By the end of this year, 270 million people could be living in famine conditions, according to the United Nations World Food Programme, up from an already staggering 149 million before COVID-19. Add in the disruptive effects of climate change and our planet’s ever-increasing population, and we’re looking at difficult times ahead.

By 2050 — the year when a growing list of nations aim to have zeroed out their contributions to climate change — the UN projects the global population will be 9.7 billion, on its way to topping out at 11 billion in 2100. The pressure to produce more food, or at least to make more money from agriculture, is driving nations to clear forests and wetlands for farms and divert scant freshwater to grow crops in the desert. But is that really necessary? Could we increase the food supply while also protecting precious wild resources?

Bloomberg Green embarked on a thought experiment: Given the number of people on Earth and the amount each one typically eats, could we feed them all using just our existing agricultural infrastructure? The answer turned out to be yes — and then some. Hypothetically, at least, we could feed the population of two Earths without clearing a single new acre of land. Here’s how we’d get there.

There are around 7.8 billion people on the planet now, and each needs about 1.4 kilograms of food on average per day, not including water. That means we require about 3.7 billion metric tons of food a year to feed everyone.

At the moment, the world produces about 4 billion metric tons of food per year — but about 1.3 billion tons goes to waste, according to the UN’s Food and Agriculture Organization (FAO). One international study led by researchers from the University of Edinburgh put that number even higher, suggesting that as much as 44% of agricultural production is never consumed.

Developing nations waste as much food as developed ones — what’s different is how it’s wasted. In rich countries, more than 40% of the losses occur at the retail and consumer levels. While some are trying to limit those losses by diverting expiring groceries to the poor, safety regulations often leave restaurants with little choice but to discard uneaten food.

In poorer countries, where households waste far less, the UN estimates that more than 40% of crop losses occur between the fields and store shelves. In India, for example, some of the biggest losses occur during the harvest itself, and fresh produce often rots on the way to market because of transportation delays and a lack of refrigeration. Research by Esri, a global geographic data supplier, suggests that if we factor in wasted potential from ill-used agricultural land, then only 30% to 50% of what we could grow ends up in our stomachs.

Most of the potential gains in production would come from farming the land we have more efficiently, through mechanization, better seeds and crop choices, and improved irrigation. Some waste and inefficiency is inevitable — poor harvests, pests, bad weather, and logistical difficulties are just functions of an imperfect system. But — in a perfect world, using Esri’s estimate, along with estimates of food waste from the UN and others — we could supply 9 billion metric tons of food a year, or 2.4 times as much as we’d need to feed every person on the planet.

It’s possible to increase the world’s food supply without relying on improved efficiency or reduced waste. According to research cited by the UN, 2.2 billion acres of degraded former farmland could be restored through more responsible use of fertilizer and irrigation. Some of that land would be used for reforestation, but even so, the rest could produce about 500 million tons of food annually.

Better eating habits could also help. Research from Italy suggests that as much as 140 million metric tons of excess food — enough to feed more than 270,000 people — gets consumed each year, leading to obesity and chronic health problems such as high blood pressure and Type 2 diabetes. (Getting that food to the undernourished, of course, would be a whole other matter, and colossally complicated.)

Not all edible crops are destined for the food shelf — almost 600 million hectares of land is used to grow crops for ethanol, for instance. That means vast quantities of corn, wheat, rice, sorghum, sugar cane, cassava, and sugar beets are going into cars’ fuel tanks instead of people’s mouths. The move toward electric and hydrogen-powered vehicles could take a significant chunk out of this market. If all of that land were converted back to food production — a big assumption, but go with it — we could feed 280 million more people, according to one study from the University of Virginia.

Finally, there’s the food we grow to feed the animals we eat. Almost half the grain in the world goes to feed livestock, according to the UN. Animal husbandry is an ancient part of many cultures; livestock are used to produce other things besides meat, such as wool and dairy products; and many cattle, especially in developing nations, consume pasture or vegetable matter not eaten by humans. Still, reducing meat consumption by half could free up a quarter of the world’s grain supply for humans.

Any number of combinations of the actions described here would put global food output well above the 7.4 billion metric tons needed to feed almost 16 billion people. Of course, these are just back-of-the-envelope calculations. Politics, economics, culture, and commerce all play major parts in development and trade, ensuring that no agricultural system will ever be optimal. But even with more realistic targets for reducing waste, reclaiming lost farmland, and redistributing our food supply, we have more than enough land to take care of a growing global family.

That’s not the only benefit that could come from better agricultural management. Ruminant animals alone — e.g., sheep, goats, and cows — account for at least 2.5% of global greenhouse gas production; eating less of them means raising fewer of them, so emissions would shrink. Meanwhile, the UN estimates that restoring carbon and nutrients to degraded soil for cultivation would cost a relatively reasonable $300 billion and suck up the equivalent of all our carbon emissions for the next two decades, buying us precious time to bring the rest of the global economy in line with net-zero.