How many crops feed the world?

A guest post for @plantvillage by Dr Colin Khoury, Visiting Research Scientist | International Center for Tropical Agriculture (CIAT) 


(Banner image by Neil Palmer at CIAT)

Recently I was involved in a study that documented dietary change around the world, concluding not-so-surprisingly that we are all eating more and more of the same foods. Most of us have noticed in one way or another these changes during travels- finding pizza in Tokyo, or McDonald’s in Rome- or simply by reading ingredient labels in the supermarket and realizing that derivatives of maize, wheat, and soy are in just about every processed food on the shelves. For many of us there is something reassuring in recognizing these brands, food products and specific crop plants we know so well. But there is an equal feeling of disquiet- is this standardization and globalization of food across the world good for human health? Is it good for planetary health?

The original inspiration for our research dates back a few years. At that time, I was a scientist at the Global Crop Diversity Trust, a foundation dedicated to the conservation of crop diversity for the purpose of long-term global food security. In working to convince donors and the public of the value of crop diversity in agriculture, I noticed that just about every article written about current or future food production included a key sentence that said something like: “only X number of crops feeds the world”. For a few examples, see here, and here, and here. The implication behind this ubiquitous sentence about humanity’s dependence upon a very short list of plants for our survival seemed to me to be that we are interconnected, and at the same time that we are fragile. Perhaps, on a deeper level, it is not so different than the cultural message so often translated via science fiction films (like this and this) and literature- that our own ingenuity and hubris might in the end be our downfall…unless of course we have a hero to save us at the last minute.

It was peculiar to me that each sentence varied somewhat in the actual number of crops listed that feed the world. I wanted an answer- was it 3, or 12, or 30? Most of these statements source back to production or consumption data compiled by the FAO across the world’s countries, and then aggregated into a global average. While searching for an answer to this seemingly simple question I came across a couple of researchers who had not been satisfied with a boiled-down (globally aggregated) answer. One paragraph of their discussion convinced me:

“Crops such as fonio, Digitaria exilis Stapf, and quinoa, Chenopodium quinoa Willd., are lost in global production data; but to conclude that they are unimportant is to conclude that the people of Guinea, Gambia, and Bolivia who rely on them are unimportant.”

Prescott-Allen R and Prescott-Allen C (1990) How many plants feed the world? Conservation Biology 4(4):365-374

Instead, these researchers used national level food supply data from the FAO, which represents the most disaggregated data available for comparing consumption around the world. They measured the importance of different crops within diets within each country around the world, and for all available measurements- not only calories, but also protein, fat, and food weight. Published in 1990 but using a dataset from 1979-1981, the researchers were able to show that a considerably longer list of crops (over 100) are quite important in contribution to calories or protein or fat or food weight in at least one country around the world. I found this answer compelling and yet not fully satisfying, and it raised even more questions, not the least because their answer was already 30 years old.

A second and equally ubiquitous assertion appearing in just about any article about crop diversity is some variation on “the world has lost xx% of the agricultural diversity we once had”. Again the precise number varies (although it is often ‘75%’). Most of these quotes source back to few studies also by FAO that measured the loss in named crop varieties, for example of vegetables in the U.S, or in production area of different varieties in a few countries. It turns out that the question of just how much crop diversity has been lost in modern times is a very complicated and highly contested one, with strong opinions all around. Researchers have done some work to look more deeply at changes in diversity in crop varieties, and in the genetic diversity within those varieties. Surprisingly, though, little work has been done to understand changes in diversity of crop species- maize versus rice versus fonio versus quinoa and so on.

I wondered if it was possible to ascertain if the importance of different crops had changed in the 30 years since the ‘how many plants feed the world’ article’s data was compiled, that could help us both to give an updated answer to the question of how many crops feed the world, and at the same time to assess the amount of loss of crop species diversity globally.

I quickly realized in analyzing current FAO data that it wasn’t going to be possible to compare to the previous research. On one hand, the way that crops were reported had changed. On the other, the countries in the world had changed! Fortunately, though, FAO provided data in the current format for every year dating back to the inception of consumption data compilation. So we crunched the numbers from the beginning (1961) to the present, standardizing the crops and the countries across the years, to provide a picture of change in food supplies around the world over the last 50 years.

One of the surprises that quickly emerged from the results was that we could not see any complete losses of food crop species from diets over the past 50 years. That is, we saw no extinction. This is due in large part to the degree of comprehensiveness of FAO food supply data- only 52 crop specific commodities are able to be analyzed, and a number of these are group commodities that include in one name (e.g. ‘cereals, other’) a bunch of different crops, including for instance both fonio and quinoa. Due to this lack of specificity, the data was not sufficient to show trends in the local crops that are most vulnerable to changes in the global food system, but we do know from targeted local studies that many of these crops are in decline.

Wheat Pic by Neil Palmer (CIAT)

A second surprise was that over the last 50 years, many national diets had actually gotten more diverse, not less. Again this finding requires some qualification- due to the generality of crops reported in the data, we were not able to assess overall complete diversity in diets (that is, every last food plant that people are eating). What we were able to do was assess change in the globally and regionally significant crops that are reported in the data, and we found that in fact diets that were primarily based upon particular staples, for instance rice in East Asia, had diversified to include other staples such as wheat and potatoes.

Due to the diversification of national diets in regard to these big crops around the world, we were able to show that diets have become much more similar worldwide, with increasing amounts of global staple crops (in particular, staple cereals and oil crops) in the diet, and decreasing amounts of regionally and traditionally important food crops. A key paragraph from our article:

The total number of important crop species we identified remained relatively consistent in comparison with a previous point estimate based on national-level data from 1979 to 1981 (Prescott-Allen and Prescott-Allen, 1990), but the spread and abundance values of these crops have changed measurably. The rate of movement toward homogeneity in food supply compositions globally continues with no indication of slowing. This trend implies a likely deterioration in importance of unreported minor and geographically restricted food plants, along with the measured cereal, oil, starchy root, and other crops that displayed significant declines in abundance in national food supplies. Thus, even as the number of measured crops available to the consumer in a given country has increased over the past half-century as a global trend, the total diversity of crops contributing significantly worldwide has narrowed.

To find out the current answer to the question of how many crops feed the world, using methods that parallel the 1979-1981 study, take a look at the supporting information. Interestingly, the homogenization of global food supplies means that a globally aggregated average is now probably a much truer answer to the question than it was 50 years ago. While it is still much too simple to simply say that wheat, rice, sugar, maize, and a few other crops feed the world, the statement is more correct now than it ever has been in the past.

In the end, of course, an exact answer to the question of how many crops feed the world is not that important. There are many ways to answer the question, and varying thresholds for what we might consider significant versus insubstantial. What is important is the direction we are heading globally, which brings us around to the question as to whether we can validate the ubiquitous statement about the loss of crop diversity globally, on the species level. This we know- global food supplies are becoming increasingly composed worldwide of a pretty short list of exceedingly important staple food crops and their derivatives, and alternative crop species are becoming further marginalized. We need to develop much better consumption (and production) information systems worldwide, though, in order to fully track changes in threatened crop species.

What are the implications of increasing homogeneity in global food supplies? We think there are many, both on the agricultural production side, and the human health side. The impacts of the drivers of dietary change are complex- with both positive outcomes (e.g. extremely productive major crop systems and reduced hunger) and significant concerns (e.g. increase in monocultures leading to greater vulnerability to pests and diseases, and greater susceptibility to the negative effects of the nutrition transition, including heart disease, diabetes, and other diet-related diseases.

Given the concerns raised by the homogenization of global food supplies, and the deep complexities of the underlying drivers of these changes, what is the potential for change in food systems in a direction that is better both for our agroecosystems and our health? One final surprise in the data- we are already seeing some national food supplies beginning to move in somewhat different directions- less animal products and other energy dense foods, and more diverse cereals and vegetables in particular. Change is still occurring, and the future does not appear to be fixed. Who is the hero that might just save us from the dangers of our own ingenuity? In my opinion, much needs to be done on all levels- by crop scientist’s choices in the crops they work on, by policy makers in the research they fund and the laws they enact (and here) that impact the price and availability of different foods, and by consumers’ choices in what they eat and how they eat it.