Poverty’s most measurable effect: whether children live to five

Poverty touches almost everything, from what people eat to how long children stay in school, but one of its effects is counted for every country in the world each year: whether children live to see their fifth birthday. In 2024, 4.9 million did not¹.

Nobody records poverty as a cause of death, so no official count says how many of these children poverty killed. The risk, though, follows income closely. A child born in a low-income country is about 14 times as likely to die before five as one born in a rich country, and inside poorer countries the children of the poorest families face about twice the risk of the richest¹.

4.9 million children a year, and why the estimate keeps moving

The UN’s best estimate is that 4,858,000 children died before their fifth birthday in 2024¹. The true figure is very likely between 4.7 and 5.2 million, because in the poorest countries many deaths are never registered and the numbers are pieced together from household surveys and censuses. A further 835,000 children aged 5 to 14 died the same year².

The figure is down from 13 million in 1990, a fall with its own story, told on the page about why child mortality has declined¹.

You may have seen other numbers. This page used to count 6.2 million deaths of children under 15 in 2018, a wider age group from an older round of estimates. And the UN first put 2022 at 4.9 million³, then revised it to 5.2 million as more survey data came in². The extra 300,000 children had died all along, and better data found them.

A child in a poor country is 14 times as likely to die

Yes, and by a wide margin. A child born in a low-income country in 2024 has about a 1 in 15 chance of dying before turning five (68.8 per 1,000), almost 14 times the risk in a high-income country, where it is about 1 in 200 (5.1 per 1,000)¹.

Family income matters inside countries too. Across the 106 low- and middle-income countries with good survey data, 57 of every 1,000 children in the poorest fifth of households die before five, against 28 in the richest fifth¹. That is about 1 in 18 against 1 in 36.

Other markers of disadvantage follow the same line. Children in rural areas are about 1.5 times as likely to die as children in towns and cities, and children of mothers with no schooling face a rate of 76 per 1,000, against 21 for children of mothers with higher education¹. The UN’s estimators put the cause plainly: these inequities “reflect unequal access to life-saving interventions and services”¹.

Most child deaths happen in sub-Saharan Africa and South Asia

Sub-Saharan Africa accounted for 2.8 million of 2024’s child deaths, 58%, and Southern Asia for 1.2 million, 25%. Together they had more than 80% of the deaths but 59% of the births¹. So the concentration is not only about where babies are born: a child born in these two regions is more likely to die than the world’s average child, which is the income gap above drawn on a map.

Sorted by income instead of geography, low- and lower-middle-income countries accounted for about 87% of the world’s child deaths in 2024, by our own division of the UN’s figures. All the world’s high-income countries together lost about 65,000 children under five, against 1.4 million in low-income countries alone².

Children in poor countries still die of infections

In sub-Saharan Africa, where most child deaths happen, the leading infectious diseases still cause 54% of deaths before five, against 9% in Europe and North America and 6% in Australia and New Zealand⁴. Worldwide, past the first month of life, malaria, pneumonia and diarrhoea together cause almost half of all deaths¹.

Across all ages under five, the leading causes worldwide in 2024 were being born too early (17.6% of deaths), pneumonia (13.4%), complications during birth (10.2%), malaria (9.2%) and diarrhoea (7.6%)¹.

Hunger is behind far more of these deaths than the records show. More than 100,000 children aged one month to five years were recorded as dying of severe acute malnutrition itself in 2024, and malnutrition weakens children’s defences against ordinary illnesses⁴. A widely cited 2013 estimate put undernutrition behind 45% of all child deaths in 2011⁵.

What poverty keeps out of reach: a skilled birth, a clinic, a doctor

Much of the gap comes down to care that is further away, or costs money, for poorer families. The clearest case is birth itself: being born too early and complications during birth together cause more than a quarter of child deaths¹, and those are the deaths that care at the birth itself is meant to prevent. In 2024, about 3 in 10 births to mothers in the world’s poorest fifth of households took place without a midwife, nurse or doctor present, against about 1 in 28 for the richest fifth. In sub-Saharan Africa, only 52% of the poorest mothers had a skilled attendant, against 95% of the richest⁶.

The gap repeats wherever care has to be paid for or travelled to. In sub-Saharan Africa’s latest surveys, 45% of births to the poorest families happen in a health facility, against 91% for the richest. A caesarean section, the operation that saves mother and baby when labour goes wrong, is used in 3.2% of the poorest mothers’ births and 16.8% of the richest⁶.

Sick children face the same divide. A child with pneumonia symptoms in sub-Saharan Africa is taken for care 41% of the time in the poorest families, against 63% in the richest⁶.

Bed nets are the exception. In the region’s latest surveys, 50.5% of the poorest fifth’s young children slept under an insecticide-treated net, against 45.7% in the richest fifth⁶. The surveys don’t say why, but they show the wealth gap in care is not inevitable.

The gap at birth is narrowing. In 2000 only 24% of births to the poorest fifth worldwide had a skilled attendant, against 79% for the richest; by 2024 it was 70% against 96%, and the difference had halved⁶.

Give poor families cash, and fewer children die

The most direct test of whether poverty itself kills children is to give poor families money and count what happens. In rural Kenya, researchers gave a one-off $1,000 to more than 10,500 poor households in randomly chosen villages. Infant deaths fell by 48% and deaths before five by 45%⁷. The study is a randomised trial but still a working paper, not yet peer-reviewed.

Most of the effect came through care. Deaths from causes around birth fell by three quarters, hospital births rose by 45%, and the effect was biggest for families living near a doctor and for the poorest households. When the money ran out, death rates went back to where they had been⁷. In what they called a more speculative implication, the authors drew a cautious lesson.

Government cash programmes point the same way, more modestly. A comparison across 37 low- and middle-income countries found under-five deaths about 8% lower where national programmes paid out⁹⁸, with more births attended by skilled workers, more children vaccinated against measles and less diarrhoea¹⁰.

Brazil’s Bolsa Família was associated with 17% lower under-five mortality where it reached most poor families, and deaths from malnutrition and diarrhoea fell furthest¹¹. Mexico’s Progresa was followed by 17% fewer infant deaths in poor rural areas, though not deaths in the first month of life¹². These are comparisons rather than experiments, so they show an association, and their effects can’t be added together.

Richer countries are not automatically healthier ones

If household money saves children, it might seem that national growth would too. The evidence is mixed. Economists David Cutler, Angus Deaton and Adriana Lleras-Muney found almost no link between how fast countries’ economies grew and how fast their life expectancy rose over 10, 20 or 40 years, and noted that China’s and India’s growth spurts came with slower, not faster, falls in infant deaths¹³.

The opposing case is that wealthier countries are healthier countries. Lant Pritchett and Lawrence Summers estimated that every 10% rise in income cuts child death rates by 2 to 4% in the long run, and attributed more than half a million child deaths in the developing world in 1990 alone to poor economic performance in the 1980s¹⁴.

Both can be right at once. Read together with the cash studies, they suggest money matters most in the hands of poor families who can spend it on care, and much less as a national average that may never reach them. If so, poverty harms children mainly through the clinics, midwives and vaccines it keeps out of reach, and public money can provide those whether or not a country grows rich.

Most child deaths could be prevented with care that already exists

Most of the 4.9 million children who died in 2024 could have been saved with proven, low-cost treatments and access to decent health care, according to the UN agencies that count them⁴. If every country had the death rates of rich countries, 17.4 million more children would live to five by 2030¹.

On current trends 27.3 million children would die before five between 2025 and 2030, almost 13 million of them newborns. Meeting the UN’s 2030 goal in every country would save 8 million of them¹. So many of those deaths falling in the first month of life suggests the hardest part will be care at birth, the care the poorest mothers are least likely to get.

The gap at birth is closing, slowly. If births to the poorest fifth keep gaining skilled attendants at the pace of 2020 to 2024, about one and a half percentage points a year, they would reach the level the richest have today around 2040, by our own arithmetic on UNICEF’s figures⁶. That is a rough extrapolation rather than a forecast, and the last stretch of any such gap is usually the slowest. The Kenya trial is the warning attached to it: when the money stopped, the deaths came back. Whether the money for care keeps coming after the aid cuts of 2025 is covered on the page about why child mortality has declined, and foreign aid has a page of its own.