Is the world getting better or worse?

This may be the most important line in human history. For almost all of it, roughly four newborns in ten never reached their fifth birthday — a grief so routine it shaped how people named and counted their children. Today it is under four in a hundred, and still falling. Nothing else on this page has travelled so far.

Much of that survival was bought cheaply. Routine childhood vaccination went from reaching one infant in five in 1980 to more than four in five — a few dollars a dose standing between a child and diseases that once emptied nurseries.

The rich world’s deadliest habit is finally receding: adult smoking has fallen by roughly a third since 2000, the slow payoff of taxes, advertising bans, and plain packaging. The cigarette took a century to conquer the world and is taking decades to lose it.

Survival is still an accident of geography. A child in Sub-Saharan Africa is more than five times likelier to die before turning five than one in East Asia — and though every line is falling, the gap measured in actual lives remains enormous.

The global average hides a gap measured in children’s lives. A baby born in Nigeria is roughly forty-eight times more likely to die before turning five than one born in Finland — about 116 deaths per 1,000 against barely two. The hopeful half: India and Bangladesh, once near the top of this grim list, have driven their rates down into the twenties and thirties, closer now to Brazil than to Chad or Somalia. Where a child is born still decides whether they live — only less ruthlessly than it did a generation ago.

Average output per person has roughly tripled since 1960 alone. An 'average tripling’ is the gentlest possible description of what that meant on the ground: hundreds of millions of people moving from bare subsistence to something like security inside a single working life.

Pull the camera back two hundred years and the hockey stick appears. For all of history before about 1800, almost everyone everywhere was poor; the roughly fifteen-fold rise since is the single most consequential economic event there has ever been — and it is barely two centuries old, a blink in human time.

The catch-up is the defining economic story of our era: East Asia has closed much of the gap with the West inside one generation, and South Asia is now climbing the same curve a few decades behind. Sub-Saharan Africa is the exception — growing, but falling further behind in absolute dollars even as it does.

Average world income tells you almost nothing; the spread is the reality people live in. A typical Swiss earns around $90,000 a year and an American $66,000, while in DR Congo or Burundi the figure is a few hundred dollars — a gap of more than two hundred to one. The movement is in the middle: China has climbed to about $13,000 a head, several times India’s $2,400, which is itself more than double Ethiopia’s. The ladder is real, and millions have climbed it within a lifetime — but the rungs stay brutally far apart.

In 1990 more than two-fifths of humanity lived in extreme poverty; today barely one person in ten does. That is well over a billion people crossing the line in three decades — the fastest mass escape from material misery ever recorded, most of it in China and India.

Raise the bar to a still-meagre $8.30 a day — the World Bank’s poverty line for an upper-middle-income country — and the triumph deflates: nearly half the world remains beneath it. 'No longer extremely poor’ is a long way from comfortable, secure, or reliably fed.

The remaining extreme poverty is collapsing across Asia and concentrating in Sub-Saharan Africa, where it has fallen far more slowly. The next billion lifted out will be far harder won than the last — poorer, more rural, and increasingly in places fractured by conflict.

Pull the regional average apart and the spread turns brutal: in DR Congo and Mozambique four people in five still live below $3.00 a day, while China has all but erased that kind of poverty and Vietnam is nearly there. India — for decades the largest single reservoir of the world’s poor — is now down to about one person in twenty. What remains is not spread evenly across the poor world; it has pooled in a band of African countries where it falls slowest, and in some years not at all.

Two countries make a point a regional average buries. In the late 1970s nearly six in ten Indians lived below $3.00 a day — a deeper hole than Nigeria’s, where about half did. Then they split: India’s rate fell to roughly one in twenty by its latest survey, while Nigeria’s gave only a little, still above four in ten. Same starting misery, opposite endings. Both lines stop at 2022 because poverty is counted by household survey, not every year — India went a full decade between counts — so only the modeled world and regional lines above run on toward today.

The line climbs about half a percentage point a year — real progress, but slow enough that at this pace the last billion people are not reached until well into the second half of the century.

The fastest-moving water statistic on Earth: open defecation fell from one person in five to fewer than one in twenty in a single generation, driven by vast toilet-building drives across India and Asia.

The divergence the world average hides — most regions now sit near three-quarters covered, while barely a third of Sub-Saharan Africans have safely-managed water at home.

Safely managed drinking water is the quiet dividing line of development. In Germany and Japan it reaches essentially everyone; in Chad just 6 in 100 have it, in DR Congo about 12. The middle holds the demographic weight: India has pushed safe water to roughly three-quarters of its people, Bangladesh to under 60. A tap that runs clean is so ordinary in rich countries that it goes unseen — and so absent across the Sahel that fetching water still shapes the day.

Two centuries ago reading was a rare skill — perhaps one person in ten, mostly men, mostly wealthy. That it is now the near-universal inheritance of humanity is among the quietest revolutions on this page, and the foundation beneath almost every other gain here: a population that can read is one that can be taught everything else.

The frontier is Sub-Saharan Africa, where adult literacy has climbed from under half to roughly two-thirds in a single generation — rapid by any historical standard, yet still the world’s widest remaining gap.

Reading is now so close to universal across most of the world that the exceptions stand out sharply. China and Brazil sit above 90%, and India has reached 82 — a staggering climb for a country that was majority-illiterate within living memory. The hold-outs cluster in the Sahel: in Niger and Chad only about a third of adults can read, the legacy of schooling that never arrived and that fell hardest on girls and the rural poor.

The line tells two stories at once: a steady decline through the 2000s, then a reversal after 2015 as war, climate shocks, and the pandemic pushed the share of the hungry back upward. Progress against hunger is not a ratchet — it can be undone, and recently has been.

The clearest case on this whole page of an average that lies. The world figure has been flat-to-falling while undernourishment across Sub-Saharan Africa has climbed back above one person in five — two decades of progress quietly erased behind a reassuring global line.

Hunger has retreated to particular places rather than vanishing. In Madagascar and DR Congo around four in ten people do not get enough to eat; across China and the rich world the figure sits at the measurement floor, near zero. The unfinished middle is the telling part: India, despite its growth and its grain surpluses, still leaves about one in eight undernourished, and Nigeria closer to one in five. Enough food exists — reaching everyone with it is the part that stalls.

Measured by law alone, the direction is unmistakable: across 190 countries the average woman now holds about three-quarters of the legal rights of the average man, up from under half in 1970. What a statute grants and what a society concedes are different things — but the statute, at least, moves only one way.

In the classroom the oldest inequality has essentially closed: girls worldwide now enrol about as often as boys, a parity that would have stunned every century before this one.

Outside the classroom the gap holds. Women’s share of paid work has barely shifted in thirty years, stuck near 70% of men’s — equality in school has not carried into the economy or the boardroom.

The gain measured in lives: a woman is now less than half as likely to die giving birth as in the late 1980s. Childbirth is still far deadlier in poor countries than rich ones, but that gap is narrowing faster than almost any other in women’s health.

One composite of reproductive health, political power and paid work — and on it, every region is falling toward equality. South Asia has moved fastest, from the most unequal region in 1990 to about the global average today; Sub-Saharan Africa now sits highest, while Europe and Central Asia approaches the floor. A lower line is a more equal one.

The gap between women’s and men’s lives is widest where you would guess and narrowest where you would too. Yemen and Afghanistan sit near the top of the inequality scale, their women shut out of schooling and paid work; Norway and Denmark sit at nearly zero, about as close to parity as any society has reached. India lands in the awkward middle at 0.40 — better than a generation ago, still far from its own stated ideals. The United States, at 0.17, trails much of Europe.

In 2005 fewer than one person in six had ever been online; today it is nearly three in four. Few technologies in history have reached most of humanity this fast — and what that connection does to us, for better and worse, is the harder question no coverage figure can answer.

The divide is real but narrowing fast: Sub-Saharan Africa has gone from almost entirely offline to roughly a third connected in under twenty years, even as it stays far behind the rich world’s near-saturation.

The internet went from rich-world luxury to near-necessity in two decades, but the on-ramp is uneven. The Gulf states and South Korea are effectively fully connected; the United States and China sit in the low 90s. The frontier is the bottom of the list: India has crossed 70%, a billion people brought online fast, while Ethiopia and Chad remain under 20%. The people still offline are disproportionately the rural poor, and disproportionately women — last to get this, as they were last to get most things.

Safer cars and roads have bent the death rate down even as the number of vehicles on Earth exploded — quiet, unglamorous engineering progress. Yet more than a million people still die on the roads each year, most of them young, and the toll is climbing where cars are arriving fastest.

The danger has migrated to the developing world: a road death is now roughly four times likelier in Sub-Saharan Africa than in Europe, where half a century of seatbelts, speed limits, and safer design cut the rate in half.

Dying on the road is one of the most unequal risks there is. Saudi Arabia and DR Congo lose more than 30 people per 100,000 to traffic each year; Sweden and the United Kingdom lose around three. The pattern mostly tracks income, with one loud exception. The United States, at 12.7, kills roughly four times as many of its people on the road as its peer democracies do — a gap measured in tens of thousands of lives a year.

Every year humanity burns more carbon than the last. Annual emissions reached almost 39 billion tonnes in 2024 — another record — and the curve has bent upward, not down, for two hundred years. Until this one line turns over, everything else only slows the rate at which the problem grows.

The one line bending the right way, and fast: wind and solar climbed from a rounding error in 2000 to roughly a sixth of the world’s electricity by 2025, the share doubling every few years. It is the quickest-scaling source of power in history. The only open question is whether it scales quickly enough.

This is the warming already banked. The planet sits about 1.2°C above its mid-century average, and the sharpest part of the climb is the most recent stretch. Every tenth of a degree means more heat in the oceans, longer droughts, and a higher floor under every summer still to come.

Asia now emits more than the rest of the world put together — over 20 of those 39 billion tonnes — having passed Europe and North America early this century. Yet the ledger of who caused it runs the other way: the rich world burned first and longest, while Africa, home to a sixth of humanity, still accounts for under 4% of the carbon.

The damage and the blame point in opposite directions. A person in Qatar emits about 41 tonnes of CO₂ a year and an American 14, while a Nigerian emits well under one tonne and a Congolese barely a tenth of one — a gap of several hundred to one between top and bottom. China now sits in the upper-middle at about 8.7 tonnes, having industrialised on coal. The cruelty of the arithmetic is that the lowest emitters, across Africa and South Asia, stand first in line for the floods and heat the highest emitters set in motion.

The Red List Index measures how close assessed species — birds, mammals, amphibians, corals — sit to extinction, where 1.0 means none are threatened. It has slipped from 0.82 in 1993 to 0.72 today, each downward step a species sliding a category nearer to gone. Nothing on this line has been reversed; the only question is the speed of the descent.

The one hopeful line here, and a fragile one. The world still has less forest each passing year — cover fell from about 4.34 to 4.14 billion hectares since 1990, an area the size of Mexico — but the loss is widening more slowly than it once did, as replanting across Asia and Europe offsets some of the tropical clearing. A wound still open, bleeding less fast.

This is the figure behind the headline that wildlife has collapsed by two-thirds. Among the vertebrate populations scientists track, average size fell about 73% between 1970 and 2020 — though it is an average of trends, not a head count, and a few crashing populations can drag it down further than the typical species has actually fallen. Real and alarming, and more contested than the round number lets on.

Two directions at once. Latin America — the Amazon above all — has lost the most, from 53% forested in 1990 to 46% today, and Sub-Saharan Africa has thinned from 34% to 26%. Europe and East Asia, having felled their forests centuries ago, are slowly putting a little back.

The line nobody puts on the progress charts. Humanity killed about 8 billion land animals for food in 1961; in 2024 it killed over 86 billion, almost all of them chickens, and the curve rises with the world’s appetite for meat. Every gain in human wealth on this page has so far meant more animals in cages — the one column where a climbing number is unambiguously worse.

The numbers here are almost too large to feel. China slaughters more than 16 billion land animals a year for meat and the United States about 10 billion — overwhelmingly chickens, which now outnumber every other farmed animal many times over. The totals track population and how industrial a country’s farming has become, not appetite alone: France and Germany each kill under a billion, yet eat about as much meat per person as anyone. However the line is drawn, the system takes more animals every year than the last, with little public attention to the curve.

Seven centuries of records, pointing one way. A person in medieval England or Italy was dozens of times more likely to be murdered than someone in Western Europe today — the rate fell from above 30 per 100,000 in the 1300s to well under one now. Daily life grew so much safer, so gradually, that we mistake the calm for the natural state of things. It is the longest record of human behaviour on the page.

The reversal that broke the long peace. Deaths in state-based conflict had fallen to about 12,000 in 2005 — tiny by any historical measure — before Syria, Ethiopia, Ukraine and Gaza drove the toll back above 120,000 by 2024. The series is jagged by nature, one large war dominating a decade, so a single year matters less than the slope — and the slope since 2020 points up.

A direct readout of that fraying. The number of people driven across a border by war or persecution has climbed to around 31 million, the most in the six decades of record here, up from a few hundred thousand in 1960. Behind the line sit a handful of catastrophes — Syria, Ukraine, Afghanistan, Sudan — and it leaves out the even larger number displaced inside their own countries.

The Cold War’s receding shadow, and the most double-edged 'good’ line on the page. Stockpiles fell from a peak near 70,000 warheads in 1986 to about 12,000 now, as Washington and Moscow stood down most of their arsenals. But the decline has flattened, modernisation is back, and 12,000 warheads remain enough to end civilisation several times over — a smaller sword over a still-bare neck.

Where lethal violence lives now is not where history would guess. Latin America is the deadliest peacetime region on Earth — around 20 homicides per 100,000 people, roughly ten times Europe’s rate and twenty times East Asia’s — driven by gangs, drugs and weak courts rather than war. The medieval decline reached everywhere; the modern burden settled unevenly.

Murder is wildly uneven across the map. Jamaica and South Africa run above forty killings per 100,000 — common enough to shape daily life — while Singapore and Japan sit so near zero that a single murder can make national news. Latin America carries much of the world’s lethal violence, with Mexico and Colombia near twenty-five and Brazil close behind. The United States is the outlier among rich nations at 5.8 — roughly four times France’s rate and nearly thirty times Japan’s — proof that wealth alone is no shield.

Two centuries of upheaval, and the top still takes the same cut. The richest 1% captured 19.7% of world income in 1820 and 20.7% today, with a dip in the mid-20th century — when war and steep taxes briefly squeezed the very top — long since clawed back. The world transformed beyond recognition; the share flowing to the top of it barely moved.

The floor rose even as the ceiling held. Measured against an absolute line, the bottom is far better off: the share living under $8.30 a day fell from 70% in 1990 to about 44% now, as Asia’s poor climbed toward the middle. Shares of the pie can stay fixed while billions still escape destitution — which is exactly why the floor and the ceiling tell opposite stories.

The other end of the same ledger. The bottom half of the world’s adults — about four billion people — share under 8% of global income between them, down from roughly 14% in 1820. For every dollar earned worldwide, the poorest half splits less than eight cents, while the richest hundredth takes more than twice that from far fewer hands.

Inequality is not one number but many. Latin America has been the most top-heavy region for two centuries, its richest 1% still taking nearly a quarter of all income; Europe is the most equal, having roughly halved the top’s share since 1900. Where the rich ended up taking less, it was politics — unions, taxes, welfare — that made it so, not growth alone.

Inequality within nations varies as much as the gulf between them. South Africa and Namibia, still shaped by their histories, are among the most unequal societies ever measured — Gini scores in the mid-to-high 50s, a sliver of the population holding most of the income. The United States, near 42, is markedly more unequal than Germany or Sweden, which sit around 30. China, once imagined as egalitarian, now reads about 36 — above Germany and Britain, if still short of the US. The rich world has reached nothing like a shared answer on how much gap is bearable.

Two of every five units of the world’s electricity now come from a source that burns no carbon — wind, solar, hydro and nuclear together — up from about a third in 2000. The climb is real but gentler than the headlines imply, because surging demand keeps refilling the fossil side as fast as clean power is added. The share is rising; the carbon, not yet falling.

The number that spoils the celebration. The world now generates almost twice as much fossil-fuelled electricity as in 2000 — roughly 18,000 terawatt-hours against 10,000 — because clean power has so far been stacked on top of coal and gas, not swapped in for them. Until this line turns down, the rising clean share is winning a race the atmosphere is still losing.

Perhaps the most consequential line in all of energy. A watt of solar cost about $128 in 1975 and costs around 26 cents today — a fall of roughly five hundredfold, among the steepest sustained price collapses ever recorded for anything made by people. Cheap sunlight is why clean power is no longer a subsidy story but an economic one: across most of the world, new solar is now simply the cheapest electricity there is.

One transition running at four speeds. Europe already draws more than three-fifths of its power from low-carbon sources; Africa, barely a quarter — and the gap is less about will than money, since the grids and solar farms that decarbonise a country cost what poor ones do not have. The cleanest electricity belongs, for now, to those who burned the most coal first.

More research pours out each year than anyone could read in a thousand lifetimes. The world published under 100,000 scholarly works in 1900 and more than 10 million in 2024 — roughly a hundredfold rise, steepest in the last two decades. But volume is not discovery; the same flood feeds a growing worry that ever more papers are chasing ever thinner slices of the unknown.

A quieter gain hides inside the flood. More than half of all new research is now free to read: open-access works rose from a trickle to over 6 million a year, around three-fifths of the total, prising science loose from paywalls that long shut out anyone without a university login. The knowledge is multiplying, and more of it than ever is public.

The centre of gravity in research is sliding east. East Asia and the Pacific now spends a larger share of its economy on research than Europe does — about 2.8% against 2.1% — a reversal from a generation ago, when the West held an easy lead. China above all has gone from scientific bit-player to rivalling the United States in sheer output.

The geography of science has been redrawn inside a generation. China now publishes nearly 900,000 scientific papers a year, almost twice the United States, with India third and climbing fast — a reversal that would have read as fantasy in 1990. Germany and Japan each still sit near 100,000, producing far more per person than the giants above them. Raw output rewards size and says nothing about whether the work is any good — but the centre of gravity has plainly moved east.

Ask the world to rate its life from zero to ten and the answer barely moves — about 5.4, near-flat across every year it has been measured well. The strangeness is in the comparison: incomes, lifespans and schooling all climbed steeply over the same span while this line held still. Wealthier, healthier, no more satisfied.

The hopeful counter-current beneath that flat average. The global suicide rate fell by about a third this century, from roughly 13.6 to 8.8 deaths per 100,000 — a quiet retreat at the very edge of human despair, led by sharp declines in China and India. Fewer people are reaching the last resort, even as everyday contentment refuses to climb.

Ask people to score their own lives from zero to ten and the extremes rarely surprise: Finland leads at 7.8 with the Nordic countries just behind, while Afghanistan sits at the floor, 1.4, a figure that reads like a measure of collapse. The middle is where it gets interesting. China reports 6.1, comfortably above India’s 4.5 and Nigeria’s 4.8, despite India’s far louder optimism about its future. Income buys a great deal of life satisfaction — but past a point, peace and health appear to buy more.

The average person lives in a freer country than their great-grandparents did — and a less free one than they did in 2012. The index rose for two centuries to a high around 0.41, then slipped to about 0.37 as elected leaders hollowed out courts and presses from within. This is the 'third wave of autocratisation’, and weighted by population it is unmistakable.

The retreat is sharpest where democracy seemed safest. North America, long held up as the model, has fallen hardest — its score down by nearly a third from a 2010 peak as institutions strained under pressure. Even Europe, still the most democratic region, now sits below its 2009 high. The backsliding runs through the wealthy democracies that thought themselves immune — while South America, a generation on from its dictatorships, is the rare region still well above where its past left it.

Democracy is not a switch that is on or off; it runs by degrees. Denmark and Sweden sit near the top of the scale, around 0.85, while the United States now occupies an uneasy middle at 0.57 — still a democracy, a notch below the nations it once lectured. India, the largest democracy by headcount, scores lower again, about 0.26: V-Dem now rates it close to an 'electoral autocracy’, its institutions worn by a decade of majoritarian pressure. Russia and China sit near the floor, where elections are ritual or absent.

Having a job once did little to keep you out of extreme poverty; now, far more often, it does. The share of the world’s workers in extreme poverty collapsed from about a third in 2000 to under one in twelve today, on the back of rising wages across Asia. Work is paying its way out of misery — even where the work itself stays informal and insecure.

The catch falls on the young. Even as work lifts older adults clear of poverty, the global youth unemployment rate has drifted upward — from about 11.5% in 1991 to over 13% now, spiking near 17% in the pandemic year. A generation better schooled than any before it is finding the first rung of the ladder harder to reach.

Unemployment is the rare measure where a low number can hide bad news. South Africa’s 32% is a genuine catastrophe — a third of everyone who wants work cannot find it. But Thailand’s 0.8% and Vietnam’s 1.6% do not mean those economies are healthier than the United States or Germany; across much of the developing world almost no one can afford to be openly unemployed, so people take whatever informal, unprotected work they can find, and the count records them as employed. The figure captures whether you have any work at all, not whether it pays a living or lasts the year.

The world stitched itself together, then stopped pulling the thread. Trade more than doubled as a share of world output — from a quarter in 1970 to nearly 60% today — carrying hundreds of millions out of poverty as factories moved to wherever labour was cheap. But the climb ended with the 2008 crisis; the line has wobbled sideways ever since, just as politics turned toward tariffs and walls.

The deeper retreat hides in capital, not goods. Foreign direct investment — factories and firms built across borders — peaked near 5% of world output just before the 2008 crash and has sunk to barely 1.3% since, as companies grew wary of far-flung supply chains. Money, even more than trade, has been quietly heading home.

Trade openness tracks size and strategy, not just wealth. East Asia tripled its trade-to-GDP ratio as it became the world’s factory — from under 20% in 1960 to nearly 60% — while Europe’s tightly-knit economies run above 80%, each member trading heavily with its neighbours. Big, self-contained economies trade less as a share of themselves, which is why the world average sits well below its most open regions.

Trade as a share of GDP measures how open an economy is, not how rich or how well — which is why the ranking can mislead. Tiny, trade-built Singapore clears 300%, re-exporting far more than it makes; Vietnam, the world’s new factory floor, runs near 175%. The giants sit low for the opposite reason: the United States and China move enormous sums in absolute terms, but their vast internal markets leave trade at only a quarter to a third of output. Open is not the same as large, and neither is the same as fair.

The population bomb defused itself. The average woman bore nearly five children in 1950 and bears about 2.25 today — a hair above the level that holds a population steady, and still falling. It happened with almost no coercion: as children survived and women gained schooling and a say, families simply chose to be smaller.

The flip side of fewer births: a world growing old. People over 65 were about 5% of humanity in 1960 and are over 10% now — a doubling — and the share rises fastest exactly where birth rates fell first, in Europe and East Asia. Fewer workers supporting more retirees is the central strain of the century’s second half.

Two demographic worlds are pulling apart. East Asia and Latin America have dropped far below replacement — to about 1.3 and 1.8 children per woman — and face shrinking, greying populations, while Sub-Saharan Africa still averages over four and will supply most of this century’s population growth. The future is being born in very different places than the recent past was.

The world is splitting into two demographic futures, with replacement set at about 2.1 children per woman. At one end, Niger and DR Congo still average close to six — young, fast-growing populations that need schools and jobs faster than they can build them. At the other, South Korea has fallen to 0.75 and China to almost exactly one, a collapse that means each generation roughly halves the next. India has just slipped below replacement, to 1.96, quietly joining the shrinking half of humanity. Too many and too few are both real problems, and almost nowhere now sits at the steady middle.

Public health’s slow, enormous victory. The share of adults who smoke fell from about 34% in 2000 to under 22% today, pushed down by taxes, advertising bans and plain packaging — the patient dismantling of a century-old habit. The cigarette is losing the rich world; the open question is the poorer countries where the tobacco industry now sells hardest.

The other drug, holding roughly level. Global alcohol intake has barely changed this century — around 5 litres of pure alcohol per adult a year — but the steady average hides a split: drinking is falling across the rich world and rising in fast-growing middle-income countries. A flat line stitched from two opposite trends.

Falling everywhere, at wildly different speeds. Latin America and Sub-Saharan Africa have roughly halved their smoking rates since 2000 — Africa’s is now under 10%, the world’s lowest — while East Asia, home to around a third of the world’s smokers, has barely shifted, still near 25%. The habit is fading from the West faster than from the places it reached last.

Tobacco still kills more people than almost anything else on this page, and where it grips varies enormously. More than a third of adults use it in Indonesia and Greece; in Nigeria barely three in a hundred do, a rare case of Africa faring best. The single number folds together very different habits — India’s is heavy with chewed tobacco, China’s with cigarettes smoked overwhelmingly by men. Britain, which once led the epidemic at far higher rates, has pushed use down to 14%, proof the habit can be unlearned at scale.