Final Report: 2017 Kenya General and Presidential Elections, March 7, 2018

Originally published
View original



Kenya’s 2017 general electoral process was marred by incidents of unrest and violence throughout the extended electoral period, and by harsh attacks by top political leaders on electoral and judicial authorities that seriously undermined the independence of the country’s democratic institutions and the rule of law. The confrontational tactics and actions of Kenya’s political leaders polarized the country and exposed the deep tribal and ethnic rifts that have long characterized its politics. Regrettably, the elections represent a major setback in Kenya’s democratic development.

Overview of Key Findings

Pre-election Period. The pre-election period included a reasonably adequate but flawed voter registration process, which included an independent audit and corrective actions to address some of the many errors in the list. While the voter list retained serious problems, the introduction of the KIEMS biometric voter identification system provided a strong safeguard against multiple voting on election day. The level of political competition in the various races was high, and there was a significant increase in the number of independent candidates.

Aug. 8 Election. Carter Center observers found that the voting and counting processes during the Aug. 8 election were generally well-administered. However, observers noted problems during the subsequent processes of electronically transmitting polling station results and tabulation of results at county-level tallying centers. While these were designed to allow for thorough checks and verification of election results, and should have been possible to implement, the process of tallying and tabulating final results suffered from delays and a lack of transparency. The situation was made worse by heightened political tensions and confrontational statements by political leaders.

By the morning of Aug. 9, the day after the elections, as the Independent Electoral and Boundaries Commission (IEBC) began announcing provisional results showing President Uhuru Kenyatta with a roughly 10 percent lead, opposition presidential candidate Raila Odinga and his National Super Alliance (NASA) claimed that there had been massive fraud and that the results had been hacked. On the afternoon of Aug. 9, the IEBC’s nearly complete provisional results showed President Kenyatta with 54 percent of the vote, leading by a margin of some 1.4 million voters over opposition presidential candidate Raila Odinga, with about 45 percent of the vote. Although the IEBC noted that the process of tallying and submitting official results forms to the national IEBC center was proceeding at a slow pace, it did not highlight or clarify that the results broadcast on TV were unofficial provisional results received electronically directly from polling stations. Unfortunately, for unexplained reasons, the IEBC did not utilize the full seven-day period provided by the law to consolidate and post all of the official polling-station results forms.

Instead, the IEBC hastily declared the final presidential election results on Aug. 11, just three days after election day, based on the constituency-level results forms, and prior to the receipt of all polling-station level results forms. Worse still, election authorities failed to ensure that parties had timely access to official polling-station level results in the days following the announcement of official results, which made it impossible for parties and observers to fully verify and cross-check the results against their internal data and reports in time to include any key evidence in court petitions.

While the lack of IEBC polling-station data hindered the parties’ ability to verify results, it is important to note that a Kenyan citizen election observation organization (Election Observation Group, or ELOG) conducted a parallel vote tabulation (PVT) that provided an independent verification of the official results. ELOG’s PVT was based on results data gathered from a representative random sample collected by about 1,700 observers deployed around the country. The PVT’s estimated results, which were released on Aug. 11, were consistent with the IEBC’s official results.