Below are excerpts from the international observers report on the voter registration process carried out in the Republic of Somaliland. (Please find the complete document attached.)
The Somaliland voter registration process began on 16 January 2016, concluding on 26 September 2016, and was conducted by the Somaliland National Electoral Commission (NEC) in all six regions of the country. International development agency Progressio, the Development Planning Unit (DPU) at University College London (UCL), and members of Somaliland Focus (UK), observed the process at different points during this period. Our international election observer (IEO) mission followed previous observations by these three organisations of Somaliland’s House of Representatives
election in 2005, presidential election in 2010, and local council elections in 2012.
Perceptions of the 2016 voter registration
Our observations on perceptions do not purport to be representative of the Somaliland population, as they are based on a limited number of interviews. Nonetheless, we feel that the views expressed offer a valuable insight into some possibly significant patterns.
Many of the registrants we interviewed felt that the experience of the 2016 voter registration was far better than previous political processes, particularly the 2012 election where multiple voting was widespread. A number of people also noted that they felt that an honest registration process should lead to a credible election.
“It is different because the last one we registered many times so that the political party we wanted to win could have more votes. I registered 5 times and I told everyone I know to do the same. This process is better, though. It is more honest and you can only register once. Hopefully, we will get a good and honest president from this process.”
Older woman, Gacan Libaax VRC, Hargeysa, Maroodi Jeex
However, a few individuals from amongst both registrants and VRC staff suggested that the voter registration process was too bureaucratic. Others complained that elections were not leading to actual change for citizens, referencing dissatisfaction with government service delivery. These opinions corresponded with perceptions of corruption and ‘clannism’ in Somaliland’s political processes.
“If everything goes well we will hopefully get an honest president and parliament. It
seems as if the current parliament does not care about poor people and those that are
struggling. They swore on the Quran to help develop the country but there are still people that are struggling. People that do not have houses or access to water. To me it seems as if they have given up on those promises. Somaliland has become a corrupt country.
I wanted to register my vote so that I can put forward an honest candidate for the
presidential election. One that will help the poor people, one that will see the potential of all the young students in our universities. It is their country and it’s wrong that they cannot find jobs in their own country when foreigners can.”
Young man, Biyo Dhacay VRC, Hargeysa, Maroodi Jeex
Somalilanders’ perceptions of voter registration are varied, however, and depend on a number of factors, including age, location and clan affiliation. In general, younger interviewees tended to be sceptical and less informed about the registration process. They were also more likely to express concerns about data security and the possible (mis)use of their personal details. We gained the impression from interviews that a proportion of younger registrants were concerned that registration details might somehow disadvantage them if they were to find work overseas after (legally or illegally) travelling outside Somaliland. Some expressed concern that registration might
compromise their chances of securing refugee status in Western countries, for example.
Given the history of political opposition in the east of the country, respondents in these areas, and particularly in Sool, were initially more inclined to distrust voter registration as a whole. This clearly related to mixed feelings towards Somaliland itself. It is notable that, in particular amongst Dhulbahante respondents, initial levels of distrust reduced as the registration in Sool continued. This was due to the active participation of traditional leaders in voter education and mobilisation by the newly appointed Minister for Interior (who is from the region) encouraging people in Sool’s rural areas to come out and register. The fact that the government appointed an interior minister from
Sool increased levels of trust in the process, and also fostered some sense of competition amongst Dhulbahante sub-clans to mobilise registrants.
In other areas, there was generally a high level of enthusiasm for voter registration, with many respondents commenting that they considered it an essential step towards legitimate elections. One interviewee even hoped that it might lead to international recognition, though another felt that recognition in itself was less important than the display of Somaliland’s capacity for self-reliance.
“I am here today for my country. For registering to vote in the election and for the voter card. So that I can vote for an honest president.”
Young woman, Ceerigaabo
Others felt that registration held personal significance, bringing with it the possibility of an ID card that would permit greater freedom of movement, travel to neighbouring countries and identification at banks and schools. These opinions highlight the difficulty posed by the near simultaneous process of registering for a national ID card (the MINT’s ‘civil’ registration process), which could be used for these purposes, and voter registration, which leads to an ID card for voting only. Indeed, we observed widespread confusion as to the differences or complementarities between the civil and voter registration processes.
There was significant evidence in respondents’ comments that awareness of the practicalities and purpose of voter registration was patchy. We gained the impression from interviews that the awareness-raising aspect of the registration had not been as effective as it could have been. Many respondents did not understand the cyclical nature of the 28-day registration period – specifically, when VRCs would be open in or near their locations, and when they would close. Unsurprisingly, this kind of confusion was most prevalent in more remote locations, and in Sool, and there was some specific indication that the now customary methods of announcing a VRC’s presence (notices
in newspapers and on broadcast media, or a car with a loudspeaker mounted on the roof) had proven ineffective, especially amongst young, eligible registrants and those living in rural areas.
Many were also unaware of any plans for the distribution of voter cards once registration was complete. This was unsurprising, as the NEC plans in this regard were themselves unclear. However, this uncertainty had the worrying consequence of leading registrants to surrender their registration certificates to elders for ‘safe keeping’. Some interviewees suggested that the individuals collectingb the certificates believed they could ‘sell’ them to one or other political party once cards were ready
for collection (see Voter verification).
Assessment of the process
The NEC and its partners designed the 2016 BVR system to be utilised at over a thousand voter registration centres on a regional rolling basis. This required that the process be identical across time and space, replicating each registrant’s experience to ensure that each eligible voter would be registered accurately and uniquely.
Each VRC was an outpost of the central NEC operation, equipped with a number of standardised biometric registration kits (two or three on average) with which trained NEC personnel could register eligible voters. These client kits were connected via an online network to the main NEC server so that deduplication (checking that a person was not already registered) could be immediately undertaken. Each BVR kit contained a laptop, camera, spotlight, iris scanning device, and printer.
The process of registration was as follows. Prospective registrants queued to enter the VRC. The first step was to have their eligibility verified by a MINT accredited official (see Voter verification). If they passed this stage, a registrant moved to the first station to be interviewed. This necessitated providing their biographical details, including name, age, year and place of birth, and telephone number, and indicating their preferred polling centre for election day. The registrant then reviewed and signed/made a mark upon the form to confirm the details were correct (or had the form read to them if they were illiterate). The registrant then took the form to one of the BVR stations,18 where the operator entered the
details from the form into the electronic registration system. A digital version of this profile was then printed out, and manually checked by the registrant and confirmed by a second signature/print.
The registrant then posed for a photograph of their head and shoulders, to be used for in-person recognition (not facial recognition software, as in 2008); and an image of the iris of both eyes was scanned (see Iris scanning). These two processes generated a third form carrying the registrant’s biographic details, photograph and a barcode, which was then reviewed and confirmed by the registrant by signature/print, with a duplicate detached as receipt of registration. All three official forms were finally stapled together and kept as a physical record of the registration to be sent to the central NEC office. Registrants left with the paper record, and were told that they would receive an SMS when their voter card was ready for collection.
Recommendations
Release of voter register data
• Our most significant recommendation relates to the eventual release of registration data. The process will be politically sensitive and must be handled with care. However, it is important that data to district level is released promptly so that discussion on siting of polling stations, parliamentary seat allocation and related election-management issues can take place in an informed context.
Operation of voter registration centres
• As with the design and management of polling stations, care must be taken to ensure that every registration centre is fit for purpose, with sufficient size and space for registration personnel to work easily and effectively, and to ensure registrants’ privacy and comfort.
• The NEC has very detailed instructions regarding how centres are to be set up. In order to ensure this is followed in all centres, there is a need to adapt training procedures of centre staff in order to ensure that practices are standardised as much as possible across VRCs.
• Electoral authorities should utilise previous registration and voter turnout statistics to ensure that there are sufficient biometric registration kits at each registration centre, and consider the ratio of interviewers to operators to ensure that queues do not form inside registration centres, potentially compromising privacy.
• Strict adherence to rules regarding official opening and closing times must be observed, without external influence or interference. To ensure that registration personnel are not overtired, electoral authorities should design the registration schedule to permit official breaks for prayer and refreshment. This timetable should be clearly communicated to registrants and NEC staff to ensure that no one is excluded from registering during publicised operating hours.
• The provision of adequate covered space for police officers is an important consideration in the location and organisation of VRCs in order to ensure that the police presence complies with NEC regulations.
• An online connection must be secured wherever possible to ensure that immediate deduplication is undertaken and thus prevent the possibility of multiple registrations or data cleaning backlogs.
Voter education
• Established methods of raising awareness of the presence and opening periods of VRCs should be expanded in future to include the use of social media, mobiles and the internet as well as the more commonly employed posters and other printed materials. Radio, TV and newspapers are restricted to the main cities, while cars mounted with speakers also have a limited range.
• While NGOs were involved in voter education, it was not clear how systematic this engagement was in terms of coverage. Also, it was not feasible in all regions. In Sool, for example, political and security concerns resulted in limited outreach early in the registration process, although the active engagement of traditional leaders in voter education later on did ensure high levels of participation in the end. Customary elders could be engaged in a more systematic way in the future. A central coordination mechanism for civil society groups and the NEC could also be established to agree on regional coverage, though that would require funding arrangements for such activities to be in place sufficiently ahead of time.
Voter verification
• Until reliable demographic records are available in Somaliland, it will remain difficult to
implement effective safeguards against the registration of underage individuals. However, there is sufficient evidence that this problem may be significant to warrant greater emphasis in training of NEC staff, and especially those managing VRCs or polling stations, in methods of age verification, and the importance of vigilance in this area.
Inclusion of minority and non-dominant clans
• Caaqils in VRCs were inevitably from the dominant clan in the area. While this generally works well for those from major lineages, areas in which inter-clan relationships are tense sometimes proved more difficult. Sanaag was particularly notable for instances in which registrants complained that caaqils were inadequately addressing the verification needs of some registrants.
While the sensitivities involved in such cases are significant, it is important that verification
procedures are generally accepted as fair, so care should be taken in areas affected to provide verification capacity that caters for all groups.
• This recommendation applies also to minority groups, although in that case, the solution must be national in scope rather than focusing on specific areas. Priority should be given to ensuring that a caaqil from minority clans is available to verify the eligibility of minority registrants when necessary.
Voter card distribution
• There is some evidence from our observation of individuals collecting registration certificates with the intention, in some cases, of ‘selling’ them on to political parties prior to card collection. We suggest that stringent conditions be put in place to ensure that cards are not distributed in bulk to individuals holding multiple certificates.
Voter register maintenance
• The announcement that the election scheduled for March 2017 would be delayed to October 2017 raises the issue of register maintenance. We are unaware of specific plans for procedures to remove registrants who die prior to the next election, or to add those who become eligible to vote in the same period. We are also unaware of procedures to deal with registrants who misplace voter cards. For the register to be considered legitimate at the time of the next election, it is imperative that such procedures be determined in advance, that they are able to accommodate the needs of eligible voters, and that they are publicised in sufficient time to permit full and methodical update prior to the election.
• While we acknowledge that the Somaliland voter register should be considered a ‘periodic’ register that is valid only for a single electoral event, we note that a second combined parliamentary and local council election is scheduled for 2019.
We therefore urge the government and the NEC to give consideration to maintenance of the voter register that is generated from this registration process in a manner that enables its use in future elections.
Download the whole report here: progressio_voter_registration_process_in_somaliland_final_170317
