Table of Contents
ASK ANYONE, in Lagos or in Kinshasa, African migration tells more than what the pictures portray. A boat in the Mediterranean. A desert crossing. A young man arriving somewhere he was not invited. Almost every element of these pictures describes a minority of what actually happens, and the parts it leaves out are the parts that carry the real economic weight.
Let us start with the arithmetic. In 2024, the United Nations revealed that the number of international migrants worldwide was more than 304 million, which is 3.7% of the global population. That number has moved only modestly since 1990, when it stood at 2.9%.
Interestingly, the continent that supposedly empties is also a destination. Africa hosted 29.2m international migrants in 2024, up from 25.3m in 2020, a rise of about 15% in four years. However, of those migrants, 86% came from elsewhere in Africa. Sub-Saharan Africa received more people from within its own borders than it sent to Europe.
African migration in reality
Measured as a share of resident population, sub-Saharan Africa is one of the least migrant-dense regions on earth. Migrants made up 2.0% of the population there in 2024. In Europe the figure was 13%, in Northern America 16% and in Oceania 21%. The perception of a continent in flight survives largely because the small share that does reach Europe arrives visibly, and often dangerously.
Individual African countries tell a different story again. Gabon’s foreign-born share reached 17.7% in 2024, marginally above the United Kingdom’s 17.1%. Côte d’Ivoire hosts 2.9m migrants, about 9% of its population, drawn largely from Burkina Faso and Mali into cocoa and construction. Uganda’s migrant population quadrupled between 2010 and 2024, from 515,000 to just over 2m, most of it refugee arrivals from South Sudan and eastern Congo.
Exhibit 1· Africa’s ten largest migrant destinations are all in Africa
International migrants' resident in African countries, 2024.
Source: UN DESA, International Migrant Stock 2024. Figures count people resident in a country other than their country of birth, including refugees.
None of this makes emigration harmless. It relocates the question. If the volume is modest and most of it stays on the continent, the damage cannot lie in the number of departures. It lies in which people depart.
Why African brain drain is a selection problem
Here the evidence is unusually consistent. Afrobarometer, which has surveyed African public opinion since 1999, found in 2024 that almost half of respondents across 24 countries had considered moving abroad, including 27% who had thought about it a lot. That is a marked rise from the 37% recorded across 34 countries in 2016 and 2018.
The averages hide the important part. Before young adults and highly educated citizens were the two groups most likely to have considered leaving, with around half of each saying so. South Africa’s 2022 survey put the gradient starkly. Among South Africans with primary schooling or less, 10% had considered emigrating. Among the most educated, 38% had. The share rose to 42% among the wealthiest respondents and fell to 18% among those experiencing moderate deprivation.
The desire to leave rises with education and with income. So does the ability to act on it. Emigration selects for exactly the people a developing economy has spent the most to produce.
According to the World Bank, in most countries of origin the workers who emigrate have higher education levels than those who do not. What varies is the severity. Small, low-income and conflict-affected countries suffer most, because the denominator is tiny. In 2010 roughly 44% of Zimbabweans with tertiary education were living in an OECD country. The OECD has reported physician emigration rates above 70% for Liberia. About one in five African-born physicians works in a high-income country.
Exhibit 2· Where the wish to leave is strongest
Share of adults who say they have considered emigrating, Afrobarometer surveys.
Source: Afrobarometer. The 2024 and 2016/18 averages cover different country samples and are indicative of direction rather than strictly comparable.
The subsidy nobody records
When a Ugandan medical school trains a doctor who then registers in Manchester, a poor health system has made a capital transfer to a rich one. The training was financed domestically, through public subsidy, family savings or both. The return on it accrues to the National Health Service. No line item in either country’s accounts records the transaction, and no compensation flows back.
The scale is easier to feel than to measure, because training costs are rarely disaggregated and because doctors are the visible tip of a much broader category. Engineers, actuaries, software developers, agronomists, pilots and senior nurses all move on the same logic. The wage gap between a Nairobi salary and a Gulf or European one is large enough to overwhelm almost any patriotic argument, and it is widest precisely in the professions where domestic capacity is thinnest.
The compounding effect matters more than the annual outflow. A country that loses its consultants loses the people who would have trained the next cohort. Teaching hospitals hollow out from the senior end down, which lowers the quality of the training that remains, which raises the incentive for the next graduate to leave. Several West African medical faculties now describe this openly as their central operational problem.
The case on the other side
The pessimistic reading is incomplete, and the counter-argument deserves a fair hearing.
Yes, there is the case of remittances back to the continent. Reports show that remittances to Africa reached about $95bn in 2024, up from roughly $53bn in 2010, and rose over the same period from 3.6% to 5.1% of the continent’s GDP. Foreign direct investment was nominally larger in 2024 at $97bn, though about 36% of that figure was a single Egyptian property development, which leaves the comparable total nearer $62bn. Remittances are also steadier than investment flows and they behave counter-cyclically, arriving in larger volumes precisely when a country is in trouble.
Roughly a fifth of Africa’s remittance inflows, around $20bn in 2023, now originates within Africa itself. Zimbabwe drew about 37% of its inflows from South Africa in 2021. The intra-African economy that the migration data describes is also a financial system, and a growing one.
There are second-order gains too. The World Bank notes that the prospect of emigration can raise domestic investment in education, sometimes producing more graduates than the country loses. Diaspora networks carry trade, investment and knowledge back along the same routes. The Indian and Chinese technology sectors were both built partly by returnees, and no serious account of either leaves the diaspora out.
The honest objection to all of this is one of substitution. Remittances mostly fund consumption, school fees and medical bills at household level. They are a welcome income stream and a poor replacement for the output of the person who left. A surgeon’s remittances do not perform surgery. Where the diaspora dividend has genuinely materialised, in Bangalore or Shenzhen, it followed decades of domestic capacity building that made return worthwhile. The dividend is a consequence of a functioning home economy rather than a substitute for one.
Two systems that make the problem worse
The first is external. Rich-country visa regimes are calibrated to admit the highly credentialed and to refuse almost everyone else, while charging both the same non-refundable fee. Skilled worker routes are comparatively open. Visitor routes, the ones used by traders, conference delegates and families, are refused at rates above 50% for many African nationalities. The system filters for permanent extraction of talent and against the short, circular, commercial travel that would let people build something at home while staying connected abroad.
The second is internal, and it is harder to blame on anyone else. The African Union adopted its Protocol on Free Movement of Persons in January 2018, alongside the AfCFTA. Thirty-two member states signed it. Four have ratified it, namely Rwanda, Niger, Mali and São Tomé and Príncipe. Fifteen ratifications are needed for it to enter into force, and the most recent one was Niger’s in July 2019. Seven years on, the agreement governing the movement of Africans within Africa remains inoperative while the agreement governing the movement of African goods proceeds.
That gap has a direct bearing on the selection problem. Where regional movement is easy, as between Kenya, Rwanda and Uganda under East African Community arrangements, skilled people can take a better job two countries away without leaving the continent, the currency zone or the return journey behind. Where it is hard, the cheapest way for a Cameroonian engineer to improve their position may genuinely be a flight to Calgary.
What would change the arithmetic
Four measures come up repeatedly, and they vary considerably in difficulty.
Ratifying the free movement protocol is the cheapest and the most stalled. Eleven more ratifications would bring it into force, and the safeguards that anxious governments worry about are already written into the text, including the ability to suspend application. The obstacle is domestic politics rather than legal design.
Training compacts are the most direct answer to the unrecorded subsidy. Where a destination country systematically recruits from a specific origin country, it can fund training places there, as some bilateral health agreements already do. The principle is straightforward, which is that the beneficiary of a trained professional should contribute to the cost of producing them.
Reducing remittance costs is the fastest win available. Sending money to Africa remains the most expensive such transaction in the world. The average cost fell from 9.8% in the second quarter of 2016 to 8.2% in the first quarter of 2025, which is progress at a glacial pace and still roughly double the global target. Every percentage point removed is close to a billion dollars a year returned to African households.
Counting returnees is the least glamorous and possibly the most useful. No African country maintains a serious register of citizens who come back, what they bring and what they do next. Policy aimed at reversing a flow that nobody measures is policy aimed at a rumour.
The underlying point resists slogans. Africa is not being emptied, and the fixation on volume has obscured a more precise and more expensive problem. The continent trains people at considerable cost, loses a disproportionate share of the most highly trained among them, receives money back rather than capacity, and has not yet built the internal mobility that would let those people move without leaving. Each of those four things is a policy choice, and three of them are choices Africans make.
