ARC
The latest on the accelerating Arc of Human Progress.  Seize yours.
Human Progress

Famine Used to Kill Millions Every Decade. It Almost Doesn’t Exist Anymore.

Famine is one of the oldest killers in human history. For most of recorded civilization, a bad harvest, a drought, a blight, a war — any of these could translate into millions of deaths. Not hundreds. Not thousands. Millions.

The Bengal famine of 1943 killed between 2 and 3 million people. The Chinese famine of 1959–61 — the worst in recorded history — killed between 15 and 55 million, depending on which estimate you trust. The Ethiopian famine of 1983–85 killed between 400,000 and 1 million. These were not ancient tragedies. They happened within living memory.

Since 2000, major famines have killed tens of thousands, not millions. The scale of the catastrophe has compressed by more than 95%. A phenomenon that once periodically killed populations equivalent to modern major cities has been reduced — not eliminated, but reduced — to a fraction of its former lethality.

That change is one of the most underreported achievements in human history.

What Actually Changed

The Green Revolution is where the story begins, though not where it ends. In the 1960s and 1970s, agronomist Norman Borlaug and his colleagues developed high-yield varieties of wheat and rice that produced two to three times the grain per acre of traditional varieties. Applied across South Asia — India, Pakistan, Bangladesh — the Green Revolution transformed the food supply of a region that had seemed chronically on the verge of mass starvation.

India, which faced severe famine risk in the 1960s and required emergency food aid from the United States, became a net food exporter within two decades. Bangladesh, once cited as a nation that could never feed itself, achieved food self-sufficiency. The mechanism was not charity. It was agricultural science, irrigation infrastructure, fertilizer, improved seeds — all scaled through market-based adoption by farmers who could see the yields with their own eyes.

The global food supply per person has increased by 32% since 1961 — from roughly 2,200 kilocalories per day to 2,900 kilocalories per day — while the world population more than doubled. We grew more food per person even as there were far more people to feed. That is the core fact. Everything else follows from it.

The Numbers on Hunger

In 1990, approximately 19% of the world's population was undernourished — about 1 billion people out of 5.3 billion. By 2019, that share had fallen to 9.2% — while the world population had grown by 2.5 billion people. The undernourished share nearly halved even as billions more joined the population who needed to be fed.

Sub-Saharan Africa — the region most associated with persistent hunger — saw cereal yields increase by more than 50% since 1990 (FAO data). The continent still faces serious food security challenges. But the trend is in one direction: better yields, better storage, better trade logistics, better market access for smallholder farmers.

The mechanism behind these gains was not primarily foreign aid, though aid played a role in specific crises. The primary drivers were: improved crop varieties, fertilizer adoption, irrigation, reduction in post-harvest losses through better storage, and — critically — the expansion of trade that allowed regions with surplus to supply regions with deficit. A crop failure in one country no longer automatically produces famine when global food markets can reroute supply within weeks.

The Famine Researcher's Uncomfortable Conclusion

Alex de Waal is one of the world's leading scholars of famine — a researcher who has spent decades studying its causes and mechanisms. His conclusion, uncomfortable for those who prefer simpler narratives, is pointed:

"Famine is now almost entirely a political phenomenon — it happens where governments deliberately obstruct aid or starve populations. It no longer happens from crop failure alone."

North Korea. Yemen. Sudan. The Tigray region of Ethiopia. Every major famine in the 21st century has occurred where political actors — governments, militias, sanctions regimes — have deliberately or negligently blocked the food supply. The food exists. The global system can move it. What stops it reaching starving people is human politics, not agricultural limits.

This is progress framed in a grim light. The good news: nature alone can no longer starve us. The bad news: human beings can still choose to, and sometimes do. But the transformation from "famine as natural disaster" to "famine as political atrocity" is a profound shift. It means famine is no longer inevitable. It is always someone's choice.

Three numbers that define the end of famine
9.2% — Share of global population undernourished in 2019, down from 19% in 1990, while population grew by 2.5 billion Source: FAO State of Food Security and Nutrition in the World, 2021
32% — Increase in global food supply per person since 1961, from 2,200 to 2,900 kcal/day — while population doubled Source: FAO Food Balance Sheets; Our World in Data
95%+ — Reduction in famine mortality since the 1970s peak; today's famines kill tens of thousands where past events killed millions Source: FEWS NET; Alex de Waal, Famine Crimes, updated analysis 2022

"Famine once killed millions in a single event. Today it is almost exclusively a weapon wielded by authoritarian governments — because nature alone no longer has the power to starve us."

What the Pessimists Predicted

In 1968, Stanford biologist Paul Ehrlich published The Population Bomb. His opening sentence: "The battle to feed all of humanity is over." He predicted that hundreds of millions would starve to death in the 1970s and 1980s. He was specific. He was confident. He was wrong in every directional detail.

The 1970s saw the Green Revolution expand. The 1980s saw food production continue to grow faster than population. The famines that did occur — Ethiopia, Sudan, Mozambique — were driven by war and political dysfunction, not by a global food system that had run out of capacity. The global food supply per person was higher at the end of each decade than at the start.

Ehrlich's error was a failure of systems thinking. He modeled population as a demand curve and farmland as a fixed supply, and drew a line to intersection. He did not model technological change, price signals, trade, or agricultural innovation. The same error that leads forecasters to miss solar cost declines also led Ehrlich to miss the food supply expansion: linear modeling applied to a system that changes nonlinearly.

The arc does not follow linear extrapolations. It follows the logic of human ingenuity applied to constrained problems. Give human beings a pressing food problem and sufficient market incentives, and they will engineer their way past it. They did.

What Remains

The work is not done. More than 700 million people still face chronic food insecurity. Climate change poses real risks to agricultural systems. The political famines of the 21st century are real atrocities that kill real people. None of this should be minimized.

But the long arc is clear: we went from a world where famine was a recurring, unavoidable feature of human civilization — where any region could be struck without warning by the failure of nature — to a world where food production has outpaced population growth for sixty consecutive years, where the global food supply is large enough to feed everyone alive, and where famine now requires active human malice to produce.

That is not a small thing. That is one of the most important transitions in the history of our species. And it happened quietly, in agricultural research stations and irrigation projects and grain markets, while the headlines focused on everything going wrong.

The arc bends. The pessimists keep getting surprised by it.