Could there be a historic run-off in Nigeria?
Sign up here to receive our regular Election Note!
Published 20 January 2023
Can Labour’s Peter Obi break up the APC and PDP’s grip on power? A data-led analysis of the race for Aso Rock.
Key takeaways
Obi’s candidacy presents a challenge to Nigerian two-party politics. After decades of rule by the All Progressives Congress (APC) and the People’s Democratic Party (PDP), Nigeria’s megaparties, Obi’s candidacy for Labour signals a volatile three-horse race
Polls are likely to overestimate Obi’s lead. Given the challenges of Nigerian polling, particularly with regard to voter intimidation and elite networks, the current polling average is likely to be biased in Obi’s favour
It is likely that we will see Nigeria’s first ever run-off in a presidential election. No candidate has a clear path to 50%, suggesting that a second round of voting will be necessary to decide the race
State of play: A three-horse race?
On the 25th of February 2023, Nigerians will head to the polls to elect a new president, vice-president, Senate, and House of Representatives. The stakes are high. The incoming president will have to deal with rising levels of political violence, widespread corruption, and surging inflation. Moreover, as Africa’s most populous, and arguably most influential, country, the winner will be able to shape politics not just domestically but across the continent. This note outlines the current state of play of the Nigerian presidential election and presents an early view of what polls can tell us at the moment.
Ever since the establishment of the Fourth Nigerian Republic in 1999, Nigerian politics has effectively had a two-party system. However, this electoral cycle could be different. At the time of writing, it appears that three candidates have a realistic chance of winning the presidency.
Tinubu (APC): The incumbent centre-left APC has put forward Bola Tinubu as their candidate, since current Nigerian president Buhari (also APC) has reached his 8-year term limit. Tinubu is the former governor of Lagos, the most populous of Nigeria’s 36 states, and (a hugely important facet of Nigerian politics) he is a southern Muslim. His running mate is Kashim Shettima, former governor of Borno state and a northern Muslim.
Abubakar (PDP): The second major camp is the PDP, with their candidate Atiku Abubakar. With the exception of the period between 2015 and 2023, the PDP have continuously held the presidency since the establishment of the Fourth Nigerian Republic in 1999. A veteran politician, Abubakar has previously ran for president 5 times (all unsuccessful). In 2019, he lost to Buhari (APC) by just over 14 percentage points. He is a northern Hausa-Fulani Muslim, and is likely to draw strong support from the vote-rich northern regions of the country.
Obi (Labour): The Labour Party has never been a serious force in Nigerian politics, but their support and membership has exploded after PDP-defector Peter Obi joined their ranks last year. Obi, former governor of southern state Anambra, previously served as Abubakar’s running mate in 2019. His brand is one of economic renewal and anti-corruption politics, appealing to young people in urban centres in particular. Although he lacks the established party organisation structures of the APC and PDP, early polling suggests that Obi is a candidate to be reckoned with (see next section).
Before turning to polling data: how exactly does one become president of Nigeria? The President of Nigeria is elected using a modified two-round system. In the first open round, a candidate is elected if they receive over 50% of the overall vote and over 25% of the vote in at least 24 of the 36 Nigerian states. If no candidate meets these conditions, the top two candidates progress to a head-to-head run-off. Given Nigeria’s hitherto two-party system, a second round has never been necessary. Nevertheless, with Obi’s attempt to establish a third pole of Nigerian politics, it is a real possibility in 2023.
Why the polls overestimate Obi’s lead?
Table 1: Overview of public polling for the 2023 Nigerian Presidential Election
Although the market for Nigerian pollsters is growing, political polling is still seen as very difficult to get right. As a result, we have only seen three pollsters publish vote intention estimates for this race so far. In September, a foreign pollster, Premise Data, estimated that Obi would crush his opponents and collect a jaw-dropping 72% of the vote (see Table 1). However, just months later, SBM and experienced Nigerian pollster NIO Polls put the three significantly closer together. NIO Polls put Obi just 10 percentage points ahead of Tinubu and emphasised that the number of undecided voters was large enough to swing the race in any direction.
What explains this wide range of forecasts? For the Premise poll, conducted for Bloomberg, the diagnosis is quite straightforward: a non-representative sample. The 3,000 respondents were exclusively polled via smartphones. Given the distribution of the levels of smartphone usage across Nigeria, sampling from this group only is likely to skew heavily towards urban dwellers and young people. These demographics are assumed to be most closely aligned with Obi’s political brand. This makes the high vote intention figure for Obi both unsurprising and non-representative.
The two remaining polls (SBM and NIO) differ on one fundamental aspect. Whereas the NIO poll reports the percentage of respondents stating that they “don’t know” or “prefer not to say” who they will vote for, the SBM poll does not. This seems minor, but it is a crucial difference when it comes to inferring what these polls tell us. In the UK, political psychology generally assumes that people who say they do not know who they will vote for in an upcoming election either a) genuinely have not yet made up their mind or b) are using the “don’t know button” as a short-cut to express more complex emotions of voter apathy and political alienation. If we can assume that the votes of the former group will be distributed proportionally across the candidates and that members of the latter group will not vote, SBM’s decision not to report ‘don’t knows’ is defensible. It would not affect their voting estimates.
However, in the Nigerian context, “don’t knows” and “prefer not to says” may derive from another source: voter intimidation. Unfortunately, election-related violence is quite common in Nigeria (for example, see ACLED’s ‘election violence tracker’). In many parts, someone’s vote choice can be a highly sensitive piece of information, and one that many survey respondents would be very hesitant to share with a researcher they do not know or trust. Vote buying, another common facet of Nigerian electoral politics, may have similar effects, leading respondents not to disclose their vote choice as their vote is for sale to the highest bidder. Both voter intimidation and vote buying can therefore artificially inflate the proportion of respondents declining to state a candidate preference in vote intention questions.
Why does this matter? This is because knowing the (weighted) proportions of respondents declining to state a preference can help us infer something about the direction and magnitude of bias in a poll. In the case of the NIO poll, it is likely (but not certain) that Obi’s lead is overestimated. Specifically, it is likely that vote buying and voter intimidation is more common in more insecure rural areas and northern Nigeria. Given the fact that the APC and PDP are substantially more established in these parts than Labour, voters that have been exposed to these practices can be expected to skew towards these parties. Assuming vote buying and intimidation can lead respondents to decline to answer vote intention questions in surveys, we can thus infer that a part of these “decliners” will skew towards the APC and PDP.
Add to this the fact that surveys in Nigeria have tended to overrepresent the urban (and more pro-Obi) part of the population, and the direction of the polling bias is clear. The likely reality is that, if Obi is leading the pack, this lead is significantly smaller than the polls suggest.
Paths to victory
Map 1: Nigeria coloured by geopolitical region
If the race is tighter than the polls suggest, what can we say about each candidate’s winning chances? Perhaps the best way to assess this is to outline their respective paths to victory. In the 2019 elections, Abubakar’s PDP won an average of ~80% of the votes in its stronghold seats in the Christian South-South and South-East (see Map 1). However, Tinubu’s APC (then led by Muhammadu Buhari) won handily by dominating in the Muslim north of the country, specifically in the vote-rich states of Kano, Katsina and Kaduna. Add that to the fact that the APC secured 55% of the vote in Nigeria’s swing regions - North-Central and South-West - and their path to victory becomes clear. In fact, this has been the story of Nigerian electoral races for its two decades of two-party competition. Each party tries to maximise its vote share in their strongholds (North for APC, South-East for PDP) and secure at least 50% of the vote in the North-Central bellwether seats.
Can Tinubu and the APC repeat their 2019 map to victory? The short answer: it’s unlikely. Their strategy consists of two components: expanding their vote share in the Muslim north (North West + North East) and flipping the South West in favour of the APC (see Map 2). Regarding the former, Tinubu is running a so-called “Muslim-Muslim” ticket, where both he and his pick for vice-president Kashim Shettima are Muslim. This is a potential vote-winning factor in the north. Shettima’s networks in the North East - as former governor of Borno state - should also help in this regard. Tinubu’s strategy in the South West is built largely around his own political profile. As governor of Lagos - the largest state in the region, he is hoping to increase the 53% of votes that the APC received in 2019.
Although the APC will definitely pick a good number of votes, both components of this strategy have their drawbacks. Firstly, Tinubu’s brand as a southern Muslim may see him struggle to draw votes in the north, especially given that his rival Abubakar is a northern Muslim. State polling by SBM suggests that Abubakar may flip the northern states of Bauchi, Gombe and Sokoto and make inroads in the previously APC-dominated North West and North East zones. Secondly, Tinubu’s Muslim-Muslim ticket will be an obstacle to picking up votes in the Christian south and has already caused significant backlash in Christian communities. Even if Tinubu has decided he does not need votes from the South East and South South, it may also end up hurting his vote share in the South West. With these challenges in mind, it is difficult to see how Tinubu will reach 50% of votes in the first round.
Map 2: Nigerian states coloured by Buhari’s (APC) vote share in the 2019 presidential elections
The PDP and Abubakar’s path to 50% is equally, if not more, difficult. They have to hold onto their fortress seats in the south while also syphoning votes off the APC in the north. Regarding the latter, the PDP are likely to make inroads in the north given the relative strength of Abubakar’s brand in the area (as mentioned above). However, the PDP faces pressure from Obi’s Labour Party in the South East and South South. Based on state projections, Obi is likely to flip states like Anambra (where Obi was governor), Ebonyi, Enugu, and Imo. To emphasise the scale of those swings: these are states that Abubakar and the PDP won by margins between 40 and 80 percentage points. The fact that Labour is competitive here highlights Obi’s strong brand in the region, but also painfully exposes the drawbacks of the PDP’s focus on northern Muslims. Without their southern base, it is impossible to see how the PDP will reach 50% in the first round.
Obi has severely disrupted Nigerian status quo politics. His brand, which is strongest in and around his southeastern constituency, simultaneously appeals to urban youths who are tired of politics-as-usual and Christians frustrated by the Muslim focus of the election. However, even if Obi manages to break PDP hegemony in the South East and South South, he is very unlikely to break 50% in the first round of voting. As results come in, key states to watch will be Lagos as well as Niger, Kaduna, Plateau, and Nassarawa. If Obi can secure at least a third of the vote in these religiously and ethnically mixed areas of the country, he has a path towards Aso Rock in the second round.
In brief, it is unlikely that any of the three major candidates will break 50% and secure a victory in the first round of voting. Although polls give Obi the best shot at reaching the run-off, this remains to be seen given the polling bias in his direction. In our view, Tinubu remains the most likely candidate to make it to the second round and face either Obi or Abubakar. Obi and Abubakar are locked in what seems to be a zero-sum game for voters in the South South and South East, making it unlikely that both of them will make it to the second round.
Without any second round polling, the eventual winner of the run-off is impossible to predict. However, should Nigeria indeed witness its first ever second round of a presidential election, we will make sure to keep you informed with another Stack Election Note.