In the Ugandan elections of February 18, 2016, President Museveni won convincingly with over 60.75 per cent of the vote. Or did he?
The main opposition leader, Kizza Besigye, alleges that the elections were rigged. Election monitors have been extremely critical, but European Union observers have so far refused to categorically state that the elections were not free and fair. So how clean were the polls? And why is it so hard to tell? These are important questions, because they have implications for how we monitor and evaluate future elections – including the Kenyan polls scheduled for August 2017.
What’s a free and fair election?
The official results in Uganda reveal that the Besigye performed significantly better than in 2011. According to Jamie Hitchen’s analysis for the Africa Research Institute, he won 14 of 112 districts — three times more than in the previous election — and dominated the vote in Kampala.
Moreover, there is evidence that Besigye should have done much better. From high levels of intimidation during the campaign to the fact that the opposition leader was arrested multiple times, it is clear that there was not a level playing field. But deciding whether the elections were “free and fair” is complicated by the fact that different groups apply this standard in very different ways.
Academics and civil society organisations tend to understand “free and fair” in a fairly broad way that asks whether the political landscape was so biased that it significantly impacted the result. On this basis the Ugandan elections were clearly not free and fair. The NRM used state officials and funds to support its campaign and outspent the opposition about 8 to 1.
At the same time, the media was clearly biased in favour of the incumbent, opposition activists were harassed by the security forces, and the Chair of the Electoral Commission publicly stated that Besigye was not “presidential material”. There can be no doubt that the opposition would have performed considerably better in the absence of these constraints.
International election observers tend to use a much narrower standard that gives less weight to the broader context and focuses more on the process of the casting and counting votes. Other flaws will be listed in great detail in the main bodies of their reports, but are unlikely to sway their overall evaluation. In other words, if no irregularities show up in the tallying process it is very rare for international monitors to condemn the process, however problematic the rest of it might have been.
Foreign observers also tend to take into account whether or not electoral manipulation changed the outcome of the election, rather than simply evaluating the quality of the polls. If it seems likely that the incumbent would have won anyway, monitors tend to pull their punches.
The reason for this is simple: Few international actors have the stomach to take on a sitting government if ensuring that the “correct” results are respected would not actually lead to a transfer of power. Based on this approach, it is far less clear whether the Ugandan polls were “free and fair” — which is why, despite everything that has happened, the EU mission is still sitting rather awkwardly on the fence.
There are two fairly obvious problems with the approach taken by election monitors. The first is that it does a disservice to opposition parties, who fight elections with one hand tied behind their backs only to be told that the polls have been given an international stamp of approval.
The second is that it gives authoritarian ruling parties an incentive to try and exaggerate the margin of victory, in the knowledge that this is likely to mute any criticism from the international community.
By far the easiest and least controversial ways to fix the political process is to begin early. Effective authoritarians start by manipulating the voter registration process so that opposition supporters are less likely to be able to vote.
This has been a perennial complaint in both Uganda and Kenya, where the opposition alleges that the government has disenfranchised voters, either through the use of repression and intimidation (Uganda), or by using employing unnecessarily demanding criteria, such as the requirements for citizens to have national ID cards in order to be able to register (Kenya).
The advantage of these kinds of approaches is that they happen well in advance of the arrival of election monitors. Although the European Union now deploys long-term observers to pick up on this kind of activity, it still tends to be given less weight in their final reports.
Other strategies are easier to detect and more high profile, and are thus more dangerous. Ballot box stuffing and manipulating the tallying process are two obvious examples that, if they are exposed, can trigger political unrest and international condemnation — as was the case in Kenya in 2007. For this reason, governments typically use these strategies only when they need to. So what is the evidence that Museveni felt the need to embrace such tactics in 2016?
Initially it looked as if President Museveni might have an easy time of it. Early polls suggested that he was on course to match the 66 per cent of the vote he won in 2011, while many commentators wondered if Besigye, contesting his fourth election, had the energy for the fight. But over the course of the campaign Besigye’s Forum for Democratic Change (FDC) made considerable headway.
Hundreds of thousands of people turn up to its rallies, and many supporters donated to its campaign. All of this built up into considerable momentum that Besigye managed to sustain until election day.
Growing public support for Besigye put Museveni under pressure, because the elections were not simply a contest between rival parties; they were also a referendum on the president’s thirty years in power. Senior NRM leaders are becoming increasingly worried about public attitudes to Museveni’s long stay in office, especially given high levels of support for presidential term-limits and the fact that the party has no meaningful succession plan. It was therefore important for Museveni’s authority not just to win, but to win big.
This concern appears to have been the reason that the Electoral Commission took the unusual step of announcing the result on Saturday afternoon, despite the fact that returns from 1,787 polling stations, many of which were in opposition strongholds, had yet to be counted.
When the result was declared, Museveni had 60.8 per cent of the vote, and the rushed announcement appears to have been designed to ensure that the media would report the figures early, minimising the potential embarrassment if the president slipped below 60 per cent when all votes were cast.
MANIPULATE TALLYING PROCESS
This was symbolically important, because the only time Museveni has failed to secure more than 60 per cent of the vote was in 2006, a discredited election that many believe was actually won by the FDC. We, therefore, have good evidence that the president felt the need to win big- but did this lead him to manipulate the tallying process?
How do you prove that an election was rigged?
Under the Ugandan electoral law, the FDC was given 10 days to file an election petition. Ultimately, it failed to do this, not for lack of trying, but because of the numerous barriers that were put in its place. First, the clock started not when all the votes had been counted, but on Saturday when the figures from many polling stations were missing.
Second, the commission only released a (almost) full set of results towards the end of the period, reducing the time for cross-checking and verification. Third, Besigye was arrested and the police stormed the party’s headquarters, removing essential materials.
As a result, it was practically impossible for the FDC to prove that the election was rigged with the kind of evidence that would stand up in a court of law. Significantly, this is not an isolated example: A similar thing happened to the Coalition for Reform and Democracy in Kenya in 2013.
We can only speculate as to how much Museveni has learnt from his Kenyan counterparts during their East African Community meetings. However, one thing is clear: As the strategies adopted by ruling parties become more sophisticated, it is becoming harder to detect and prove electoral fraud.
This is not to say that we do not have clear evidence of major failings. The official results include at least 28 polling stations in which 100 per cent of voters turned out, and all of them voted for Museveni. Strictly speaking, this is not an election offence, but neither is it credible.
The problem for the FDC is that the official margin of victory is so wide that examples of malpractice here and there will be insufficient to persuade international observers to condemn the polls. As Jamie Hitchen has written, the 9,181 votes cast in these areas will not overturn a 25 per cent gap, however they were counted.
The outcome in Uganda demonstrates the limitations of international election observation. The narrow lens adopted, combined with the Museveni’s high vote share, has meant that the EU team have been unwilling to state what everybody knows: Whoever won the most votes, the elections were anything but free and fair.
Dr Cheeseman teaches African politics at Oxford University; @fromagehomme