Summary

• Nigeria's official population (~220-230 million) may be significantly inflated and could be closer to 170 million
• This overcount is likely driven by political and financial incentives for states
• I'm unsure of the implications if this is accurate
  • If states have uniformly inflated populations than the distribution of resources could still be divided evenly
  • Nigeria would still be the biggest country in Africa and companies/governments/NGOs would have similar cost benefit analysis for working and investing there
• This is a very shallow investigation

Why did I bother looking into this?

The below text sparked an investigation into Nigeria's population claims. It was slightly hidden in the 4th section in one of Yaw’s excellent Substack posts.

Yaw went onto explain his reasoning for thinking the population was much lower than current estimates.

Nigeria is a large country with no deep shared history among the different tribes. Due to the historical realities of different tribes hating, enslaving, and warring each other, it is a low trust society. Most Nigerians don’t trust other Nigerians and corruption is rampant.

In a low trust society, in order to ensure loyalty, you must buy off elites - political patronage. Nigeria has one of the lowest tax revenues as a proportion of its economic output among nations, even by Africa standards. Instead 65% of Nigeria’s government revenue is oil export proceeds. Then oil is distributed to provinces to “buy” them off. The oil revenue is distributed in proportion to the population of each province. This incentivizes a system where more money is concentrated in more populace regions. This incentivizes provinces to over report fertility rates, undercount child mortality rates, undercut emigration rates, over report population sizes, and not check population estimates accurately. Nigeria hasn’t done a census since 2006. Nigeria was supposed to do a census in 2023, but it was postponed. Even as of June 2024, the census has not been done yet.

To be honest, I wouldn’t be surprised if Nigeria’s population was closer to 170M.

My instant reaction was that this sounds potentially exaggerated or mistaken, surely you can't lose 50 million people without realising. And wouldn't there be incentives for the UN, other governments, companies to know how many people live in Nigeria. So I set about trying to find out if I'm wrong, and if the hours I spent learning the populations of different countries were even more pointless.

Other Sources

First I went for the most official stats. 

CIA Factbook (2024) - 236,747,130

UN (2024) - 230,271,000

National Population Commission (for Nigeria for 2022) - 216,783,381

The National Population Commission data comes from a 423 page report that is partially based of the 2006 census with various metrics assumed to remain constant.

I then  looked to see if other people had reported on this and saw a few articles that suggested there was a difference in actual and census populations.

From the Population Reference Bureau in 2006.

...in Nigeria, last month’s national census was met with protests, boycotts, charges of fraud, and at least 15 deaths. Thousands of other enumerators walked off the job because they hadn’t been paid, and many people in large swaths of the country say they still haven’t been counted.

Passion and conflict are nothing new for Nigerian censuses. Their results decide the division of federal money and the balance of political power in a nation split almost evenly between a largely Muslim population in the North (which has traditionally controlled the government) and the coastal, more-urbanized, and largely Christian South, which regularly accuses northerners of rigging the census for political gain...[The North’s political strength arose initially from the results of the 1952-3 Census, which showed 54 percent of the country’s population resided in that area]

And again the Population Reference Bureau in 2007.

Bola Tinubu, the governor of Lagos state, called for a recount and said the census figures were false. Nigeria’s president Olusegun Obasanjo called those who dispute the results “confusionists,” adding that when they saw the census didn’t break the country, they sought to sow confusion...

Explaining his objection to the census results, Tinubu claimed that a parallel census conducted by Lagos in collaboration with the National Population Commission put the state’s population at more than 17.5 million, not 9.0 million, as the 2006 census suggested...

In a report on the 2006 census process, the chairman observes that even before the census was conducted, highly placed individuals and organizations in several states had already determined to the decimal point the population of a particular area or region. “Almost all these wild guesstimates apparently were based on their perceptions mostly out of ignorance or possibly ulterior motives to suit their agenda,” he notes.

And on the incentives behind the inflated numbers.

Census numbers in Nigeria guide political redistricting for each of the country’s 36 states, the distribution of federal funds, and even civil service hiring. However, historically, southerners tend to think resources are not distributed equitably. For instance, more than 90 percent of the country’s revenue comes from oil in the south, but the north controls about 55 percent of Nigeria’s revenue

I'd like to highlight a section in the article linked above, which would suggest that the census issues could start from the households themselves.

On the first page of the census form, there was enough room to list up to nine household members. Oddly enough, data collected from some states in the South showed all households reporting exactly nine people.

From The Economist in 2015.

Its population, too, at an estimated 183 million, is the largest of any African country...There is just one snag: they are almost certainly wrong. Many of the figures about the country that are making the rounds are patently absurd, and few more so than the population statistics, for an obvious reason: allocations of revenue from the central government and voting power in the capital depend on population estimates, so every region has an incentive to bump up its own count.

...The first post-independence census in 1962 was already shamelessly rigged. An initial count suggested massive growth in eastern and western districts, which claimed that their population had increased by an average of 70% over the previous decade, compared with a 30% increase in the north. That would have shifted power from the northern elites who controlled the country, so they quickly scrapped the count and started again. This time, miraculously, the north’s population was found to have increased by 84% (an extra 9m people), just enough to ensure it had slightly more than half Nigeria’s population Almost every census since then has been disputed. One in 1991 was ditched when it seemed to show that the country’s total population was about 30% smaller than expected.

The latest one, conducted in 2006, put the total at 140m, a number that provides the basis for current estimates and forecasts. But some academics reckon that the population of some northern states was inflated by about a quarter, whereas that of some southern areas was trimmed in response to political pressure on the statisticians. The number of people in Lagos, the main southern commercial centre, was said to be just over 9m. That allowed northern rulers to claim that Kano, the main northern commercial centre, had more people and was thus entitled to more resources. Lagos officials subsequently did their own count and claimed 17m.

Even allowing for all these swings and roundabouts, some researchers, using sophisticated satellite imagery and geographical information systems, reckon that the 2006 census considerably overstated Nigeria’s urban population, mainly in the north but also in some southern cities. That means Nigeria’s current population may be closer to 160m than 180m.

It seems from the above posts, that there are plenty of incentives for each state to bump up population numbers.

• Funding & resource allocation (mainly from oil money)
• Political power
• Civil service positions
• Military positions

Potential Data Sources

I then started to look into other ways to see if the population projections were inaccurate. 

National Identification Numbers

Nigeria has a national ID card run by the National Identity Management Commission, which can be registered from birth. The number of Nigerians with National Identification Numbers (NIN) rose to ~104 million by the end of 2023.

This is a 10.8% increase from the 94 million that was recorded as of the end of December 2022. This signifies that only 10.13 million Nigerians registered for NIN in 2023. A monthly average of enrolments reveals that 844,167 Nigerians got NIN per month in 2023.

This is a far cry from the Federal Government’s target of 2.5 million registrations per month.

I then looked into enrollment per state to see if there were any discrepancies (more enrollments than population or a large difference in registration between states). It looks like the smallest states have the lowest percentage enrolled but that could be plausible for harder to reach and poorer rural areas.

There is a reasonably large gap between Lagos and Kano in percentage registered, but that could be a cultural/economic difference. I'm not sure I can take anything conclusive away from this but it doesn't disprove a smaller population.

Tech usage

I then thought that it may be easier to track usage of online services that a high percentage of the country use. This may undercount younger, older and more rural populations, but may be relatively accurate as a lower bound for 16-65 year olds (although people could have multiple accounts).

It was hard to find accurate^[1] data, none of the large companies, as far as I can tell, release usage data on a country level. 

Sim Cards

Another path was to look for sim card usage, and recently there was a NIN-SIM verification exercise that would incentivise people without a NIN to get one so that they could keep on using their sim.

The recent National Identification Number (NIN)-SIM verification exercise has caused Nigerian mobile network operators to lose 64.3 million subscriptions, as the Nigerian Communications Commission (NCC) concluded the programme on September 14, 2024. 

The verification, aimed at linking SIM cards with verified NINs to improve security and user accountability, saw major subscription losses across the nation’s top telecoms 

• Globacom’s active subscriptions dropping from 62.1 million in March to just 19.1 million
• 9mobile went from 11.6 million subscribers to only 3.6 million
• MTN retained a lead with 78 million active subscriptions despite losing 3.7 million
• Airtel ended the quarter with 53.7 million users, shedding about 9.6 million

The NIN-SIM verification mandate began in December 2020, with multiple extensions and deadlines given before the final September 14 cutoff. The delays led to chaos at telecom outlets, with customers scrambling to verify their NINs in time.

This suggests there are at least 154 million phone subscriptions, but then anecdotally people have multiple subscriptions (back to square one).

I personally own 4 lines. One obtained for personal reasons, held since 2003, two picked as corporate lines in the course of my career, the last embedded in my portable wifi device.

Both my parents have 3 each, several colleagues, friends and contacts have more than two lines.

I was on a call with representatives of a client on Jan 3rd and we took a quick poll on the subject. Two of them (like me) had four lines each and the third person with had two.

For several reasons, connectivity, business, it is very common for Nigerians to own more than one number.

UN Population Estimates and Projections

This is more tangential to current population, but the UN has revised their estimates for how large Nigeria will be over the years. I only picked a few years so can't confirm that it was smooth lines vs lots of ups and downs.

 In the 1990's their current estimates were around 90-100 million. After the 2006 census which said there was 140 million people they increased their historical estimates and future projections. Although, at some point more recently, they have revised down their long term projections for 2100 from ~730 million to ~477 million. It may be a case that their pre-2006 census estimates were more accurate than using those figures.

All of the below are the UN's medium estimate.

The 1990^[2] estimate of the population in 1985 was 92 million.

The 1994^[3] estimate for 1990 was 96 million.

For 2025

• 1990 projection - 281 million
• 2000^[4] projection - 203 million
• 2010^[5] projection -  230 million

For 2050

• 1994 projection - 339 million
• 2000 projection - 279 million
• 2010 projection - 390 million
• 2019^[6] projection - 401 million
• 2024^[7] projection - 374 million

For 2100

• 2010 projection - 73o million
• 2019 projection - 733 million
• 2024 projection - 477 million

Our World in Data has a post on the accuracy of UN projections for global population with most projections differing from the latest estimates by just 1% to 2%. Although if the estimates are wrong, that wouldn't show up using this method. 

I think given the variance in the projections for Nigeria, some, if not all, have to be quite inaccurate.

Global projections have been relatively accurate, but this isn’t true for every region
We should not assume that because global projections have been reasonably accurate, projections for individual countries or regions have been equally so. This also applies to specific age groups.

Over- and under-estimates of particular populations canceled each other out, leading to a total figure that seemed more accurate than it was for individual groups. For example, an overestimate of the population in Asia could be partly canceled out by an underestimate in Europe.

In a paper published in Population Studies, the researcher Nico Keilman, studied the accuracy of UN population projections by looking at the errors of large regions, age groups, and specific estimates of fertility and mortality trends...

Another significant error for some regions – affecting mostly Africa, Asia, and Latin America – was inaccurate population estimates in the first place (in what we would call the “base year” when the time-series starts). Many countries in these regions had poor census records and national statistics on the size and structure of the population. If the population in the start year is wrong, the projected population could be very wrong, even if you make reasonable assumptions about changes in fertility and mortality trends.

Incentives for not caring 

Nigeria and international organisations seem to have weak/negative incentives for having accurate data. One question I had throughout, is that if this undercount is even close to being true, why hasn't anyone spent more time/effort looking into this. Maybe at the census conferences this is the hot topic, but I didn't find much conversation online. Here are some potential reasons why the different groups are not incentivised to investigate.

Within Nigeria

• State governments benefit from inflated numbers through increased federal allocations
• The national government benefits from appearing larger on the global stage and supporting favourable states
• Opposition parties want to challenge numbers, but only in regions where it benefits them politically
• Academics and researchers might face political pressure not to investigate

International Organisations

• The UN works with official government statistics and may be hesitant to challenge member states' data
• Aid organisations might prefer higher numbers to justify larger programs
• International organisations need to maintain working relationships with the government and don't have the capability to run a national census
• The cost and complexity of conducting independent verification will outweigh the benefits
• Development banks and funding agencies use official statistics for consistency

Private Sector

• Companies care more about specific market metrics (purchasing power, consumer behaviour) than total population
• Multinational corporations can adjust their internal projections without publicly challenging official figures
• The difference between a subset of 170 and 220 million potential consumers might not change investment decisions significantly

Implications

I'm more persuaded that Nigeria's population is significantly overcounted. However, I'm still very uncertain about the implications of this, there aren't many areas where it is obvious that actions need to be changed given this update.

International Standing

• Nigeria would remain Africa's largest country by population, ahead of Ethiopia (~120 million)
• Its geopolitical importance would be largely unchanged
• International organisations and businesses would continue viewing it as a key market

GDP

• GDP per capita would increase by 15-25%
• This could affect international development metrics and funding
  • Could change eligibility for certain development programs
  • Might affect terms of international loans

Development Indicators

• Higher per capita metrics could improve Nigeria's standing in
  • Healthcare (doctors per capita)
  • Education (teachers per student)
• Some indicators might worsen (it depends on whether you are measuring a positive or negative indicator, and the denominator gets decreased)
  • Child mortality rates if these numbers aren't similarly inflated
  • Malaria prevalence

Domestic Politics

• Resource Allocation:
  • Federal revenue sharing between states could be significantly impacted
  • Alternatively, if all states are inflating their populations uniformly, existing allocation could be accurate and need minimal change

International Aid

• Donor calculations might shift
• Overall aid levels unlikely to change dramatically
  • Need still substantial
  • Political considerations often drive aid more than population

Future Research

Nigeria

I'm not a researcher, so it would be useful if someone has more time to spend looking into this. Here are some other areas that might be useful to dive into.

• Import/export data
• Interviews with people who have looked into this before
  • Population people (including the UN)
  • Some sceptics are also referenced in The Economist article
• Satellite imagery
  • Housing data combined with sampling
  • Comparing to similarly wealthy countries with more accurate data
  • WorldPop combines satellite data with census info
• Emissions data
• Electricity consumption data
• Migration data to countries with accurate statistics
  • Although this depends more on country relationships than size of the country
• Vaccination coverage for verifiable vaccines (i.e. BCG)
• Prediction markets
  • I'm sceptical as there isn't an easy way to verify what is true, if you made it based on the next census numbers, that could be (a) years away and (b) inaccurate

Other Countries

The UN population reports also include other countries^[8] that haven't had a recent census, and population estimates could be widely off the mark. This is mainly because of conflicts but is sometimes politically motivated. If someone is interested it may be useful to investigate population numbers in these countries.

• 1932 - Lebanon^[9]
• 1979 - Afghanistan
• 1984 -Democratic Republic of the Congo, Eritrea
• 1987 -Somalia
• 1989 - Uzbekistan
• 1997 - Iraq
• 2001 - Ukraine
• 2003 - Haiti
• 2004 - Syrian Arab Republic, Yemen
• 2005  -Cameroon, Nicaragua, United Arab Emirates
• 2006 - Libya, Nigeria
• 2007 - Ethiopia
• 2008 - Algeria, Burundi, Dem. People's Republic of Korea, South Sudan, Sudan

1. ^{^}

   And the data I did find from marketing consultancies sometimes conflicted in the same paragraph.

   WhatsApp: With over 95% of internet users aged 16-64 years, WhatsApp remains the leading social media platform in Nigeria. Its user base is estimated to be around 8.8 million.

   Facebook is the second most popular platform, with a significant user base of over 36 million in 2022. This represents a user penetration rate of around 16.7%.

   Unless someone from a large tech company wants to come forward with country usage stats and how that compares to the GDP/other metrics and expected usage given a country of that size and wealth, this is likely to be a dead end.

2. ^{^}

   Page 526 of 627 (1990 UN population report)

3. ^{^}

   Page 780 of 905 (1994 UN population report) 

4. ^{^}

   UN 2000 population estimates

5. ^{^}

   UN 2010 population estimates

6. ^{^}

   UN 2019 population estimates 

7. ^{^}

   UN 2024 population estimates 

8. ^{^}

   Whilst writing this up El Salvador released their 2024 census results (after a 17 year gap), with a population of 6.03 million, which was ~5% less than the 2024 UN projection of 6.324 million. They also had the bluest kick off census rally I've ever seen. 

   

9. ^{^}

   Politically not allowed