announced by the ZEC Chairperson on 15 June with 5,681,604 names in total (78.6% of the estimated eligible population). In reality the roll was not made available for sharing until 18 June. Candidates complained that printed copies were not available at that time, but the voter roll was shared as a searchable soft copy, in the form of a an excel database on a CD-ROM, with candidates, political parties and CSOs, which is the format foreseen in the law. Unfortunately, ZEC’s handling of the matter did not enhance transparency and the fact that it made the roll available late meant that the matter did not bolster the level of confidence of stakeholders in ZEC. Various NGOs and parties conducted audits in order to determine their level of confidence in the accuracy and inclusiveness of the final version33. Following a further series of corrections a revised final voter roll for the election was announced by the ZEC Chair on 9 July. It comprised 5,695,706 names, which is exactly 11,770 voters more than the roll distributed to stakeholders in June. The addition in Mashonaland West province alone was more than 10,000 voters. Overall, the capture rate of registered voters by district shows a few anomalies, including the under-representation of urban areas compared to rural areas (see below) and also cases of possible under-registration, in a few rural districts34 in Matabeleland North and South. Province Bulawayo Harare Manicaland Mash Central Mash East Mash West Masvingo Mat North Mat South Midlands TOTAL Final VR 2018 258,567 900,728 733,370 531,984 633,410 655,133 617,212 339,135 264,185 761,982 5,695,706 ZIMSTATS 2018 projected population 409,389 1,345,818 912,762 620,279 735,790 834,414 754,314 389,592 356,586 865,185 7,224,129 Percentage of registered voters 63.2 66.9 80.3 85.8 86.1 78.5 81.8 87.0 74.1 88.1 78.8 ZEC data shows significant differences between provinces (see table above), with the main urban areas of Harare and Bulawayo with markedly lower capture rates. The rest of the country, without the two metropolitan provinces, showed a capture rate of 82.9%, so the difference between Bulawayo and Harare with the rest of the country was 16%. The reasons for this may vary, but voter apathy may account for some of the variance. Reports from the Election Resource Centre pointed out that some registration centres in Harare closed early after achieving a specific number 33 USAid and the EU funded two external audits, respectively an IT-based audit on internal consistency and integrity of the database and a sample-based check, using both list-to-voter and voter-to-list methodologies. 34 Bulilima 58,6%, Mangwe 69,3% and Tsholotsho 68,8% – country capture rate is 78,8% using mathematical projection of the population. 15