The Centre for Democracy and Development (CDD) West Africa does not take institutional positions on public policy issues; the views represented herein are those of the author(s) and do not necessarily reflect the views of the CDD or West African Insights.
This paper may not be reproduced in whole or in part, in any form, in any registered online or print publication without express permission from the CDD.
Nigeria’s next elected president will lead the country into celebrating 63 years as an independent nation on October 1, 2023. In that time, Nigerians have seen attempts at democracy interrupted three times, but democracy has been the system of government for the last 24 years, with the 25 February polls the seventh since 1999. In that time, Nigeria has witnessed two presidents stand down after serving the constitutional limit of two terms (2007 and 2023), as well as an electoral defeat for an incumbent (2015).
For the presidential and national assembly elections on 25 February 2023, CDD deployed over 4,900 trained and accredited observers to systematically monitor the polls across all 36 states of the federation plus the federal capital territory (FCT). These insights, supported by analysis by data clerks, the work of fact-checkers and social media monitors, and the expertise and input of leading analysts provide the basis for this report on the elections.
It identifies and analyses four key themes – election day operational challenges and the use of technology, violence and voter suppression, voter behaviour and the influence of money and online falsehoods – that shaped the process, and provides some initial reflections on emerging trends from the voting so far as well as recommendations for improving future processes in the short, medium and long time.
Election day operational challenges
The 2023 elections were held as scheduled on February 25, 2023. A feat not achieved in the past three elections, which have all been delayed. This was in spite of prevailing insecurity across the country, perpetual fuel scarcity and the November 2022 decision of the Central Bank of Nigeria’s (CBN) to redesign the naira, which created currency shortages and threatened to derail operational plans. Although the Independent National Electoral Commission (INEC) confirmed that it had received the necessary resources shortly before the elections, deployment plans appear to have been affected by the uncertainty it caused in some instances.
INEC officials were present at polling stations before they opened at 0830 in just 36.7 per cent of cases according to our observers with 32.5 per cent arriving an hour or more after voting was scheduled to have commenced. This was a particularly pronounced challenge in the southeast and is likely linked to concerns for personal safety given that two-thirds of the attacks recorded against INEC since the start of 2021 have been in this geopolitical zone. In some cases, INEC officials arrived shortly before 1430 – the time polls were officially due to close.
CDD EAC preliminary press conference
When voting did begin, some challenges with the use of the Bimodal Voter Accreditation System (BVAS) were noted. Despite training delivered to all ad-hoc INEC staff on the use of the new technology tools, recruitment shortcomings and ongoing capacity gaps affected its effective deployment. This was most acute in the Northwest where 23 per cent of observers noted incidents of voter accreditation being completed in polling units, where BVAS was used, even if the individual was not captured using the BVAS reader. T
This was in contravention of the legal framework, which requires that all voters must be registered by BVAS in order to vote. Section 64 (6) of the Electoral Act also designates a primacy to the results from BVAS in the event of a disputed result. Although less pronounced elsewhere in the country 8.7 per cent of observers witnessed this taking place in south-south, 5.9 per cent in southwest, 9.6 per cent in the northeast, 4.2 per cent in the southeast and 7.7 per cent in the north-central zone.
Technical issues were also noted by our observers, with 23.1 per cent having witnessed issues of BVAS malfunction. This was particularly acute in the northeast, with 42.7 per cent of observers recording a malfunction, but was also high in the northwest (28.7 per cent). Positively, INEC’s contingency planning was broadly effective, with more than two-third of the technical hitches resolved within one hour, and where the technology was working accreditation happened at a good pace.
However, in some instances the malfunctioning of the reader created long delays for all those queuing to vote. For example, in Ile-Bale polling unit (002) in Ajase-Ipo ward, Irepodun local government area (LGA), Kwara State, many prospective voters waited for more than three hours due to malfunctioning BVAS machines. In less than 1 per cent of polling units, no BVAS reader was seen during opening by our observers.
A further consequence of the late opening of polls was that voting continued into the evening in states such as Plateau, Lagos and Imo, following INECs election day confirmation that those in the queue to vote by the original 14:30 cut off time would be allowed to cast their ballot. However, keeping track of those in the queue for such a long period beyond the deadline is difficult.
Beyond that, there have been anecdotal reports of large numbers of people arriving to vote, in urban areas of the FCT for example, but leaving before doing so after queuing for several hours, in areas often outside and open to the elements. This was not only a result of opening delays and technical challenges but also a consequence of problematic polling station allocations, which led to scenarios where some polling units had just a handful of voters whilst others had more than 2,000.
A final issue relates to the INEC Result Viewing Portal (IReV), which was designed to provide real-time transmission of election results (Form EC 8A) from polling units to the central collation centre in an attempt to enhance the transparency of the electoral process and reduce the incidence of vote rigging and election manipulation. However as of 21:00 on election day, there were no results uploaded to the platform for the presidential results.
By 11:00 on Monday February 27 just 53,154 polling unit results out of a total of 176,846 were publicly available on the platform despite the expectation that results should be uploaded to the platform as soon as they are declared at the polling unit level.
INEC stated in a February 26 press release that the challenges were a result of “technical hitches related to the scaling up of the IReV platform” but the delay in making this statement provided room for unproven rumours to flourish that point to the incomplete polling unit results on the platform as evidence of vote rigging.
The fact that some senatorial and Federal House of Representatives elections were uploaded also fuelled rumours of vote manipulation. Despite optimism from both INEC and citizens that technology could be a game changer in Nigeria’s elections, the way the technology was deployed during results transmission in particular has arguably weakened the public perception of transparency and accountability, rather than strengthened it.
Provision in the 2022 Electoral Act, laid out in Section 54 (2), provide for improved access for persons with disabilities (PWD). Whilst some polling unit locations made them inaccessible to those with a physical disability, we did not record significant incidents of those with disabilities being unable to vote. Just under 20 per cent of polling units had specialised equipment for PWDs, with this allocation done by INEC through a targeted approach based on an assessment of the voter register.
Despite well-documented instances of insecurity in all six geopolitical zones, and notwithstanding the fuel and currency scarcity that threatened to derail the electoral process, Nigerians, particularly the youth, had shown a strong interest in participating in this election.
Of the 9.46 million voters added to the register by the Independent National Electoral Commission (INEC) before the 2023 election, over 70 per cent were youth (aged 18-34); meaning that they comprised 39.7 per cent of the total electorate of 93.4 million.
But this election looks likely to suggest that ethnic identity remains an important factor shaping voter preference, with youth not a monolithic block. Results so far declared show that Peter Obi’s campaign has failed to penetrate in northern states – he won 0.21 per cent of the vote in Jigawa for example – although his narrow victory in the presidential vote in Lagos, defeating the state’s former governor and ruling party candidate, Bola Tinubu, slightly bucks that trend.
However, turnout as a percentage of total registered voters looks likely to be lower in 2023 than the 35 per cent figure recorded in 2019 based on the results announced so far, continuing a pattern of declining electoral participation in Nigeria that began in 2003. This is despite INEC confirming the national average of PVC card collection at 93.3 per cent ahead of the elections.
Whether this reflects the reality or is partially skewed by the fact that voter register was not updated between 2019 and 2023 to remove those who had died or moved abroad or is impacted by the fact that 6.2 million permanent voter cards were not collected before the polls according to INEC figures, the fact that a significant percentage of Nigerians fail to engage in elections is a concern and perhaps points to growing disillusionment with their ability to shape a more democratic society.
Afro barometer data published shortly before the election found that 9 in every ten respondents did not believe that the country was headed in the right direction. If elections are not seen as a way of addressing that, more focus needs to be put on other tools and platforms for civic engagement beyond elections, to reduce the risk of sustained democratic decline.
Although vote buying appears to have reduced as compared with recent off-cycle elections, 11 per cent of observers reported some form of vote buying taking place in or around their polling units. This was highest in the northwest, 18 per cent, with the second highest number of incidents, 12 per cent, reported in the southwest.
Just 5 per cent of observers reported voting buying in the northeast. Vote buying is used more by political parties in their strongholds, where they want to ensure that those inclined to vote for them are encouraged to do so. For the most part vote buying appears to be the preserve of the two major parties in the last two elections – the All Progressives Congress and People’s Democratic Party – with the Labour Party’s limited structures and resources meaning that it was less able to compete on this front.
Naira scarcity was one of the defining features of the last month of this election campaign. The macro monetary policies were principally meant to preserve the integrity of the naira, reduce the significant amount of cash in circulation outside the banking system and its use for criminal activities, and strengthen the economy by controlling spiralling inflation and corruption.
But it created palpable frustration and anxiety among citizens, with the inability to access cash having grounded many small businesses, further increased daily hardships and increased the difficulty of travelling home to cast a ballot. It may even have made voters more susceptible to selling their votes.
CDD issued this report on the February 25 Presidential and National Assembly elections.
The Centre for Democracy and Development (CDD) was established in the United Kingdom in 1997 as an independent, not-for-profit, research training, advocacy and capacity-building organisation. The purpose was to mobilise global opinion and resources for democratic development and provide an independent space to reflect critically on the challenges posed to the democratisation and development processes in West Africa.
This election also saw an increased use of basic foodstuffs, household goods and materials being exchanged for votes, with observers in states such as Kwara, Imo and Bayelsa all reporting numerous incidents. In Lagos, there was even evidence of politicians willing to provide bank transfers to voters under the guise of “business assistance schemes”, ostensibly a creative way to buy voters given the limitations on the availability of cash. But despite the shortages politicians were still able to secure cash even if the crime agencies caught some individuals before they could distribute it.
An individual in Lagos was detained on the eve of the polls by the Economic and Financial Crimes Commission (EFCC) with N32.4 million new naira notes that it believed were to be used for vote buying. Elsewhere in the run up to the polls, in Rivers State, a PDP aspirant for the House of Representatives was caught with almost US$500,000 in cash, which investigating agencies claimed he intended to use to buy votes.
Violence and voter intimidation
Intimidation, in many cases linked to identity or political party affiliation, was recorded across the country and was reported by 3.4 per cent of observers. It was most commonplace in the northwest, south-south and southeast. One instance of voter intimidation reported by an observer in Anambra State detailed how party agents were standing aside the voting cubicle directing voters as to how they should cast their ballot in full view of INEC ad-hoc staff.
For the most part conducted by political parties thugs or loosely affiliated individuals, some of the intimidation designed to suppress votes centred on ethnic identity. In Lagos, threats were made against Igbos by prominent political thugs like MC Oluomo. BVAS readers were also targeted in some parts of Imo and Anambra states. Without a reader, voting is unable to continue and although replacements were found, this tactic for voter suppression could become more widespread in the future. In Bayelsa State polling in 141 units were postponed after they were disrupted by violence. Although they were to go ahead the following day, safety concerns could result in reduced turnout.
Despite the prevailing insecurity in the country this was not a significant factor that contributed to incidents of violence during the election processes, with incidents mainly the preserve of political thugs. Overall, 6 per cent of our observers reported incidents of violence at the polling units they observed, with the northwest and south-south reporting the most incidents, with the southwest and north-central at the other end of the spectrum.
Although not reported by our observers, a review of media coverage of election day suggests that there were ten election-related deaths, two each in Abia, Rivers, Kano, and Ondo states, one in Kogi and one in Delta State.
Even though vandalism and destruction was only observed in a small percentage of polling units where observers were stationed – 1.2 per cent – this still translates to as many as 2,000 polling units, given the huge number of points where people can cast their ballot in Nigeria. Whilst it should not overshadow the fact that the polls generally were free from violent incidents it still impacted on a sizeable number of voters seeking to exercise their democratic right. For example in Lagos state, there were documented gunshots and attacks on polling units to deter voters. CDD observers reported gunshots in Oshodi at 14:00 while monitoring nearby polling units.
Meanwhile, media reports also highlighted the snatching of ballot boxes in Surulere by armed thugs in the state, and thugs also attacked a collation centre in Kano state.
In the Nigeria information landscape, while some renewed level of sophistication was observed in the disinformation campaigns during this election, many of the strategies used were similar to 2019. A rise of synthetic and manipulated media images were doctored to push certain narratives against candidates and videos were presented out of context to misinform the voting public. A video showing Peter Obi campaigning for Atiku Abubakar in 2019 reemerged hours before the 2023 vote; adapted from TikTok it was shared widely across the platform and WhatsApp.
Efforts to undermine INEC’s credibility were recorded throughout the campaign and even as polls were set to open. An audio file that started as a WhatsApp voice note and evolved into a video that was shared across platforms, claimed the PDP candidate was working with the INEC to rig elections. Disinformation peddlers stitched the voices of Atiku and two other political actors together to form a conversation about rigging the election.
Political actors and influencers started sharing unverified election results immediately after polls closed in some areas and took advantage of the small trickle of election results. There were also many reports around electoral malpractice and the lack of crisis communications from INEC allowed such accusations to go unchecked. Rumours of rigging are being enhanced in some instances by pre-election polling that presents a different picture to the reality now emerging.
Despite fact-checks of these proving the falsehoods many persons will still choose to believe the disinformation, as long as it favours their political leaning. The politicisation of fact checking is furthering this challenge. Furthermore, distributing fact-checks via media houses is a challenge as several expressed concern that they cannot publish or share some of them because it would seem like they are working for certain candidates.
Social media companies were largely non-responsive to election day malformation.
Meta took down one post about misleading images about INEC offices in northern Nigeria, and Twitter suspended an imitation account claiming to be the wife of President Muhammadu Buhari on the eve of the election for sharing misinformation about naira redesign policy, but no activity was taken in response to more than 20 pieces of false content flagged by CDD fact checkers. Almost 72 hours after the fake audio on rigging was debunked by CDD and other fact checking partners, much of the video is still available on TikTok, Twitter and Meta platforms.
Expectations of an increased voter turnout appear to have been unfounded. This may be partially skewed by the fact that the voter register has not been regularly updated to remove those who had died or moved abroad and the fact that 7 per cent of permanent voter cards, according to released INEC figures, remained uncollected. But could also be linked to INEC performance issues and wider security concerns highlighted in the sections above. The fact that a significant percentage of Nigerians fail to engage in elections is a concern and perhaps points to growing disillusionment with their ability to shape a more democratic society.
*Initial signs point to a growing realignment of the political landscape, independent of the outcome of the presidential results. National legislature candidates of the “third force parties” such as the All Progressives Grand Alliance (APGA), Labour Party (LP), New Nigeria Peoples Party (NNPP) and Social Democratic Party (SDP) appear to have won their seats. This has seen several prominent politicians lose their election bids, including the outgoing governors of Abia, Benue, Cross River, Enugu, Plateau and Taraba – who lost their senate campaigns. The House and Senate Minority Leaders also failed to secure re-election in Delta state and the FCT respectively. There were also unexpected instances where senators and House of Representatives candidates outperformed their presidential candidate such as in Ebonyi State, where two APC senatorial candidates are performing significantly better than their presidential candidate.
*Early voting patterns suggest that ethnic identity may have been a key determinant in voters’ choices in this election. The PDP, prior to this election, had always secured the majority of votes in states of the southeast and south-south but strong support for the Labour Party presidential aspirant from early results suggest this will not be the case in 2023. Likewise, Yobe, which had never voted for a PDP presidential candidate, did so in 2023, with Atiku the only northerner among the three leading candidates. However, the result from Lagos, where Obi was able to edge Tinubu offers some optimism for a generational change independent of the ethnic cleavages that have historically dominated Nigerian politics.
*Nigeria will see another wave of electoral litigation in its courts that will likely reverse some electoral outcomes and perhaps even impact on governance in the short-term, with challenges likely to be determined by the weighting that the judiciary gives to procedural failures.
CDD issued this report on the February 25 Presidential and National Assembly elections.
The Centre for Democracy and Development (CDD) was established in the United Kingdom in 1997 as an independent, not-for-profit, research training, advocacy and capacity building organisation. The purpose was to mobilise global opinion and resources for democratic development and provide an independent space to reflect critically on the challenges posed to the democratisation and development processes in West Africa.
This report has focused on analysing the conduct and actions that shaped election day and identifying key themes that have emerged and how they tie to pre-election factors that shaped the landscape on which polls were held. It notes that new electoral technology appears to have struggled to meet expectations and improve transparency in the results collation process in particular, whilst the digital landscape continues to amplify existing tensions, with narratives that play up identities and party affiliations and question INECs performance and credibility. INECs election day performance was not seamless, with the late start to voting problematic given that the period polls are open for it relatively short, even if it was extended in response to this. Finally, behaviour on polling day – low turnout and a tendency to exchange cash for votes – remained a feature of Nigerian elections, with violence, predominantly carried out by actors affiliated to political actors also an enduring feature.
To support improvements in Nigeria’s electoral process the following recommendations are proposed:
*INEC continues to face a perception versus reality challenge. To ensure voters better understand the reasons behind delays or technical hitches, for example, more proactive communication is needed. This can allow INEC to set the agenda and explain the reasoning for operational issues, rather than allowing others to set a more politicised narrative, which becomes difficult to counter. INEC should revise its communication policy and train staff to support proactive communication with voters. Doing so will help improve transparency and accountability to citizens and reduce the risk of social media rumours setting the narrative.
*INEC also needs to ensure a dedicated and prompt page for communicating updated decisions or positions around election-day. This includes, but is not limited to, reports around postponed elections, reassigned polling units and logistical delays.
*The presence of IReV has brought some accountability and helped democratise the result reporting structure for INEC. However, there is still a challenge in verifying results – especially when there is no dedicated page run by the commission to confirm election declarations. Running a dedicated site for verified results can help citizens distil between projections, reported results and misinformation.
*Investigations into voting day violence, and those involved in voter suppression and vote buying should be thorough and result in convictions where proven.
*A more robust stress testing of IReV is needed to avoid its malfunction undermining the forthcoming governorship process, more than it strengthens transparency around it.
*Effective collaboration between INEC and the judiciary in properly adjudicating complaints concerning the election process will be crucial to avoid significant instances of candidates being ‘elected’ by the courts, rather than by the electorate.
*Political parties must take more seriously their commitments to peace laid out in the National Peace Accord by reigning in thugs acting at their direction.
*Social media companies need to expand their ability to respond promptly to falsehoods circulating on their platform, particularly on and immediately after election day. There should be improved collaboration with recognised fact-checking platforms to allow for quick response and action.
In the future
*Updates to the voter register should not just be limited to the addition of new voters but the removal of the deceased and the no longer eligible before each election cycle so as to better reflect the levels of voter turnout.
*A review and redistribution of voter allocations to polling units is needed to reduce scenarios in which some have only a handful of voters and others have thousands.
*To further build the capacity of INEC ad-hoc staff in the conduct of polling day activities, the Commission should seek support to undertaken more robust and timely training, particularly when it comes to the use of technology, to ensure knowledge is embedded by the day of the election.
*Ahead of future elections, interpretations of the constitution as they relate to the requirements for a presidential winner to be declared or prospective run-off candidates to be selected should be clarified to reduce the risk of post-election contestation.
*Security agencies and government must also provide stronger support to INEC, which has to undertake significant logistical operations in a difficult environment, by working to address prevailing insecurity and ensuring that INEC has the financial resources at its disposal in sufficient time.
*INEC needs to ensure that the different stakeholder groups that if often draws ad-hoc personnel from, such as the Nigerian Youth Service Corp and Academic Staff Union of Universities (ASUU), have robust training schemes to mirror its efforts. If need be, INEC needs to make a decision on either prioritising quantity (large PUs) or quality (potentially having single voting sites per LGA with enough trained staff, albeit inevitably longer voting times).
*An independent post-election review by a group of civil society organisations that observed the election, alongside representatives from legal bodies, should be constituted to provide a report and recommendations on a way forward for continued improvement in the management of elections and the operational deployment required and the role for INEC and other key stakeholders.