museveni from lampposts and left them broken in the gutter. It is not hard to see why Besigye, and many Kampalans, believe that this election was stolen. But Kampala is not Uganda: elsewhere, in the villages and among older people, Museveni has genuine, if less passionate, support.

Did Besigye win? The debate in Uganda is tense, and potentially explosive. But in fact there are two different questions to ask. First, did Besigye win this election? Second, would Besigye have won a fair election? The distinction is not merely academic. It will shape an upcoming court challenge, lodged by a third candidate, Amama Mbabazi. It will also govern political debate in the turbulent and unpredictable months ahead.


Broken Museveni poster in Kampala

‘Deliberately disenfranchised’

On 18 February, millions of Ugandans queued patiently to vote for their leaders. The mood was hopeful, even excited. Turnout was up. In most places, the voting was peaceful. But as the day progressed, it became clear that there were major problems: lengthy delays, allegations of ballot stuffing, the arrest of Besigye, and a social media shutdown. The stories that have emerged over the past fortnight are a mixture of the shambolic and the sinister, played out in an atmosphere of deep suspicion towards the Electoral Commission, the police, and other state institutions.

The first problem was the late arrival of voting materials. Many polling stations did not open until the afternoon; voters had to wait for hours in the hot sun or, if they had other commitments, were unable to vote at all. The Electoral Commission eventually agreed to keep open the affected polling stations for longer, but not before police had used tear gas to disperse angry demonstrations.

Delays were longest in Kampala and Wakiso: the most central, most developed, and best connected districts in the country. They are also Besigye strongholds. Voters were ‘deliberately disenfranchised’, say Martin Mwondha of the Citizen Election Observers Network – Uganda (CEON-U), which deployed 1250 observers on election day. Though there in no proof of intent, adds Charles Mwanguhya Mpagi of the Citizens’ Coalition for Electoral Democracy (CCEDU), it certainly looks like ‘selective inefficiency’. Just over one in two voters in Kampala and Wakiso cast their vote, against a turnout of 68% nationwide.

A second, shadier problem was allegations of ballot stuffing. These were widespread. At 9am on election day a woman in Mbarara, a town in the west, gave me the registration numbers of a car and a motorbike allegedly transporting pre-ticked ballot papers (I have been unable to follow-up these charges). By the afternoon, voters I met in Masaka were recounting rumours of ballot stuffing in nearby Sembabule. Besigye himself was arrested while trying to force entry into a house in Kampala that he said was a co-ordination centre for vote-rigging.

In the days after the election, some of these allegations were given more substance. Joy Kabatsi, an independent parliamentary candidate in Sembabule, claimed to have confiscated pre-ticked ballots from agents of Museveni’s National Resistance Movement (NRM); she has handed the evidence over to police. A polling official in the district told Al-Jazeera that results were altered with the collusion of the army and police.

Indeed, there were 130 polling stations across the country that reported 100% voter turnout; in 113 of these, Museveni supposedly won 90% or more of the vote. Of course, a 100% turnout is not impossible; but it requires that no voter has died, fallen sick, or moved away since registration was completed last autumn. Tellingly, these suspicious polling stations are concentrated in just four districts: Nakaseke, in central Uganda, and Mbarara, Sembabule and Kiruhura in the west. Museveni himself is from the west, and the NRM is strongest there; Mwondha of CEON-U reports that his observers reported the most problems from that region.

All this suggests clear evidence of vote-rigging in these districts, if not elsewhere. A further mystery is the high number of invalid ballots: nearly half a million, or 4.6% of the total (if ‘invalid’ had been a candidate, it would have come third). Museveni, with no little chutzpah, has said these votes were rightfully his: ‘my supporters ticked on the [party] symbol’, he says, instead of in the box. The opposition contend that some voters were presented with papers pre-ticked for Museveni, but defiantly added their own tick for Besigye, invalidating the ballot.

Underlying all these failures is a third, deeper problem: the role of the Electoral Commission. Few trust it. Its chief, Badru Kiggundu, is a Museveni appointee who has made openly partisan comments about the candidates. In the words of European Union observers, the Commission ‘lacks independence, transparency and the trust of stakeholders’.

The Commission declared Museveni the winner before all the votes had even been counted. It then took a whole week to release data from individual polling stations, fuelling popular suspicion of the vote aggregation process. Opposition activists have struggled to get hold of the official Declaration of Results forms, making it difficult for them to monitor vote counting. And police have hindered any attempts at scrutiny, targeting opposition vote tallying teams in raids on party headquarters and arrests across the country.

In short, election day was characterised by fraud and incompetence: but given the lack of transparency, it is difficult to know how much. One neat visualisation (reproduced below) is provided by analysts at Development Seed, an American organisation that promotes open data. They plot the percentage of votes for Museveni against the turnout in each polling station. The smudgy fingerprint in the middle shows that turnout was higher in pro-Museveni areas. The spot of red in the top right is those 130 improbable polling stations: clear evidence, the analysts say, of fraud.


A plot of all the polling stations in the country, with the percentage of votes for Museveni on the y-axis and the voter turnout on the x-axis. Thanks to Drew Bollinger at Development Seed,

Numbers game

That there was vote-rigging and malpractice is beyond doubt. But was it enough to ‘steal’ a victory from Besigye? Perhaps not, the numbers suggest.

In official results, Museveni won 5.97 million votes (61%), against Besigye’s 3.51 million (36%). The gap between the candidates was almost two and a half million votes. That’s a lot of votes to steal.

First, consider those 130 dodgy polling stations. In fact, let’s go further, and deem that anywhere with a turnout over 90% was rigged. There were 539 polling stations that meet this criteria; Museveni ‘won’ 188,760 votes in them, against 17,174 for Besigye. So scrapping the results from these polling stations only closes the gap by around 170,000 votes.

Next, take the problem of low turnout in opposition strongholds (especially Kampala and Wakiso, where ballots arrived late). Divide polling stations into two groups: those where Museveni got more votes than Besigye, and those where he got fewer. Ignoring the 539 ‘dodgy’ polling stations, which we’ve already scrapped, the average turnout in pro-Museveni areas was 71%, compared with 63% in pro-Besigye areas.

What would happen if we increased the turnout in pro-Besigye areas, so that it matched turnout in pro-Museveni areas, keeping the proportion of votes for each candidate the same? Besigye would catch up by 125,000 votes. Even if we assumed that all the extra voters would choose Besigye – which is obviously unlikely – then he would still only gain 430,000 votes.

So clear cases of ballot stuffing and differences in turnout do not amount to much: perhaps 300,000 votes. For the sake of argument, throw in those half million invalid votes as well, and assume they were all for Besigye. Even on the most generous (and unrealistic) assumptions, we have closed less than half the gap between the two candidates. A fairer estimate would be that we have accounted for perhaps half a million votes – leaving 2 million still to be stolen.

Could the ruling regime have found a way to steal those 2 million votes? Worse things have happened in other countries. So far we have considered only the most egregious examples of malpractice; a regime with thirty years’ experience could find plenty more subtle ways to fiddle the numbers. A few extra ballots could be stuffed in here and there; some Besigye votes could be counted for Museveni; the figures could have been altered at the aggregation stage.

But consider this: such a plan would involve adding an extra 70 votes to Museveni at every polling station in the land (or, equivalently, transferring 35 votes from Besigye to Museveni). That’s a lot: the average polling station has 370 voters. And the vote was closely watched, by party agents and ordinary Ugandans (see ‘Watch and Wait’, 18 February). In some places the level of scrutiny would make such rigging impossible, meaning that even more votes would have to be added on elsewhere.

Nobody, so far, has produced a smoking gun. Besigye’s party, the Forum for Democratic Change (FDC), say they have such evidence. So too does Mbabazi. But neither has yet presented it.

There is plenty of reason to think that Museveni could have won. A 2012 study by Jeffrey Conroy-Kutz and Carolyn Logan, academics at the University of Michigan, found that prosperity and improved security were the most important reasons for his victory last time round. In a December 2014 survey by the International Republican Institute (IRI), an American think tank, 69% of Ugandans said their country is heading in the right direction, and 77% approved of Museveni’s performance. A more recent survey, by British academics, found that Museveni still has strong support in the countryside. And no pre-election opinion poll put Museveni on less than 50%.

Of course, survey data is unreliable: 54% of respondents in the IRI study said that they ‘always’ or ‘often’ have to be careful what they say about politics. But Museveni has real achievements to his name: economic growth that has averaged 6% a year, albeit slowing; stability, in contrast to the chaos that came before; a much-admired campaign against AIDS. His supporters acknowledge the problems, from poor roads to missing medicines, but point out that development takes time. Besigye would not be able to fund his grand promises, argues influential journalist Andrew Mwenda. Many Ugandans have bought into Museveni’s message of ‘steady progress’, and look on Besigye’s talk of change as empty rhetoric.

On current evidence, then, a Museveni victory seems plausible. Absent any dramatic revelations, the answer to the question ‘did Besigye win this election?’ is: ‘probably not’.


Museveni supporters on their way to a rally in Mubende

Three decades, not three weeks

Do not misunderstand this argument. The assertion is a simple one: that a majority of Ugandans who wanted to vote 18 February voted for Museveni. That statement does not address more subtle issues: why people voted Museveni, under what pressures, and why some did not want to vote at all. The key question is not ‘did Besigye win this election?’, but ‘would Besigye have won a fair election?’

Long before polling day, Museveni already had this election sewn up. The modern Ugandan state has been built to ensure the perpetuation of his power, and the control of his NRM party. As Gabrielle Lynch, an academic at the University of Warwick, has observed, Museveni has always been mistrustful of multiparty politics, blaming it for the ‘sectarianism’ and ‘opportunism’ that tore the country apart in the 1970s and 1980s. Preserving the dominance of the NRM is not a betrayal of his principles, but the apotheosis of them.

This election was won over three decades, not three weeks. For an example, consider the role of the security forces. The Ugandan army, like the NRM itself, is the direct descendant of the National Resistance Army, created and led by Museveni during the five-year ‘Bush War’ of 1981-86. Museveni controls it with an iron grip; dissident voices are forced out or suppressed (David Sejusa, a general with opposition sympathies, is currently in jail). The police force has been increasingly militarised, and is headed by a general, Kale Kayihura, with a penchant for aggressively partisan rhetoric.

In the run-up to this election, overt intimidation was less marked than in previous years. There were instances of violence, notably a clash between Museveni and Mbabazi supporters at Ntungamo, western Uganda. FDC activists from Serere, an eastern district, say they were beaten by pro-government thugs, assisted by soldiers. But on the whole the campaigns were more peaceful than Ugandans have come to expect. This time, says Charles Mwanguhya Mpagi of CCEDU, the role of the military and police was more ‘subtle’.

That did not make the threat of violence any less real. Police chief Kayihura, for instance, organized a force of 150,000 ‘crime preventers’: a community policing initiative that more closely resembled a pro-government militia. The secretary general of the NRM warned that ‘the state will kill your children’ if they take part in protests which ‘destabilize the peace’ in Kampala. In an Afrobarometer poll, conducted in May last year, some 63% of respondents said that they feared becoming a victim of political intimidation or violence during election campaigns.

More perniciously, the rhetoric of the regime stoked fear of violence, and perhaps even civil war, should Museveni be defeated. Museveni has compared the opposition to ‘wolves’, while Kayihura was quoted as saying ‘we shall not hand over power to the opposition to destabilize the peace we fought for’ (remarks he says were taken out of context). In a survey last year by Research World International, a Ugandan polling company, 61% of people said they did not think that Museveni would leave peacefully if he lost; the country has never changed leaders in a peaceful election. Some will have voted for Museveni, or stayed at home, because they fear the consequences of an opposition victory.

Alongside violence, the regime used its control of resources to buy support. The line between state and party is blurred: the Alliance for Campaign Finance Monitoring (ACFIM), a civil society group, lists multiple cases where government resources, such as official vehicles, were used for NRM campaigning. A scheme to distribute 18 million hoes to Ugandan farmers – at a stated cost of 135 billion shillings ($40 million) – is widely seen as a bribe: EU observers spotted the hoes being given out at NRM rallies. In official campaigning alone, Museveni spent twelve times as much as Besigye and Mbabazi combined in November-December 2015, says ACFIM.

‘The monetisation of the election campaign is higher than in 2011’, has said Jo Leinen, head of a delegation from the European Parliament. But it was also more structured, argues Mwanguhya, of CCEDU. That makes it harder to pin a charge of bribery on Museveni. And there are other ways the state uses resources, in a manner that is unfair but not straightforwardly illegal: for instance, hinting that regions which vote for Museveni are more likely to receive funding for roads or healthcare.

Finally, Museveni dominates the media, making it difficult to subject him to proper scrutiny. There is an independent and critical press, but journalists face harassment: in the last fortnight, 20 journalists have been detained while covering Besigye, estimate the Human Rights Network for Journalists – Uganda (HRNJ-U). One was pepper-sprayed in the face. The most widely-read newspaper, New Vision, is government-owned: it published a 111-page supplement last month to celebrate 30 years of NRM rule.

Most Ugandans get their news from the radio, broadcast in local languages. Not surprisingly it is local journalists, away from Kampala, who face the most interference. The media regulator closed and seized equipment from 13 radio stations in January, sometimes as direct retaliation for hosting opposition candidates. Self-censorship is common. To take two examples, cited by EU observers, 93% of coverage on King’s Radio in Masindi was devoted to Museveni, while 66% of coverage on Kabale’s Voice of Kigezi was paid for.

Would Besigye have won a fair election? It is impossible to tell. But this was not a fair election, even before the shenanigans of polling day. Museveni used his control of the state to threaten opponents, pay off supporters, and suppress scrutiny. In the circumstances, it is impressive that Besigye performed as well as he did.


Police vehicles on the streets of Kampala

Looking for fingerprints

The most interesting thing about this election, says Charles Mwanguhya Mpagi, was the state’s reaction: ‘There was no celebration, and from the moment the results were announced the government was on the defensive.’ Heavy deployment of security and the repeated arrests of Besigye suggest that the regime has something to fear.

The unusual mood that greeted the result is the mirror image of the atmosphere beforehand, when a Besigye victory seemed possible. Besigye’s claim that he won, only to be thwarted by rigging, is believed by many. In 2014, say IRI, a third of Ugandans believed it ‘very’ or ‘somewhat’ likely that the announced result would not reflect the actual voting; that figure may be higher now. The evidence of malpractice is clear, says Mugisha Muntu, the FDC party president: ‘this time round [the regime] forgot to put on their gloves, because they have left their fingerprints all over the place.’

And there are fingerprints, certainly: the chaotic organisation of the ballot, strong evidence of ballot-stuffing, and a huge number of invalid votes. But are these enough to account for Museveni’s 2.5 million vote lead in official figures? No. On the balance of probabilities, using currently available evidence, Besigye did not win this election.

He was never meant to. As Martin Mwondha of CEON-U points out, ‘political parties were kept in limbo for 20 years after the NRM came to power’. The opening to multiparty democracy since 2005 has not been accompanied by the institutional changes that would create a level playing field. Misuse of state resources, corruption of institutions and the threat of violence were used to buttress the position of the NRM. ‘By the time we reached election day,’ says Mwondha, ‘the chances of Besigye winning were very slim.’

Does it matter how, or when, the election was won? Yes, for two reasons. Firstly, there is the court challenge ahead. To annul the election, Mbabazi’s lawyers must show either that Museveni was personally responsible for electoral offences, or that electoral malpractice ‘affected the result of the election in a substantial manner’. Simply deeming the contest to be unfair is not sufficient grounds for annulment. In 2006, for example, the judges unanimously found that the election had not been ‘free and fair’, but a majority of 4-3 ruled that the result was not substantially affected. They upheld a Museveni victory, as they had also done in 2001.

Secondly, and more importantly, Besigye’s claim to be the true president-elect will shape the politics of the streets. He is calling for non-cooperation with government and mass defiance. Whether that will materialise, and on what scale, is still unclear. But there is a world of difference between merely decrying the unfairness of the election and claiming to be the winner – the legitimate, de jure ruler of the country.

More evidence will doubtless emerge in the coming months. Some will be shown in court; some will be uncovered in the final reports of election observers, or through the investigations of journalists. The FDC are calling for an independent audit of the whole process. They are extremely unlikely to get one. But the debate over what happened in this confused, confusing election will not go away: it lays the foundation for Ugandan politics in the years ahead.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


This entry was posted on March 4, 2016 by in African Politics and tagged , , , , , .
%d bloggers like this: