Cornstalks Photo: Big Grey Mare After several decades of rapid rise in world grain yields, it is now becoming more difficult to raise land productivity fast enough to keep up with the demands of a growing, increasingly affluent population. From 1950 to 1990, world grainland productivity increased by 2.2 percent per year, but from 1990 until 2009 it went up by only 1.3 percent annually. Despite some impressive local advances, the global loss of momentum in expanding food production is forcing us to think more seriously about reducing demand by stabilizing population, moving down the food chain, and reducing the use of grain to fuel cars.

One of the key components of Plan B, the Earth Policy Institute's ambitious strategy to save civilization, is to halt world population growth at no more than 8 billion by 2040. This will require an all-out population education effort to help people everywhere understand how fast the relationship between us and our natural support systems is deteriorating. It also means that we need a crash program to get reproductive health care and birth control services to the more than 200 million women today who want to plan their families but lack access to the means to do so.

While the effect of population growth on the demand for grain is rather clear, that of rising affluence is much less so. One of the questions I am often asked is, "How many people can the Earth support?" I answer with another question: "At what level of food consumption?" Using round numbers, at the U.S. level of 1,763 pounds of grain per person annually for food and feed, the 2-billion-ton annual world harvest of grain would support 2.5 billion people. At the Italian level of consumption of close to 880 pounds, the current harvest would support 5 billion people. At the 440 pounds of grain consumed by the average Indian, it would support 10 billion.

Of the roughly 1,763 pounds of grain consumed per person each year in the United States, about 220 pounds is eaten directly as bread, pasta, and breakfast cereals, while the bulk of the grain is consumed indirectly in the form of livestock and poultry products. By contrast, in India, where people consume just under 440 pounds of grain per year, or roughly a pound per day, nearly all grain is eaten directly to satisfy basic food energy needs. Little is available for conversion into livestock products.

Among the United States, Italy, and India, life expectancy is highest in Italy even though U.S. medical expenditures per person are much higher. People who live very low or very high on the food chain do not live as long as those at an intermediate level. People consuming a Mediterranean-type diet that includes meat, cheese, and seafood, but all in moderation, are healthier and live longer. People living high on the food chain can improve their health by moving down the food chain. For those who live in low-income countries like India, where a starchy staple such as rice can supply 60 percent or more of total caloric intake, eating more protein-rich foods can improve health and raise life expectancy.

Although we seldom consider the climate effect of various dietary options, they are substantial, to say the least. Gidon Eshel and Pamela A. Martin of the University of Chicago have studied this issue. They begin by noting that for Americans, the energy used to provide the typical diet and the energy used for personal transportation are roughly the same. They calculate that the range between the more and less carbon-intensive transportation options and dietary options is each about four to one. The Toyota Prius, for instance, uses roughly one-fourth as much fuel as a Chevrolet Suburban SUV. Similarly with diets, a plant-based diet requires roughly one-fourth as much energy as a diet rich in red meat. Shifting from the latter to a plant-based diet cuts greenhouse gas emissions almost as much as shifting from a Suburban to a Prius would.

Shifting from the more grain-intensive to the less grain-intensive forms of animal protein can also reduce pressure on the Earth's land and water resources. For example, shifting from grain-fed beef that requires roughly seven pounds of grain concentrate for each additional pound of live weight to poultry or catfish, which require roughly two pounds of grain per pound of live weight, substantially reduces grain use.

When considering how much animal protein to consume, it is useful to distinguish between grass-fed and grain-fed products. For example, most of the world's beef is produced with grass. Even in the United States, with an abundance of feedlots, over half of all beef cattle weight gain comes from grass rather than grain. The global area of grasslands, which is easily double the world cropland area and which is usually too steeply sloping or too arid to plow, can contribute to the food supply only if it is used for grazing to produce meat, milk, and cheese.

Beyond the role of grass in providing high-quality protein in our diets, it is sometimes assumed that we can increase the efficiency of land and water use by shifting from animal protein to high-quality plant protein, such as that from soybeans. It turns out, however, that since corn yields in the U.S. Midwest are three to four times those of soybeans, it may be more resource-efficient to produce corn and convert it into poultry or catfish at a ratio of two to one than to have everyone heavily reliant on soy.

Although population growth has been a source of growing demand ever since agriculture began, the large-scale conversion of grain into animal protein emerged only after World War II. The massive conversion of grain into fuel for cars began just a few years ago. If we are to reverse the spread of hunger, we will almost certainly have to reduce the latter use of grain. For perspective, the estimated 114 million tons of grain used to produce ethanol in 2009 in the United States is the food supply for 370 million people at average world grain consumption levels.

Quickly shifting to smaller families, moving down the food chain either by consuming less animal protein or by turning to more grain-efficient animal protein sources, and removing the incentives for converting food into fuel will help ensure that everyone has enough to eat. It will also lessen the pressures that lead to overpumping of groundwater and the clearing of tropical rainforests, two additional trends that threaten the viability of our global civilization.

Adapted from Chapter 9, "Feeding Eight Billion People Well," in Lester R. Brown, Plan B 4.0: Mobilizing to Save Civilization