UGANDA: BY-ELECTION GOES FROM SUCCESS TO SUSPECT

C O N F I D E N T I A L SECTION 01 OF 03 KAMPALA 000067 SIPDIS E.O. 12958: DECL: 2020/02/11 TAGS: PGOV PREL PHUM KDEM UG SUBJECT: UGANDA: BY-ELECTION GOES FROM SUCCESS TO SUSPECT REF: 09 KAMPALA 01278; 09 KAMPALA 01349 CLASSIFIED BY: Aaron Sampson, Pol/Econ Chief; REASON: 1.4(B),(D) ¶1. (C) Summary: A January 25 by-election in central Uganda underscored concerns about the voter registry and the transparency of polling results in advance of the February 2011 presidential election. Local media and observers described turn out as low, and the Electoral Commission uncovered one instance of ballot stuffing. Our own review of election results cast suspicion over returns from nearly one fifth of Budiope's 140 polling stations. While the apparent fraud may not have affected the outcome in favor of the ruling National Resistance Movement (NRM) candidate, similar practices could impact the outcome of a competitive and contentious presidential election in February 2011. End Summary. -------------- Budiope By-Election -------------- ¶2. (SBU) Citizens north of Jinja in central Uganda's Budiope sub-county voted on January 25 to elect a new Member of Parliament following the death of Budiope's former MP in a vehicle accident. Three EmbOffs joined with European Union colleagues and the International Republican Institute to field a small contingent of election observers. Budiope is a rural area without any paved roads or towns. EmbOffs visited polling places in village market areas, fields, school yards, and private compounds. All polling stations visited were outdoors and open to the elements. The NRM candidate won with 23,457 votes (75.6%). The opposition Forum for Democratic Change (FDC) candidate finished a distant second with only 6,210 votes (20%). Official turnout was 51%. The Electoral Commission spent approximately $34,000 (or $0.53 per registered voter) to organize the election. The Chairman of the Electoral Commission and three Commissioners were in Budiope to observe the exercise. ¶3. (SBU) The campaign period was peaceful and all materials seemed to be in place the evening before the vote. On election morning, several polling stations opened hours late due to tardy poll workers and/or lack of voters as Ugandan law requires a specific number of voters to witness the opening of each polling station. Few voters seemed to have voter cards, but voter registries include photographs and those without cards can vote provided they match the picture in the registry and someone at the polling station vouches for their identity. The Electoral Commission voided results from one polling station where the number of ballots exceeded the number of voters. Three poll workers were arrested, charged with electoral malpractice, and released on bail. ¶4. (C) Election day appeared clam and moderately well organized, although opposition leaders alleged intimidation and rigging. EmbOffs observed riot police, armed with weapons and tear gas grenades, in areas highlighted by the Commission as potential flash points. At one polling station riot police were leaning against the string demarcating a polling station when EmbOffs arrived. The police immediately melted away, but opposition party delegates and others complained that the police had been deployed the night before to depress local turnout and intimidate villagers. -------------- --- Voter Registry Problems - Missing Voters -------------- -------------- ¶5. (SBU) On election day, the FDC announced that voter registries sold to the FDC by the Electoral Commission on January 21 contained approximately 2,800 more names than the official registries distributed by the Commission to poll workers on January 22, and accused the Commission of surreptitiously deleting the names of FDC supporters. Observers confirmed that voter registries issued to the KAMPALA 00000067 002 OF 003 FDC differed from those issued to poll workers, but the deletions appeared to be more a function of the Electoral Commission's disorganization than a concerted plot to disenfranchise opposition voters. According to the Electoral Commission, the actual voter register was not finalized until January 23 and deletions in the days preceding the election involved the removal of deceased voters and those no longer living in Budiope sub-county. The Commission defended the registry "cleaning" process, stating that all legal guidelines and requirements were followed. -------------- Suspect Participation Rates -------------- ¶6. (SBU) Local newspapers, observers, and the Electoral Commission all reported low voter turnout. The Director of the European Union's $20m Deepening Democracy Project (DDP) noted numerous factors that depressed participation rates, including traditional disinterest in by-elections, a lackluster campaign period, a race regarded by many as a foregone conclusion in favor of the NRM, and the fact that the by-election winner must stand for re-election again next year. Official numbers placed the participation rate at 51%, which is remarkably good for a low profile by-election in a rural constituency. An analysis of participation rates by polling station, however, reveals that nearly 20% of Budiope's 140 polling stations reported turnout rates in excess of 88%, with several stations carried by the NRM recording 99% or 100% participation. Average turnout at the 29 stations won by the FDC was 38%. -------------- -------------- No Answers From Electoral Commission -------------- -------------- ¶7. (SBU) The U.S. Mission shared its assessment of the Budiope election with Electoral Commission's Secretary, Sam Rwakoojo, on February 8. We noted that while the Commission's last minute changes to the voter rolls did not appear to disenfranchise specific voters, the unfortunate timing raised suspicions that could have been avoided had the Commission shared the final list with all interested stakeholders. In response to questions about the surprisingly high participation rates at some polling stations, Rwakoojo said a high turnout for a presidential election would likely not exceed 70%, but claimed participation rates of 90 to 100% for an unheralded by-election in Budiope are not cause for concern. ¶8. (SBU) Despite advance assurances from the Electoral Commission that Rwakoojo would provide us with an updated budget for the 2011 elections, Rwakoojo said the Commission is still revising its budget for 2010 and 2011. In 2009, the Commission requested $98 million for election-related activities in 2010 and 2011. This is almost double what the Commission spent for elections in 2006. According to the DDP, the Commission spent $4.30 per registered voter in 2006 but is on track to spend $8.60 per voter in 2011 (ref. A). Rwakoojo said the Commission has revised its budget downward by approximately $3.5 million by cancelling and scaling back planned activities that can no longer be implemented before 2011, but said the budget could go back up as the Commission is still working on final calculations. The Commission has so far received one third of its budget request, or 61 billion shillings ($30 million),from the Ugandan government. ¶9. (SBU) Rwakoojo was unable to provide an updated roadmap for the 2011 election cycle. He claimed the Commission completed the field component of the polling station re-organization exercise per the roadmap's February 1 date, and said opposition parties participated in this process. Two FDC parliamentarians, however, told the Mission on February 8 that they were not aware of any attempt to reorganize polling stations in their constituencies and were not in KAMPALA 00000067 003 OF 003 contact with the Electoral Commission. ¶10. (SBU) Rwakoojo was also unable to provide additional information on the timetable for cleaning the voter registry, which along with the reorganization of polling stations is one of the election's most contentious issues. Rwakoojo said procurement difficulties delayed the registry clean-up. In November, the Public Procurement and Disposal of Public Assets Authority (PPDA) canceled a $15m tender issued by the Electoral Commission for a voter registration and biometric ID system due to irregularities in the bidding process and complaints that the proposed registration program would target only urban voters who traditionally support opposition parties (ref.B). Rwakoojo said the Commission is determined to move ahead with the procurement, despite the fact that it will consume nearly half the government's current allotment to the Commission. -------------- -------------- Comment: One By-Election, Two Lessons -------------- -------------- ¶11. (SBU) A transparent voter register is critical to the success of the 2011 election. The Electoral Commission did itself no favors by revising the registry just days before election day and failing to notify the FDC. On February 9, the FDC called on the Electoral Commission to post the registry to the internet and arrange for on-line registration of voters. On-line voter registration in Uganda, where few voters have access to the internet and many have no electricity, is impractical. Posting the entire registry on line, however, would increase transparency, get the Commission out of the business of selling the registry by the page to stakeholders, and enable political parties to monitor changes to the electoral list when they occur. On February 11, the local National Democratic Institute (NDI) representative estimated that posting the registry to the internet with a built-in search engine would cost approximately $25,000 and not pose any serious technological difficulties. ¶12. (SBU) The Commission should also set a cut off date for freezing changes to the registry prior to election day, and provide a second copy of the registry to post in front of polling stations on election day to reduce voter confusion and allow voters to quickly ascertain - without standing in line for hours - whether they are lined up in front of the correct polling station. ¶13. (C) The inflated participation rates suggest scattered collusion between poll workers and delegates from opposition parties. On February 8, two FDC parliamentarians admitted that party poll watchers are not well compensated and that some in rural poverty-stricken areas like Budiope are tempted to accept small pay offs to look the other way. Although this collusion did not impact the lopsided Budiope result, it could have a destabilizing effect should the 2011 presidential election be close and competitive. LANIER