The Earth Can Feed, Clothe, and House 12 Billion People
There are about 7 billion people on the planet today, with just over 9 billion expected by the year 2050. Given that many of the poorest 1-2 billion people on Earth may not even have enough to eat today, can we really expect to adequately feed more than 9 billion people in less than 40 years time? Well, in fact, our agricultural capacity to produce food, clothing, and shelter at the standards of the western world can accommodate nearly 12 billion people. The Earth today has about 4,100 million hectares of arable land (land with adequately fertile soil and sufficient rainfall capable of supporting traditional agriculture)--that's 41 million square kilometers or about 16 million square miles. A little less than 5% of this land is part of protected parks and wildlife preserves. Of the rest, only 15 million square kilometers are presently used for agriculture according to the FAO (Food and Agriculture Organization of the United Nations). Arable land in these statistics includes forest land and pasture lands that could possibly be used for traditional agriculture, but might be realistically needed for other purposes. A small amount of actively farmed land in the world (mostly in the Middle East) is actually not arable--think desert land made viable by irrigation and fertilization--so, this is not an absolute limit on agriculture.
We would likely not be content turning all arable land, much of which exists as forest and other semi-wild ecosystems, into high productivity grain farming. The effects on wildlife and aesthetics would be dramatic. So, let's assume that the world as a whole decides to protect twice the current arable area protected by parks and other reserves, and let's assume that another 10% of the area would be made up of semi-wild managed forest, managed game lands, and similar uses. That leaves a total of 33 million square kilometers of arable land available for agriculture of which we are currently using 16 million, or about half. One might then assume that we could easily support twice the current population, but this neglects that about a billion people are malnourished today, and many more are poorly clothed and housed (agriculture also produces the fibers for clothing).
To begin, we should assure that we can generate at least 2500 calories per person per year (the average need for an adult man), which is somewhat more than necessary because it does not account for the lower needs of women and children. 2500 calories per person per year for 9.2 billion people is a global need of 23 trillion calories per day in 2050. If every acre of arable land were planted with potatoes (the highest caloric yield per acre of any crop), we could produce 8 times more than we need to support all 9.2 billion individuals' energy needs, although potatoes alone would not meet people's nutritional requirements for protein and other nutrients. {interestingly, apples might provide even more calories per acre than potatoes}
Instead, relying on a one-third each mixture of corn, beans, and squash combined with a rotation in similar crops would provide for almost all nutritional needs including protein, vitamins, and minerals (data on yields and caloric content can be found here, here, here, and here, organic farming yields were used where available). This combination produces a average total of 2.7 million calories per acre or enough with all arable land in production to feed 2.5 times the population in 2050. If we were to allow for more variety in our diets and incorporate additional servings of a wider array fruits and vegetables our average yield might fall to 2.4 million calories per acre and reduce the surplus to 2.3 times the population in 2050 (using tomatoes at 80 calories per pound and 20,000 lbs per acre as a proxy for a mix of other vegetables).
Next, however, we should account for the needs of fiber (textiles and paper), and timber (paper and construction materials). Cotton consumption in the US peaked in the 1990's at about 6.7 kg per person per year, and is currently about half of that. If we use the higher figure as a basis for worldwide needs in a fully developed economy, we would have a need of 62 million metric tons (tonnes), or a little more than twice the estimated cotton production for 2012. At current yields of 0.8 tonnes per hectare, worldwide fiber needs could be met by less than one million square kilometers (0.77 million sq. km). The demand for other fibers like wool and synthetics can be estimated at a value equal to cotton demand, for another 0.8 million sq. km. Total timber consumption in the US in 2005 (including pulp for paper, hardwood, softwood lumber, veneer, and other products) was about 20 billion cubic feet or 67 cubic feet per person. This translates into a global demand of 620 billion cubic feet or 12 billion tonnes. Using sustainable forest management practices, a yield of 6,600 kg per hectare of wood might be possible, while bamboo could yield up to 40,000 kg per hectare. This implies arable land needs of 3 million square kilometers (bamboo) to 18 million square kilometers (mix of hard and soft wood). Assuming bamboo will be able to meet half of the demand for these products, we would have a projected need of 10 million square kilometers for all timber together with the 1.6 million square kilometers for fiber.
So, the food (14 M sq.km), timber (10 M sq.km), and fiber (1.6 M sq.km) needs of the projected population in 2050 can be met with only 78% of our available arable land (26 of 33 M sq.km). In fact, every man, woman, and child on the planet would be able to consume as much of these things as Americans typically do (or more in the case of vegetables), and that level of production would satisfy up to 11.6 billion people. While annual crop rotations are assumed in these calculations, multiple crops in a given year are not even though they are common in many places. What these figures do not include explicitly are animal products, although fish and game harvested in sustainable quantities would be an addition with no impact on the other values as would animals raised on agricultural waste products including wheat and rice straw, winter cover crops such as alfalfa, and discarded produce as well as those raised on pasture lands that do not qualify as arable (hill sides, rocky grasslands, arid grasslands, etc.). Adding animals raised on grain or other primary agricultural produce would reduce the caloric and protein productivity of the land overall, but would still be possible given the 7 million square kilometers of unused arable land revealed in this scenario. Future changes in the amount of arable land due to climate change or sea level rise are not considered here.
So, why don't we feed everyone sufficiently today given that we have more than enough worldwide agricultural land in production? There are several reasons. One is waste. In the US 20-40% of agricultural produce is wasted for one reason or another. Another is high value but low productivity agricultural activities including grain fed cattle and alcohol production. Another is the combination of adverse incentives created by rich and poor world governments actively involving themselves in the agricultural markets to different ends. None of these practices needs to be eliminated in order to supply adequate food and other products to the worldwide population, but official encouragement and innovation and open trade between regions will be required to meet these needs. We can say then that we are not necessarily heading for an impending disaster, but whether we succeed in sufficiently providing for everyone will remain an open question.
