From Jacob van Etten.
Some demographic projections tell us that global population numbers will grow to 9 billion in 2100 and stabilize around this number. So how can people three generations down the line feed themselves, while still conserving biodiversity?
Lian Pin Koh offers a solution based on simple economic principles. Grow the most productive crop to produce cereal, oil or protein, and grow each crop where it grows best. He presents an interactive world map to demonstrate that no extra land needs to be taken into production to feed 9 billion people.
The results are interesting. Strikingly, Brazil is doing just fine. Just a bit more of rice and Brazilian agriculture is optimal. Other countries need to change drastically. North European countries need to switch from barley to wheat. Canada and Russia need to abolish wheat agriculture and adopt maize. West Africa and the Cono Sur needs to grow more rice but northern South America and the US need to grow more maize. Yields will go up automagically, as each crop is planted on land that is more suited to it, fulfilling the dream of the 1980s land use evaluators.
The study is still in preparation and no background info on the methodology is available yet. Transport costs and climate change do not seem to be taken into account. “Optimal” seems to refer to yields per hectare, not to labour and inputs. Overall, the trend seems to be towards more high-yielding crops, which also require more inputs. With more people, more labour is available. But other inputs, like water, are limited.
Another question would be why crop use is sub-optimal now. Is it trade barriers? Cultural preferences and agricultural traditions? Or is it economics, really?
There is of course more to conservation than making agriculture more efficient. Another study shows that intensifying land use does not in fact put a break on crop land expansion. Additional measures would be needed to ensure that more efficiency indeed stops crops taking over non-agricultural land, and impacting biodiversity.
Even so, this is an interesting thought experiment. In an ideal world, swapping crops is enough to raise crop production some 30%. Feeding 9 billion people suddenly appears a bit easier.
1. Unless I am mistaken, the 9 billion will be reached (without me) by 2050.
2. We should beware simple economics, and even economics altogether. Lian Pin Koh’s suggestions do not take account, it would seem, of the simple agronomic principles, such as climate adaptation, the desirability of a crop rotation, or the topographical constraints of rice growing. France, it is suggested, should convert cereal production to maize… That would lead to a civil war because of excessive water consumption in summer. Algeria would have to convert 62 percent of its cereals area to rice? That is a joke. The DR Congo would go for wheat? That is another joke.
3. Feeding the 9 billion requires much more than optimising, without regard for agronomy, staple food production. The biggest challenge is perhaps to ensure that people can buy the food.
4. The other study, which supposedly “shows that intensifying land use does not in fact put a break on crop land expansion” also seems to be an ivory tower production, a very puzzling one on top of it.
5. “In developed countries, there was no evidence that higher staple crop yields were associated with decreases in per capita cropland area”? In the USA, the population increased by some 25 percent between 1979 and 1999. I am not aware that the cropland area increased during that period by the same order.
6. “This may be because high agricultural subsidies in developed countries override any land-sparing pattern that might otherwise occur”? Were is the evidence? In 1992, the European Union started to impose “set aside” to reduce the land effectively under production. Woodland has increased in France during the period…
7. The above examples may be anecdotal exceptions compared to the global study, but I really need to be convinced.
Good points all of them. A thought experiment always has something of a joke of course.
It would be cool to do a similar though experiment for water. Can crops be swapped to increase crop per drop and have more water for nature?
When I wrote this piece I saw Congo DR does actually produce some wheat according to the FAO statistics… I wonder what the methodology behind Lian Pin Koh´s study is, but since this is an intermediate product I guess that the most obvious mistakes will be solved.
Actually, the map doesn´t say that Algeria should convert 62% of its land to rice. The numbers refer to the total output, I think.
On the other hand, I think the authors of the land-sparing study talk about the marginal contribution of land use intensification on land sparing. The fact that the EU needed the set-aside policy in the first place and didn´t rely on just increasing production per hectare to increase woodland actually confirmis the point. Land use intensification in itself will not lead to more woodland. Combined with other measures it might, like it does in France.
An interesting experiment in an ideal world that also shares wealth and technology. Don’t think we’ll ever see that world.