Are We Close to Peak Farmland?

I want to believe. Who in conservation doesn’t? What if we really have hit peak farmland globally, as a new study by Rockefeller University researchers argues (Ausubel et al. 2013)? Considering that land conversion is the biggest threat globally to terrestrial biodiversity, hitting peak farmland would be better news for conservation than finding a flock of passenger pigeons living next to the Gates Foundation headquarters.

Yet I have several niggles about the study that make me click my heels together three times because I am not sure it’s real. Beef consumption, intensification of agriculture, and tropical deforestation make we wonder if the new study is right. As people become wealthier, they eat more meat. There are exceptions such as India, but it’s generally true. Fortunately, the global demand curve for meat flattens out after about US$10,000 in per capita income, but most countries are well below this level (Figure 1, below).

Producing a pound of meat takes a lot of other food. A broiler chicken gains 1 pound for every 1.7 pounds it eats, giving it a feed conversion ratio of 1.7:1. A pig has a 3.0:1 feed conversion ration, and a cow has a 10.4:1 ratio (Tolkamp et al. 2010). But these ratios are a bit misleading, because a 1-pound weight gain does not equal 1 pound of marketable meat. For a cow, it takes approximately 33 pounds of feed to produce 1 pound of marketable meat, because edible meat equals approximately 30-35% of a cow’s live weight ( Granted, cattle are often raised in rangelands that are unsuitable for agriculture, but that is rarely the case with other farmed animals, and 30% of the grain produced in 2011 was used as animal food (USDA).

Figure 1: Global Demand Curve for Meat Per Capita Income (as per UNEP GEAS using FAO and World Bank data).


The Peak Farmland co-authors note that beef consumption is a potential confounding wild card. Yet beef consumption is predictable once the income elasticity of demand is known. In China, for example, income elasticity of demand for beef is high — 1.56 — meaning that demand for beef increases 15.6% for each 10% increase in average income (Masuda & Goldsmith 2010). The average person in China currently consumes about 12% as much beef as the average American (FAOStats). If China’s GDP per capita grows at only half the rate of the last decade, beef consumption will still triple between 2010 and 2030 (Masuda & Goldsmith 2010). And we are doing our part: By the end of 2013, there will be 2,000 McDonalds restaurants in China (

China is not alone on the beef issue. Mexico and Indonesia are also projected to have increases in beef consumption. To borrow a line from University of Minnesota’s Jonathan Foley, for global agriculture, “the elephant in the room is not an elephant but a cow.” Thus, my first beef with the Peak Farmland study is that there is no beef in the study — i.e., no consideration of how growing beef consumption will impact farmland.

My second niggle is with the idea that intensification of agriculture on existing lands is sufficient to feed the 2060 world population. The co-authors make the point that the contest-winning farmers in Iowa produced 18 tones of corn per hectare in 2010 compared to the US average of 10 tones/ha and the global average of 5 tones/ha, and intensifying corn production on existing land could negate the need for new farmland. Yet if the 16 most important food and feed crops were brought to within 95% of their current potential yields, this would increase production by only about 58%, or about half the increase needed to meet the projected world food demand (Foley et al. 2011).

Moreover, Fargione et al. (2010) show that exponential extrapolations of yield increase as used in the study (e.g., 1.7% per year) result in poor predictions because trends in yield increases may be linear rather than exponential. There is also the problem that, in recent years, increases in yields for wheat and rice have stagnated in more than 30% of global crop areas (Ray et al. 2013). In short, it seems overly optimistic to assume that intensification of agriculture on existing lands can meet future food needs.

Finally, agricultural expansion in tropical forests is likely to continue to be one of the biggest global threats to biodiversity conservation whether we are past peak farmland or not. The forces driving the expansion of soybean and sugarcane areas in Brazil, palm oil in Indonesia, and small-scale agriculture in sub-Saharan Africa are likely to continue even if global farmland decreases in aggregate.

I very much hope the Peak Farmland co-authors are right, and we are over the hump. But the study gives me the feeling that we are not in Kansas anymore, but are in a different and better world where I would like to live but don’t yet.

(Image: Crops with Mt. Adams in the distance, Oregon. Source: Flickr user pfly via a Creative Commons license.)


Ausubel, J.H., I.K. Wernick, and P.E. Waggoner. 2013. “Peak Farmland and the Prospects for Sparing Nature,” In McNicoll, G., J. Bongaarts, and E.P. Churchill, eds., Population and Public Policy: Essays in Honor of Paul Demeny, supplement to Population and Development Review 38. New York: Population Council.

Fargione, J.E., R.J. Plevin, and J.D. Hill. 2010. The ecological impact of biofuels. Annual Review of Ecology, Evolution, and Systematics 41:351-377.

Foley, J.A., N. Ramankutty, K.A. Brauman, E.S. Cassidy, J.S. Gerber, M. Johnston, and D.P. Zaks. 2011. Solutions for a cultivated planet. Nature 478(7369):337-342.

Lopez, J.A., and J.E. Malaga. 2009. Forecast and simulation analysis of Mexican meat consumption at the table cut level: Impacts on US exports. In 2009 Annual Meeting, July 26-28, 2009, Milwaukee, WI (No. 51986). Agricultural and Applied Economics Association.

Masuda, T., and P.D. Goldsmith. 2010. China’s meat consumption: An income elasticity analysis and long-term projections. Poster prepared for presentation at the AAEA Annual Meeting, Denver, Colorado.

Ray, D.K., N. Ramankutty, N.D. Mueller, P.C. West, and J.A. Foley. 2012. Recent patterns of crop yield growth and stagnation. Nature Communications 3:1293.

Tolkamp, B., E. Wall, R. Roehe, J. Newbold, and K. Zaralis. 2010. Review of nutrient efficiency in different breeds of farm livestock. Scotland: SAC.

UNEP. 2012. Growing Greenhouse Gas Emissions Due to Meat Production.

USDA. 2012. Indonesia: Long-Term Prospects for U.S. Agricultural Exports. Washington DC: Foreign Agricultural Service.

If you believe in the work we’re doing, please lend a hand.


  1. There are two contradictory points that are the key issues here:

    1. That economically, alternatives to animal protein are utterly uncompetitive because of the high land costs of farming vegetable protein crops. In contrast, low-quality animal protein can be produced at a trifling costs from the cheap lands of Australia and sub-Saharan Africa. These have extremely erratic and arid climates and exceptionally old and infertile soils which are totally unsuitable for cropping. However, even with poor, rough pasturage, they can produce meat at a low cost even with very low yields.

    2. Historically, abundant animal protein was confined to the extremely rich and young soils of the upper (more poleward) Enriched World where these rich soils allowed for very large herbivore densities. However, these lands, despite their low fragility, never had the levels of population growth possible in the Tropical, Unenriched or even hotter and wetter parts of the Enriched World. There may be political elements in Dharmic vegetarianism because it allows more rapid population growth, but near-vegetarianism was universal among Tropical and Unenriched foragers due to the lack of meat.

Add a Comment