Elections are supposed to flip power relations. During an electoral process, a country’s political elites remember their fundamental jobs and are – more than ever – the servants of the people. While campaigns are ongoing, the elites need voters’ support, their attention…and most importantly, their votes. During this time, it is voters who can sit back and evaluate their leaders, deciding whether or not their actions are deserving of another term in office. Over the years and around the world, however, this power structure has often been reversed. In the quest to win and/or retain power, political elites have managed to shape the electoral process to their advantage, creating loopholes and amending laws that dilute public power.
PUBLIC CONFIDENCE IN THE KENYAN ELECTORAL PROCESS
Kenya is no exception to this rule, and voters have taken note. Over the past fifteen years, the Kenyan public’s confidence in elections has dropped precipitously. In fact, between 2005 and 2015, the proportion of Kenyans who strongly agreed that elections were the best way to choose leaders decreased by more than ten percentage points. Unsurprisingly, faith in the credibility of elections has also suffered. In fact, there has never been a time over the last three election cycles (including the current one) when a majority of Kenyans has felt that the last election was completely free and fair.
With the next general election in Kenya scheduled to take place in less than three months, it is critical to think about how to urgently address this marked dearth in voter confidence. An important first step is the assessment of potential vulnerabilities. Reflection about what went wrong last time and what is at risk of going wrong again is useful, not only for policymakers but also for voters, who can and should take time to critically assess whether or not their electoral processes prioritize their roles and voices.
There are, of course, many issues to consider. These include poorly enforced electoral laws, delayed timelines, the lack of intra-party democracy, incidents of pre-election violence, and shrinking space for vigorous public debate on the most contentious and timely election-related issues. At this stage in the electoral cycle, however, it is most urgent to focus attention on the factors that most significantly impacted the credibility of the last election and which continue to haunt this election cycle. Together, unresolved questions regarding leadership and integrity, the Independent Electoral and Boundaries Commission (IEBC), voter registration and the process of counting and tallying threaten the credibility of the upcoming elections.
LEADERSHIP AND INTEGRITY
To begin, the 2013 election was the first to take place under the guiding principles of the country’s new, internationally lauded constitution, itself the result of a decades-long struggle. The constitution included many new provisions that would have a bearing on elections, but one of the most fundamental, overarching issues concerned the qualifications for and conduct expected of state officers. Indeed, Chapter Six of the constitution, devoted to leadership and integrity, is groundbreaking in the context of Kenyan political history, confronting, as it does, some of the most longstanding and deeply embedded obstacles to good governance.
Over the past fifteen years, the Kenyan public’s confidence in elections has dropped precipitously. In fact, between 2005 and 2015, the proportion of Kenyans who strongly agreed that elections were the best way to choose leaders decreased by more than ten percentage points.
Chapter Six fundamentally shifted the relationship between state officers and the people, requiring the former to selflessly serve the latter. Despite the dramatic weakness of the Leadership and Integrity Act that was passed to operationalize Chapter Six provisions, the lead-up to election day in 2013 did include bold efforts to test the letter and spirit of the law. The most notable of these was a lawsuit filed by the International Center for Policy and Conflict and five others, which asked a deeply controversial question: were Uhuru Kenyatta and William Ruto in violation of Chapter Six standards, and therefore ineligible to run for office, based on the International Criminal Court’s indictments against them for their alleged commission of crimes against humanity in the aftermath of the 2007 election?
The High Court’s judgment was disappointing, leaving the public doubting that state institutions were in fact committed to the standards of the constitution. Although the Court claimed that it lacked jurisdiction on matters related to the presidential election, its judgment did define the standard of integrity required by a person seeking public office. According to the Court, such an office-seeker should be beyond reproach and should not have unresolved questions about his/her character and commitment to the national values in the constitution. On the other hand, however, it also ruled that there is a distinction between appointive and elective positions. For the latter, the Court was of the opinion that it is up to the voters to decide who to choose, based on candidates’ “honesty, rectitude, uprightness and scrupulousness.” After the judgment was handed down, a lawyer for the petitioners remarked, “Today marks the official death of one of the chapters in the constitution. That is Chapter Six.”
Uhuru and Ruto were thus free to run for office, and Kenyans were told that questions about their character related to the indictments against them were “a personal issue.”
More importantly, since the substantive issues of the case were never appealed to the Supreme Court, Kenyans were left with an electoral context that was marked by the severe limitations of the candidate pool. After all, the Court had not considered that by attempting to leave it up to voters to decide how strictly candidates should be judged with regard to factors of integrity, it could well be that voters ended up with little substantive choice. With no enforcement of Chapter Six provisions for elective office-seekers, voters could well be faced with a slate of candidates, all of whose characters were tainted by integrity-related problems. The ruling also made it clear that there was little political will to apply the leadership and integrity laws across the board, thus cementing the status quo of elite impunity.
Unsurprisingly, unresolved leadership and integrity issues continue to plague this electoral cycle. In 2016, PricewaterhouseCoopers conducted a survey on the prevalence of economic crimes in the world and found that Kenya topped the list of 78 countries in the study. A shocking one percent of the country’s national budget had been properly accounted for in the previous year. Theft and misappropriation was the most common type of crime. As of 2016, there were 17 MPs who had been charged in court for committing serious criminal offenses, including fraud, forgery, hate speech, rape, corruption and incitement to violence. At least 90 others were under suspicion for graft.
The public has noticed. When asked how much they trusted the ruling party, the opposition and MPs, Kenyans reported significantly low levels of confidence.
Table 1: Trust in Parliament, the Ruling Party and Opposition Parties
| Not at All | Just a Little | Somewhat | A Lot | Don’t Know | |
| Parliament | 14.7% | 28.5% | 36.4% | 17.1% | 3.2% |
| Ruling Party | 16.3% | 21.0% | 31.1% | 29.9% | 1.6% |
| Opposition | 21.8% | 28.7% | 31.1% | 15.2% | 3.1% |
Source: Afrobarometer R6 2014/2015
Such findings are telling, and they are especially relevant in the context of upcoming elections. Given the lack of political will to seriously implement and enforce constitutional standards of integrity, the public has little reason to put its trust in the state, or in the electoral process. Without proper enforcement of Chapter Six provisions, voters’ choices are always restricted. Over time, such an environment can lead to increasingly low levels of public confidence. In the long run, this endangers democratic resilience.
In an attempt to address the gaps in Chapter Six enforcement, the IEBC recently convened what is known as the Chapter Six Working Group on Election Preparedness. The group, which includes several state institutions, plans to vet parties’ lists of nominated candidates to ensure that they adhere to the Attorney General’s recently published guidelines. It is unclear, however, what authority this group has to carry out its stated purpose, especially given that the Court’s ruling in the case against Uhuru and Ruto made it clear that the presumption of innocence holds until cases are concluded.
THE INDEPENDENT ELECTORAL AND BOUNDARIES COMMISSION
Closely linked to public confidence in elected leaders is trust in the body charged with administering elections, the Independent Electoral and Boundaries Commission (IEBC). In fact, one study found that public confidence in a country’s electoral management body (EMB) is tied to public faith in the credibility of the election. Specifically, in a paper on Nigerian elections, Nicholas Kerr found,
The strongest correlate of citizen’s perceptions of election quality is the performance of [Nigeria’s Independent National Electoral Commission (INEC)]…when citizens are highly satisfied with the performance of INEC, they are thirty-eight percent more likely to consider the 2007 elections free and fair… But even more importantly, it highlights that ordinary Nigerians associate their evaluations of EMB performance more with their assessments of election quality, despite how strong their partisan attachments may be.
In Kenya, public confidence in the IEBC has fluctuated dramatically over the last five years. At the end of 2012, surveys showed that 70 percent of Kenyans had faith that the IEBC was carrying out its work impartially and 72 percent believed in the Commission’s independence. One month before election day in 2013, 91 percent of respondents said that they had faith in the IEBC’s competence to manage the election and 89 percent believed in the Commission’s impartiality. 91 percent of respondents also said they believed the IEBC was independent enough to conduct the election in a free and fair manner.
These pre-2013 election levels of confidence were a high point for the Commission, which suffered from plummeting levels of public faith after the 2013 election. The drop in confidence was due to a combination of factors, the most notable of which included procurement delays and irregularities in the lead-up to election day, problems with technology during voter registration and on election day, the failure of the electronic results verification system and the subsequent delay in announcing the result. The Supreme Court case challenging the integrity of the entire process, filed by the Africa Centre for Open Governance (AfriCOG) and the Kenyan Asian Forum, also publicly highlighted the multiple technical and administrative problems throughout the process. Indeed, post-election polls revealed a precipitous drop in public confidence in the Commission. In the immediate aftermath of the election, 44 percent of Kenyans reported that they were confident in the IEBC. In Nyanza, only 8 percent of respondents felt this way. By 2015, the figures had not changed dramatically, with 43 percent of Kenyans reporting confidence in the IEBC. Among opposition supporters, 71 percent reported that they did not have confidence in the Commission.
After the judgment was handed down, a lawyer for the petitioners remarked, “Today marks the official death of one of the chapters in the constitution. That is Chapter Six.”
The IEBC has experienced significant changes since the 2013 election. Senior members of the Commission were implicated in a corruption scandal, and a series of opposition protests against the IEBC eventually resulted in the dismissal of all the commissioners. New commissioners assumed office in late December 2016.
The changes seem to have made some difference. A recent poll shows that 72 percent of respondents feel that the IEBC is prepared to conduct a free and fair election. On the other hand, Kenyans are also extremely cautious in their outlooks. In fact, in four of Kenya’s former provinces, more than 20 percent of the population is not sure that the Commission will be able to administer a credible election.
While the high level of public confidence is encouraging, the new IEBC stands at an important crossroads. In order to maintain public faith, the Commission will have to work to show Kenyans that it is serious about avoiding the mistakes that marred the last process and that it is willing to fight to maintain its independence. The signs so far are mixed. The Commission’s decision to cancel the tender process for election technology and engage in a direct award of the contract to the same firm that was partially responsible for the previous set of botched technology raises questions about how well the IEBC has learned from past mistakes. This is compounded by a more recent announcement that the IEBC may proceed with a direct award to a ballot-printing firm. Moreover, recent analysis of the last mass voter registration exercise has revealed serious administrative and technical irregularities. On the other hand, the Commission’s stated commitment to enforcing gender parity in party lists and to enforcing leadership and integrity standards in the vetting of candidates is admirable.
The IEBC is making certain attempts to keep the public updated. It does hold press conferences, and it regularly updates its website with relevant press releases. This information is useful, but the Commission must go further with regard to transparency if it wishes to maintain public confidence. There are several outstanding questions at this stage of the electoral process, the most urgent of which are related to procurement, voter registration, the ongoing audit of the voters’ register, the use of technology, and counting of results. If the IEBC begins to address some of these concerns, it could go a long way in preserving public faith, especially as it is likely that problems will continue to arise as election day gets closer. No election is perfect, but the IEBC’s honest evaluations of its strengths and weaknesses related to current concerns are critical.
VOTER REGISTRATION
In 2013, much of the public’s dissatisfaction with the IEBC was rooted in problems with the voters’ register. The register was shrouded in a certain amount of mystery, with the total number of registered voters in Kenya shifting throughout the electoral cycle. The first sign of the problems to come appeared in February 2013, when it became clear that the final, gazetted register differed significantly from the provisional register released in December 2012. Overall, the register had grown by 12,500 voters.
Without proper enforcement of Chapter Six provisions, voters’ choices are always restricted. Over time, such an environment can lead to increasingly low levels of public confidence. In the long run, this endangers democratic resilience.
While a decrease in the number of registered voters was expected (because the verification and cleaning process would expunge dead voters, multiple registrations, etc), it was unclear how the register grew in size between December 2012 and February 2013. Moreover, there were significant regional changes in the numbers between December and February. These are detailed in the table below.
Table 2: Internal Changes to the Register of Voters
| Region | Changes between December 2012 and February 2013 |
| Coast | +901 |
| Nyanza | -15,026 |
| Central | +1,848 |
| Rift Valley | +67,000 |
| North Eastern | +6,604 |
| Western | +2,938 |
| Eastern | +4,222 |
| Nairobi | -50,102 |
Source: AfriCOG/KPTJ. 2013. “Voter Registration for the 2013 General Elections in Kenya.”
These changes became more worrying when the IEBC could not commit to one total number of registered voters. In fact, there were at least six different totals announced during various parts of the electoral cycle.
Table 3: Shifting Totals of Registered Voters in Kenya
| Provisional Register (December 2012) | 14,340,036 |
| Principle Register (February 2013) | 14,352,545 |
| Special Register (March 2, 2013) | 36,236 |
| Election Results Total (March 9, 2013) | 14,352,533 |
| Green Book Total | 14,388,793 |
| Post-Election Register (July 2013) | 14,388,781 |
Source: AfriCOG/KPTJ. 2013. “Voter Registration for the 2013 General Elections in Kenya.”
This shifting total, in addition to the IEBC’s assertion that what was known as the “green book” (A green book is an unregulated, manually-recorded list of registered voters. It had been severely criticized by experts.) was being used for purposes of registration, severely compromised public confidence in the integrity of the register. Indeed, the lack of a single, verifiable register breeds suspicion about political influence at worst and basic incompetence of the electoral management body at best.
Doubts around the register have not faded. In fact, the IEBC’s two mass voter registration exercises in the current cycle were rife with problems. These included widespread problems obtaining IDs, problems with dysfunctional and nonfunctional biometric voter registration (BVR) kits, unexplained use of the green book, disorganized registration centres and poorly trained IEBC staff, registration bribery, coercive registration practices and massive amounts of transfers. At the end of these processes, the IEBC announced that the total number of registered voters had grown to 19,749,310, representing a 37 percent increase since 2013.
In the immediate aftermath of the election, 44 percent of Kenyans reported that they were confident in the IEBC. In Nyanza, only 8 percent of respondents felt this way. By 2015, the figures had not changed dramatically, with 43 percent of Kenyans reporting confidence in the IEBC. Among opposition supporters, 71 percent reported that they did not have confidence in the Commission.
The elections law also allows the IEBC to engage a professional firm to conduct an audit of the voters’ register. The stated purposes of such an audit are to verify the accuracy of the register, recommend mechanisms to enhance its accuracy and to update it. While an audit may go a long way in promoting public confidence in the register, the process has thus far been controversial. In addition to allegations that there were irregularities related to the decision to hire KPMG, there is mistrust because of the firm’s lack of expertise in conducting such audits. Indeed, KPMG’s proposed methodology does not reflect internationally accepted best practice for the audits of voters’ registers, and there has thus far been a lack of transparency with regard to KPMG’s progress and therefore its eventual findings.
In addition to these technical problems with registration, the legitimacy of the entire process was cast in doubt when it emerged that the IEBC had cancelled a public tender process for the acquisition of an integrated elections management system in order to give a direct award to a company known as Morpho, the same company that provided the problematic BVR kits in 2013.
The doubts that have arisen as a result of the above issues have been compounded by the IEBC’s refusal to make the register publicly accessible. In fact, the IEBC did not even provide the updated register to political parties during their primaries. The lack of the register made it impossible for parties to confirm that voters were in fact registered, and this contributed to the chaos that characterized the primary processes. The IEBC has also refused to give the data to civil society on the grounds that it cannot release it until after the audit is complete. Without access to the pre- and post-audit data, however, it is impossible for the public to conduct its own analyses and understand the changes.
A related issue is that of the integrated election management system (IEMS) itself. According to the IEBC, the system is meant to be wholly integrated, such that voter registration, voter identification and results transmission are linked. Since the IEMS technology did not arrive until well after the conclusion of registration, however, it is unclear how all the components will be linked, if at all.
ELECTION DAY, COUNTING AND TALLYING
Confidence is also, of course, based on the credibility of results announcements. In 2013, the IEBC used an electronic results transmission system, which was designed to allow polling station officers to transmit results to regional tallying centres and to IEBC headquarters in Nairobi via a secure, digital connection. This system was meant to protect the credibility of the count and prevent the kind of manipulation that had been seen in past elections, which often occurred during the time when tally sheets were being physically transported from polling stations around the country to Nairobi. Unfortunately, however, the system was a spectacular failure. Midway through the counting process, Kenyans watched the stream of live results freeze on television screens. Journalists based at the Bomas of Kenya, which was the national elections centre, referred to it as the Bomas screen saver.
When the electronic system failed, the IEBC again relied on the paper forms, which had to be ferried from all over the country to Nairobi. As expected, the paper forms were highly problematic. Issues with the polling station-level tallying forms (Forms 34) included:
- Many Form 34s showed that there were more votes cast than registered voters. In Turbo constituency, Polling Station 69, Stream 2, there were 784 votes cast but only 755 registered voters. In Polling Station 71, Stream 2, there were 741 votes cast but only 716 registered voters. In Kacheliba, Polling Station 112, there were 215 votes cast but only 214 registered voters.
- In some Form 34s, only some presidential candidates were listed. For example, in Baringo South, Polling Station 91, Stream 1, it was only the names of Uhuru Kenyatta, Raila Odinga and Paul Muite that appeared. Some candidates were also missing from Form 34 in Baringo South, Polling Station 68, Stream 1.
- Many of the figures on the form did not add up. For instance, one of the most glaring discrepancies occurred in Kacheliba constituency, Polling Station 102. Here, the votes cast are recorded as 0, while there are 170 rejected votes and 170 valid votes. In Baringo South, Polling Station 117, Stream 1, there were 133 valid votes and 0 rejected votes, which should total 133 votes cast. The figure for votes cast, however, was 134. In Cherangany, Polling Station 2, Stream 5, the number of valid votes is 332 and the number of rejected votes is 4, which adds up to 336 total votes cast. The number of votes cast, however, was 340. In Turkana North, Polling Station 12, the number of votes cast (340) did not equal the number of valid votes (340) plus the number of rejected votes (5).
- There were several instances of changes having been made to various figures on the form, with no authorizing signature next to the change. Such alterations affected individual candidates’ results, the total number of votes cast, the number of rejected votes, the number of valid votes and the number of registered voters. This change had the potential to affect other numbers on the form. For instance, a change to the number of rejected votes would necessarily change the number of total votes cast.
- In some forms, there was no figure indicating the number of registered voters. There was no official Form 34 for Polling Station 19 in Turkana Central. Instead, the results were reported on an ordinary piece of paper, which did not include the number of registered voters.
- Many Form 34s were missing. There was no Form 34 for Polling Station 84 or for Polling Station 99 in Turkana North. Polling Station 99 did not appear on the list of polling stations published on the IEBC website on February 24, 2013, but it did appear in the paper gazette. Forms 34 for Polling Stations 92 and 113 in Turkana Central were missing.
- Some forms did not include results for certain candidates. In Turkana Central, Polling Station 55, there were no results listed for Muite and Kenneth. In Turkana Central, Polling Station 65, there were no results listed for Kiyiapi, Karua, Dida, Muite and Mudavadi.
- There were non-identical duplicates of certain forms. In Turkana Central, Polling Station 89, there are 4 nearly identical copies of Form 34. It is unclear whether the numbers on these forms were counted multiple times. In Kacheliba, Polling Station 2, there were two forms with different entries. There were also multiple copies of perfectly identical forms, and it was unclear whether or not these figures were counted more than once.
- There were many forms in which it was extremely difficult to determine the exact value of the written figure, either because of the handwriting or because the original figure had been written over with another number. There were an overwhelming number of such cases, and the choice of which number to report was subjective.
The most worrying issues called the very accuracy of the count into question.
The failure of the electronic system was not a complete surprise. In the month before the 2013 election, tests of the electronic systems revealed significant problems. In fact, Sarah Elderkin detailed how a test of the system “had gone horribly wrong.” After one hour, only one of five mock polling stations could successfully transmit results. In this election cycle, the IEBC again plans to use an electronic system. The IEMS, mentioned above, includes results transmission. It is unclear, however, if and when the IEBC will publicly test the kits and publicly explain its plans for the kits’ dysfunction or failure. In fact, one of the most pressing unanswered questions in this cycle is related to proposed back up systems. New amendments to the electoral law allow for the use of complementary registration, identification and results transmission mechanisms, to be used in the event that the technology fails. To date, however, the election regulations only provide vague provisions about using the printed out register for voter identification in cases where voters cannot be found in the biometric list. The regulations do not clarify what, if any, complementary systems will be used in case the electronic results transmission system fails.
The lack of a clear definition of these complementary mechanisms is highly problematic, especially given Kenya’s political history and context. During mass voter registration, the IEBC used the green book in conjunction with the BVR kits. Does this mean that the green book was the complementary mechanism with regard to voter registration? Will the green book be used in addition to the printed out register to identify voters on election day?
Midway through the counting process, Kenyans watched the stream of live results freeze on television screens. Journalists based at the Bomas of Kenya, which was the national elections centre, referred to it as the Bomas screen saver.
There is one significant difference with regard to the law in the current election cycle. The High Court recently ruled that constituency level results for all elections are final and can only be appealed through a court process. The decision nullified Section 83(4) of the General Elections Regulations, which empowers the IEBC to “confirm” results before announcing them as final. The IEBC filed an appeal to the decision, with IEBC Chair Wafula Chebukati stating that constituency level officers could “make mistakes.” The IEBC’s decision to appeal has sparked some controversy, with the opposition threatening that there will be no election if the IEBC does not abandon its appeal and alleging that it means the Commission wants the power to unfairly change results. Civil society has also expressed its reservations about the appeal, suggesting that it erodes public confidence in the IEBC’s commitment to upholding the law.
Given the extreme controversy and suspicion surrounding the announcement of results in 2013 and in other past elections, the IEBC should acknowledge that there is significant public concern around the potential use of manual systems. To promote public faith, the IEBC should explain its rationale regarding the appeal. If constituency level results are erroneous, as Chebukati fears, a court process to address such problems would allow the public to see and understand the issues at hand. It would promote transparency. Surely, this option is better than a closed process in which the IEBC changes constituency results at the national tallying centre.
As it currently stands, it is unclear whether the IEBC and other stakeholders have learned from past elections. If public confidence is a priority, these stakeholders must immediately respond – at minimum – to the above public concerns in an honest and open way, remembering that it is voters who hold the power at this stage of the game, and it is voters who will ultimately decide the credibility of the election. The legitimacy of the upcoming election now hangs in the balance, but there is still time to save it…if only we are willing to learn from the past.
