The share of the world population living in extreme poverty has never declined as rapidly as in the past three decades.

The decline in China was particularly fast, and given that one in six people in the world lives there, we’re often asked whether the decline in global poverty was only due to the decline in China.

The chart shows the data that answers this question. In blue, we see the global decline. In red, we see the decline if we exclude China from the data. In the world outside of China, 33% lived in extreme poverty in 1990; by 2025, this share was down to 12%.

The large economic growth that lifted 940 million Chinese people out of extreme poverty since 1990 was a major contributor to the global decline in poverty. But the non-Chinese world also achieved a very large reduction.

It is not true that the global decline in poverty was only due to China. Extreme poverty has declined in China and the rest of the world.

Forty years ago in Japan, two babies were born for every person who died. Twenty years ago, these numbers were equal. And today, the ratio has reversed: one baby is born for every two people who die.

In the chart, you can see this change in the number of births and deaths over time.

Since deaths now greatly outnumber births, and because immigration is low, Japan’s population has started to shrink.

Debates over whether religion is booming or dying are common. What does the data say?

Most countries lack long-term data on religious identity, but results from the Pew Research Center offer insights into changes over the decade from 2010 to 2020. (Unfortunately, 2020 is the most recent year for which we have comparable global data.)

At a global level, there was barely any change. The share of people identifying with any religion dropped by just one percentage point, from 77% to 76%.

But religious affiliation did drop significantly across many countries in Europe, the Americas, and Oceania. You can see this drop for a selection of countries in the chart.

In Australia, rates dropped from 75% to 58%. In the United States and Chile, the percentage has decreased from roughly 85% to 70%.

So while religious affiliation is stable in many parts of the world, this data shows religion is becoming less prominent in others.

Note that this data is based on self-identification with any religion; it doesn’t tell us about changes in practices or rituals, such as prayer or attending services.

This chart shows one way to compare automated manufacturing across countries — it plots the number of robots per 1,000 manufacturing employees.

The chart shows very large differences between countries. South Korea stands out, with more than one robot for every ten manufacturing workers.

Singapore comes second, and China ranks third, close to Germany. The United States sits in the middle, close to the European average, below Switzerland, Denmark and Slovenia.

This perspective shows industrial robot adoption in relative terms. In another Data Insight, I looked at robot adoption in absolute terms. From that perspective, China stands out by a large margin: it’s a large economy with a huge manufacturing sector, and it has by far the largest stock of industrial robots.

Much of this expansion has happened recently: China’s annual installations increased 12-fold over a decade, helping it catch up to South Korea in terms of robots per worker.

Most people in the world are religious. When asked whether they identify with any religion, three-quarters of respondents choose one.

But in the chart, you can see huge differences in rates of religious affiliation across the world. In some countries, such as India and Pakistan, it’s almost universal: almost everyone identifies with a religion.

The opposite is true in China, where just one in ten people does. Several countries in East Asia, in particular, have particularly low rates of religious identification compared to other regions.

This doesn’t necessarily mean these populations hold no religious beliefs; they may still engage in activities that can be considered religious or spiritual, even though they don't describe themselves as belonging to any one in particular.

It might seem odd that countries would agree to import plastic waste from other countries, but many do so for the cheap materials or to feed specific manufacturing processes.

Environmentally, the trade in plastics has often been a concern, as it allows rich countries to effectively “dump” waste on poorer countries with weaker waste management systems.

The good news is that trade in plastic waste has fallen by more than two-thirds over the last decade. You can see this reduction in the chart.

China has been the biggest driver of this. It was once a large importer, but after a steep decline in trade in 2016 and a ban in 2018, many countries lost their largest export market.

In 2024, around 5 million tonnes of plastic waste were traded worldwide. For context, that is around 1% of the total plastic waste generated. What’s perhaps surprising is that most trade is now between high-income countries, which reduces the risk that this waste leaks into the environment.

Industrial robots are rapidly becoming a common part of manufacturing in some countries. The chart here shows how many new ones are installed each year in the industrialized countries for which we have available data from the International Federation of Robotics (IFR).

In this dataset, industrial robots are defined as automatically controlled, reprogrammable, and multipurpose machines used in industrial settings. The data covers only physical industrial robots, not software or consumer technologies.

The chart shows that in 2011, China, the United States, Japan, Germany, and South Korea were all installing similar numbers of these robots. However, in the decade that followed, the paths of these countries diverged. By 2023, annual installations in China had risen to 276,000 robots, a twelvefold increase.

Over the same period, installations in the United States, Japan, Germany, and South Korea also increased, but much more slowly: none of them even doubled. The United States, which saw the second-largest rise, went from 21,000 new installations in 2011 to 38,000 in 2023.

These figures refer to new robots installed each year; that is, annual additions to the existing stock of robots. The IFR also publishes data on the total number of robots in operation, and by this measure, China also had the largest installed base, at around 1.76 million robots in 2023.

Relative to its large manufacturing sector, China’s stock of robots today does not stand out – but the data here shows that this is changing quickly.

Most of our work on war and peace focuses on the people killed directly in the fighting. But war has many other costs: it worsens people’s health, leaves them without work, and pushes them out of their homes.

The chart shows this for the civil war in Syria. Since the war began in 2011, more than 400,000 people have been killed in the fighting. At the same time, annual deaths increased as more people died from other causes. Young children were especially affected: estimates suggest that the number of annual child deaths more than doubled.

The war has also forced millions of people to leave their homes: in total, more than seven million are displaced within Syria, and almost as many are refugees elsewhere.

It also became much harder for people to make a living. Average living standards, measured by GDP per capita, have more than halved since the war began. As a result, poverty and hunger have risen sharply.

These numbers come with uncertainty because conflict makes it hard and dangerous to collect data.

This shows that to understand the costs of war, we need to have a broad perspective and see its impacts on health, displacement, and living standards.

Three-quarters of people worldwide say they are religious. But rates of religious identity can vary a lot across countries, and so do the particular religions people follow.

In the map, you can see the most common religious affiliation for each country. This can include the “unaffiliated” who do not identify with any specific religion. This data is sourced from the Pew Research Center and is based on how people describe their own identity, regardless of their particular practices or beliefs.

As you can see, Christianity is the most common across much of Europe, the Americas, and Sub-Saharan Africa. Islam is the most common across North Africa and the Middle East, while Hinduism and Buddhism dominate across much of South Asia.

In East Asian countries such as China, Japan, and South Korea, the religiously unaffiliated are the biggest group. That doesn’t mean these populations hold no religious beliefs; they may still engage in activities that can be considered religious or spiritual, but they don’t describe themselves as belonging to any one in particular.

William Foege, who sadly died this week, is one of the reasons why this map ends in the 1970s.

The physician and epidemiologist is best known for his pivotal role in the global strategy to eradicate smallpox, a horrific disease estimated to have killed 300 million people.

Despite the world having an effective vaccine for more than a century, smallpox was still widespread across many parts of Africa and Asia in the mid-20th century.

Foege played a crucial role in developing the “ring vaccination strategy”, which focused on vaccinating people around each identified case, rather than attempting a population-wide vaccination strategy, which was difficult in countries with limited resources.

This strategy, combined with increased global funding efforts and support for local health programs, paved the way: country after country declared itself free of smallpox. You can see this drop-off through the decades in the map.

The disease was declared globally eradicated in 1980.

William Foege and his colleagues’ contributions are credited with saving millions, if not tens of millions of lives.

Several data sources show that theft in England and Wales has declined in recent decades.

One of those is police records — but they only capture reported crimes, and many people don’t report thefts. So it’s also important to draw on a second data source. The data we show here comes from reports based on face-to-face interviews with a representative sample of the population. In these interviews, the public is asked about their personal experiences of crimes in the previous 12 months.

On this chart, we’ve broken down the numbers by four different types of theft.

You can see a dramatic drop in vehicle-related thefts. These peaked in 1995, with an estimated 4.3 million incidents in England and Wales. While some of these incidents involved the actual stealing of a vehicle, many were either attempted break-ins or the theft of specific components, such as radios.

Burglaries — which involve someone breaking into a building to steal — also peaked in the mid-1990s.

Both types of incidents have decreased by more than 80% since then.

Pickpocketing or “snatching” has been more persistent. These crimes have decreased slightly from the 1990s and early 2000s, but have also experienced an increase in recent years.

In 2000, less than 10% of the population in Indonesia had access to clean cooking fuels. This is now over 90%, as the chart shows.

Clean cooking fuels are those that, when burned, emit less than the World Health Organization's recommended amounts of air pollutants. They reduce the burden of air pollution — and its health impacts — for the households that use them.

In 2007, the Indonesian government launched a national program to move from kerosene cooking fuels to liquefied petroleum gas.

This shift has greatly reduced particulate pollution and improved health outcomes. Death rates from indoor air pollution have fallen steeply.

Back in 1980, stomach cancer was the type of cancer that someone in Japan was most likely to die from. Its death rate — the number of deaths per 100,000 people — was over twice as high as the next largest killer, lung cancer.

But this is no longer the case. Since then, death rates from stomach cancer have dropped by more than 70%. You can see this change, compared to other cancers, in the chart.

While death rates of some other cancers have also fallen, these declines have been much smaller. Some types even saw an increase in death rates over these four decades.

Improvements in prevention, detection, and treatment have all contributed to this huge decrease in stomach cancer death rates. Stomach cancer is often caused by a bacterium called Helicobacter pylori; better hygiene and food safety have reduced its spread. Early screening for the infection has also made a big difference to survival rates.

This progress is not unique to Japan. Many countries, and the world as a whole, have seen a huge reduction in stomach cancer mortality.

Note that these death rates are age-standardized, which means they hold the age structure of the population constant. This allows us to understand how the risks of someone of a given age have changed over time.

Following the collapse of the Soviet Union and the social and economic disruption that it left behind, suicide rates in Lithuania increased rapidly. They climbed in the early 1990s and reached a peak in 1995. At 45 suicide deaths per 100,000 people, the country had one of the highest rates in the world.

But in the last few decades, rates have more than halved. You can see this in the chart.

Several factors likely contributed to the decline. Economic conditions improved, with average incomes more than doubling over just a decade from the mid-1990s to the mid-2000s and continuing to rise thereafter. In 2007, the country launched its first National Mental Health Strategy. A decade ago, it also developed a Suicide Prevention Bureau and a Suicide Prevention Action Plan.

This progress has saved many lives. Yet today it still has some of the highest rates in the world. That’s because suicide rates have not only fallen strongly in Lithuania, but in many countries — estimates for the global suicide rate suggest a 40% decline since 1995.

Since the late 20th century, astronomers and space agencies have taken steps to monitor the threat of large asteroids passing near Earth. They set up international efforts to find these objects early, track their paths, and learn more about what they’re made of, so we’d have the best chance of spotting a real collision risk in time.

As the chart shows, more than 40,000 near-Earth asteroids have been discovered and tracked since 1990. NASA estimates that we’ve already found over 90% of near-Earth objects larger than 1 kilometer. These are the most dangerous ones, because an impact at that size could cause global-scale damage.

Explore more interactive charts on space exploration and satellites →

If we look at income levels across countries in South and Southeast Asia, Malaysia is far richer than many of its neighbors. Its gross domestic product (GDP) per capita has almost doubled since 2000. It is now more than three times higher than that of Cambodia, Laos, and Bangladesh, and more than double that of Indonesia and Vietnam.

But if we look at measures of childhood nutrition, Malaysia is not doing better. You can see this in the chart. While its neighbors have made progress on childhood stunting — the share of children under 5 who are too short for their age — Malaysia has regressed. In 2000, 20% of children were “stunted”, and this has increased to 24%.

Malaysia also stands out at a global level. When we plot the share of children who are stunted against GDP per capita, the country is a clear outlier for its level of income. Most other countries at this level of economic development have rates below 10%.

Malaysia also does relatively poorly on other measures of malnutrition. On childhood wasting — when a child’s weight is too low for their height — it has one of the highest rates in the region.

The country is off track or worsening on most global nutrition targets.

In 2021, around 1.25 million people died from diarrheal diseases. Around a third of these deaths were children.

Two main factors explain why so many children still die from diarrhea, especially in poor countries: the persistence of risk factors such as poor sanitation and unsafe water, and the lack of access to effective treatment.

Here, I want to focus on the second factor: access to a particularly effective treatment, known as oral rehydration therapy (ORT), which is essentially a mixture of clean water, salts, and sugar. Simple as it may sound, researchers writing in the medical journal The Lancet called ORT “potentially the most important medical advance of the 20th century.”

The chart shows how often this treatment is used in a selection of African countries. This is based on household surveys asking caregivers of children under five who recently had diarrhea whether they received ORT.

There are large gaps: in Chad and Cameroon, fewer than one in five children with diarrhea received the treatment. This reflects a mix of challenges, including low awareness of its benefits and expensive or inconsistent supply.

Importantly, though, the chart also shows that rates are much higher in Sierra Leone, where around 85% of children received ORT. This shows that much higher coverage is possible. Sierra Leone has implemented several successful policies, including free treatment for children.

Not every child with diarrhea needs this treatment — some recover without it, depending on their health and circumstances. But ORT is cheap, safe, and easy to give. In low-income settings, especially, offering it widely as a cheap preventive measure can make a big difference for those who need it.

Over the past four decades, the global number of people dying from cancer each year has doubled. This can look like the world is losing its battle with cancer: people are more likely to develop it, and we’re getting no better at treating it. This isn’t true.

There are, of course, almost 4 billion more people in the world than in 1980. And many of those people are older. This matters a lot because cancer rates rise steeply with age.

The chart shows three different measures. Total deaths just count how many people died from cancer; this is the number that has doubled. Crude death rates, shown in yellow, adjust for population size; the increase shrinks from more than 100% to around 20%. Age-adjusted rates, shown in blue, also account for the fact that countries have older populations today; we can see that the fully age-adjusted rate has actually fallen by more than 20%.

It means that for the average person, the likelihood of dying from cancer in any given year is now lower than it was for someone of a similar age in the past. The world still has a long way to go in preventing and treating cancer, but it’s wrong to think that no progress has been made.

The map shows which European countries saw an increase in population, and which saw a decrease in the year from July 2022 to 2023.

The regional divide is stark: most countries with negative population change are located in the eastern and southern parts of Europe, while countries in the west and north saw population growth.

We focus on 2022–2023 as they are the most recent years in the UN’s latest World Population Prospects, published in 2024. Temporary shocks can influence year-to-year population changes, but this regional pattern is not unique to this particular year: you see it in earlier years too, and it also shows up when you look at population change over longer periods.

A key driver of this is migration. Most countries in Western and Northern Europe have had positive net migration (i.e., more people arriving than leaving). Many countries in Eastern and Southern Europe, in contrast, have had more people leaving than arriving.

Fertility rates have been declining across Europe, and all of these countries now have rates below the replacement level. That means that across much of Europe, since deaths now exceed births, the population would be shrinking without migration. In Western and Northern Europe, positive net migration has been offsetting this.

It’s a widespread view that child deaths are still a pressing problem in poorer countries, but not in rich ones.

I don’t think this is true, and I want to illustrate it with one example from the United States.

In 2023, 30,200 children died in the US. In the same year, 22,800 Americans of any age were killed through homicide. You can see this in the chart.

Few Americans would argue that murders are a “solved problem”. And this is certainly not what you’d take away from the news. As we showed in a recent article, homicides receive disproportionate coverage in both left- and right-leaning media, relative to the number of people who die from them.

The everyday tragedies of children dying from preterm births, neonatal sepsis, and asphyxia do not get nearly the same attention, but are no less important. These are problems that we can still make more progress on.