Asterisk: You argue that the nutritional gains of the Green Revolution have been uneven. Although overall calorie consumption has increased in some areas, dietary diversity has decreased, and micronutrient deficiencies and stunting still remain common. Can you describe this process, using India as an example?

Prabhu: When the Green Revolution began, India was facing massive hunger and starvation. The focus of the Green Revolution was a major boost to the calorie supply, and the new varieties of rice and wheat helped to do that. But the government was focused on just these two crops. All the infrastructure that was built, the policy environment, and the incentives for farmers were focused on these crops only. That resulted in the crowding out of more traditional crops that are higher in nutrition, such as millets and pulses. 

As the hunger problem was solved, incomes started to rise through small-farm productivity growth. But then we also found that although the demand for dietary diversity was rising, the supply wasn’t keeping up for foods like fruits, vegetables, pulses, and livestock products. The policy environment still focused on the big staple crops.

But all this happened in the regions where the Green Revolution was successful and incomes were rising. With rising income, people could trade with other parts of the country where other foods were grown, bringing some of that diversity into local markets. But if you were in the regions that are much poorer — the Green Revolution did not take root as strongly in the eastern part of India, for example — you did not see either an increase in the demand for diversity or an increased supply of diversity to markets. Those are the places that ended up with very high levels of malnutrition and child stunting, which continue to be persistent.

A: Can you say more about why there was better support for wheat and rice as opposed to millet or lentils and what forms that support took?

P: Beginning in the ’60s, there was a search for crops that could see a rapid increase in overall yields. This was an international effort involving large experiments that focused on very particular crops. The U.S. and Europe had a history of research on wheat, and yields were already rising as a result of that crop-improvement research. The challenge was to make crop-improvement research applicable to the subtropical conditions, such as the Indo-Gangetic plains in India. That was the work that made Norman Borlaug very famous, resulting in the Nobel Prize in 1970. Wheat materials developed in Mexico were then brought to India, China, and many other countries. 

It was a similar story with rice. Japan, Korea, and other Asian countries had a history of rice-improvement research, but these were all japonica-type varieties. The challenge was to bring that crop improvement into rice varieties grown in the tropics, such as the indica varieties. That was done at the International Rice Research Institute in the Philippines and resulted in high-yielding rice varieties, which then very rapidly spread across Asia. 

Why wasn’t the same thing done for millet, sorghum, or other traditional crops? None of these crops were grown in the West. There was no history of research on them, and the history of agriculture research played a major role in which crops were chosen. 

Today we are now saying we need higher-yielding varieties of millets, pulses, et cetera. There’s quite a bit of effort happening on this front, including at ICRISAT (the International Crops Research Institute for the Semi-Arid Tropics) in India. But the improvements haven’t been as dramatic as we’ve seen for rice and wheat. Rice and wheat productivity went from one (metric) tonne per hectare to four, five, even six times that. In the case of millet, we’re still talking about going from a tonne to maybe two. 

A: Why did Green Revolution crops fail to be as successful in Africa as compared with Latin America or in Asia?

P: It’s important to separate out the crops themselves. Wheat can be grown fairly well in areas like Ethiopia, Zimbabwe, or South Africa, and these countries have seen widespread adoption of modern wheat varieties. But much of what’s being produced goes to urban markets and urban populations. High-yielding varieties of rice are also now slowly coming to West Africa, in swampy areas especially. 

But maize has been more disappointing and there are many reasons for that. When Asia was undergoing the Green Revolution in the ’60s and ’70s, land was scarce and population densities were high. So the emphasis there was on how to increase yields on existing lands. Africa did not have that same land scarcity issue. At that time, it was still very sparsely populated, with large land areas available for expansion. As a result, much of the donor focus was on increasing labor productivity rather than crop productivity. So instead of increasing productivity per unit of land, funding emphasis was on mechanization and ways to expand land area under cultivation. At that time, it seemed like a better strategy. 

The problem is that if you look at Africa today, population densities in many areas are very similar to those in Asia in the ’70s and ’80s. Moreover, Africa is growing much more rapidly than Asia is growing now. So one would anticipate that right now you should see more intensification and adoption of higher-yielding varieties.

That’s happening to some extent, but not as rapidly as one would expect. One reason for the slow progress is that infrastructure investments have been very poor in Africa relative to other regions. The second is that after the formation of the World Trade Organization in 1995, international trade in food became very prevalent. In many of the big population centers in Africa, it’s often cheaper to import food from other countries than it is to grow domestically. The terms of trade have made it very hard to create a competitive supply system from within the continent. 

A: Could you expand on how intensive agriculture became a stepping stone for further development?

P: Much of Asian agriculture has traditionally been smallholder, where farmers own one or two hectares and traditionally grow one crop a year. One crop of rice or millets traditionally yielded about a tonne per hectare. After a farmer keeps what is needed for the family, there’s very little to sell. The margin is small and in many years — because of weather or climatic issues — farmers would end up with less than one tonne and have a food deficit as a result.

So the opportunities for agriculture as a growth sector, to lead to increased incomes, were limited. As new crop varieties and new technologies came in, farmers could grow four or five times what they’d been able to in the past. After you kept what was necessary for your own needs, you still had a significant amount to sell. 

Another factor was that these crop varieties had a much shorter growing period compared with the older traditional varieties. So you could grow two or three crops on the same piece of land. That then created an enormous boost in productivity, which then boosted market surplus. As that happened, it generated income that could be invested in better housing, purchase of consumer goods, education, et cetera. In addition, productivity increased land values of small farms, so farmers sold their land and invested it outside of agriculture. And that triggered an overall economic growth process. 

A: And places like Africa that are importing grain don’t have that same beneficial flywheel effect.

P: Exactly. In the hinterlands of sub-Saharan Africa, farmers continue to eke out a tonne per hectare. So the opportunity for growth there is very limited.

A: You mentioned earlier that demand for dietary diversity is increasing, but supply of diverse foods has not kept up with demand. Why? 

P: As incomes rise, the demand for diversity in your food consumption increases also. Very poor households will consume large amounts of basic, starchy staples and very little of vegetables, fruit, livestock products, et cetera. As incomes rise, the share of the staples in your overall diet begins to decline. In agricultural economics, this is referred to as Bennet’s law. 

But the supply hasn’t matched that demand. And one of the reasons I think that’s the case is that policy tends to be very sticky. Policy hasn’t been nimble enough to redirect itself as the demand conditions have changed. Even today, for instance, much of the food and agriculture policy in India is focused on the big staples. There is some lip service to promoting diversity, but the essentials remain the same: price supports, procurement of grain, subsidies for inputs, and subsidized credit. 

We don’t have a level playing field for other crops. A farmer can’t say, for example, “I’m getting a better price for onions in the market, so maybe I should switch from rice to onions.” Onion production doesn’t have the support system that rice production does, so onion production tends to be much riskier. In addition, we don’t have infrastructure built around non-staples. Combine that with a lack of incentives, and the switch to non-staple crops has been very limited. 

One last point here is that the people who have been most successful in growing staple crops have also benefited the most from the supports. They’re not going to give those supports up easily. So there’s a big political economy issue here. How do you make the transition for them as well as making the transition for the food system itself?

A: You’ve called this “crop-neutral agricultural policy.” Have you seen any movement away from strong support for staple grains toward nutrition-sensitive agriculture on a policy level, or is policy still too sticky? 

P: There’s a lot more talk about nutrition-sensitive agriculture and a lot more pronouncements about why this is important. However, most governments see this as an add-on, not a substitution. Rather than removing the existing supports or reducing the existing supports for staples, governments have just added supports for other crops. That creates some marginal improvement for some of the other crops, but your fundamentals don’t change. The crop-neutrality argument says: Treat all these crops on a level playing field and let market signals determine the supply responses.

A: One barrier here is that access to markets is often very poor in a lot of the developing world. What kinds of policy interventions and private investments do you think are important to make sure markets are well-developed for diverse foods? 

P: This is where the private sector needs to be playing a much bigger role than it does today. When we talk about nutrition-sensitive crops, we’re talking about perishable crops and perishable products. You need cold-storage systems, quality controls, safety controls, et cetera. These require massive investments from the private sector. And encouraging the private sector to take on these roles is really important. 

Many governments have been shy about doing that because much of the historical procurement policies and support policies have been led by governments or government parastatals rather than by private sector companies. But unless you make that transition, you’re not going to see that supply response coming in. 

A: What do you see as the role for bio-fortified crops, which tend to receive a lot of interest? Something like orange-fleshed sweet potatoes can marginally increase vitamin A status for women and children. But other bio-fortified crops — for example golden rice — have received a lot of pushback. What are the benefits and what are the drawbacks of bio-fortifcation, compared with targeting dietary diversification for nutrition outcomes?

P: When we think about bio crops, we have to be very careful to say we’re not talking about GMOs, because a large proportion of bio-fortified crops are done through traditional breeding, not through genetic modification. Golden rice, of course, is a GM crop, and that’s where it received pushback, particularly from civil-society groups. 

But for the rest — orange-fleshed sweet potato for vitamin A, zinc-enhanced wheat or rice, et cetera. — there’s been a lot of research and public investment to try to promote these crops, particularly in Africa. Orange-fleshed sweet potato has made some inroads. But outside of Africa it’s been very limited. There’s been little adoption of most bio-fortified crops. They’re available but farmers don’t choose them. 

I think one big reason for the lack of adoption is that there’s no way to physically differentiate a bio-fortified crop from a non-bio crop. They look the same. The color is the same. And once they’re out in the field, they all get mixed up and it’s hard to tell them apart. As a result, there’s very little price difference. 

The incentives, therefore, in investing in bio-fortified crops are limited for farmers. And the fact that it’s more nutritious is not a particularly persuasive argument for them. Getting the same nutrients from other sources, such as fortified products, is often easier. 

A: We’ve been talking about crops specifically, but animal-sourced foods are one of the best sources of a lot of key nutrients. How do you think about the trade-offs between animal welfare and nutrition outcomes?

P: From a food-policy point of view, there’s been a lot more attention to livestock issues than, say, millets and vegetables. Much more attention is being given to improving livestock productivity, and there have been significant improvements. 

Milk and dairy production has gone up dramatically over the last few decades in Asia, although less so in Africa. Access to nutrients from those products is improving significantly. 

Meat has been a challenge for many reasons. One, there are cultural issues around meat consumption, and therefore supply of meat has been limited. But the other factor has been that overall infrastructure around meat production, including meat marketing, has been limited. Here again, we come back to perishable-food issues — cold storage, et cetera. — so we haven’t seen supply rise to meet demand. 

A: Obtaining accurate information on nutrition and agriculture seems really hard. Even in high-income contexts, getting good food-consumption data is a pretty tricky problem. I’m curious what the biggest challenges are in your research and how they impact the kinds of conclusions you can draw about nutrition. 

P: Data is a big challenge for all of us. In my public lectures, I typically spend at least 10 minutes complaining about lack of data, especially when there are policy makers in the audience. I keep telling them that they have to do something about this. 

Many countries have these periodic surveys, which are nationally representative surveys that look at income and consumption and try to track what’s happening. Many of them are done on a recall basis and the time periods between surveys can be as much as 10 years or more. 

Recently, other groups have established their own surveys. One of them is the Living Standards Measurement Study of the World Bank, part of which is the Integrated Agricultural Assessment. That survey attempts to capture some of the consumption data, which is helpful. But again, it’s periodic — every five years or so, which makes it difficult to track what’s happening over time. In India there’s what’s known as CMIE data, which tracks monthly expenditures across food groups for 60,000 households around the country. So these expenditure surveys are giving us some more real-time tracking ability to look at what’s happening with food consumption. 

A lot of people use dietary diversity scores, which ask how many times you’ve had this food or that food. I find those are shaky and not very useful, but that’s the information that there is on the nutrition-status side.

A: You’ve written about the double burden of malnutrition — high undernutrition, high numbers of the overweight and obese. In India, rates of stunting are still more than 30 percent. But overweight and obesity numbers are inching up to 20 or 30 percent as well. 

In general, we don’t have much evidence of countries that have been able to eliminate undernutrition at the same time that they minimize overnutrition. How are those problems connected? And if there are countries that have evidence of maintaining a stable ground, what are the characteristics of those?

P: It all comes back to dietary diversity. What happens is that countries that have been successful in increasing staple-grain productivity end up with large volumes of staple grains. These grains are relatively cheap, which leads to surplus being used by the processed-food industry, which leads in turn to a drop in the relative price of processed foods. The other issue is oils, sugars, and sugary beverages. It’s incredible that for the same countries where we talk about access to good food, there’s no poor access to bad food. Bad-food access is easy, even in very remote areas.

Countries that have been successful in managing the nutrition transition are countries that very quickly shifted from staples to a more diversified food basket. Thailand is one that switched very early in the process. But Thailand was also historically a rice-exporting country, with much lower population densities compared with some of the other Asian countries. So they were able to make that switch much faster than others.

A: What policies did Thailand adopt that let them handle that transition successfully? And is that a model that other countries in the region can emulate, or are there factors specific to Thailand?

P: I haven’t spent enough time looking at Thailand recently, but my immediate guess is that in addition to being a rice-exporting country, Thailand also emerged as a major exporter of fruits, vegetables, cassava — a whole variety of crops. So the government policy then kind of opened up the broader food market for exports. And I think that played a big role in the transition that took place in Thailand. 

Thailand was also able to integrate its food market into the global market early. They invested in infrastructure around it — roads, transport, power supply, irrigation. And they also had a lot of private-sector investments. 

A: I’m curious about the kinds of interventions you think are effective at addressing some of these issues. 

P: Behavioral change plays a big role. Just getting consumers to demand better quality food and more diversity of food plays a big role. Even middle-class populations haven’t been as quick to change their diets and consumption practices as much as economic theory suggests they should be. Behavioral-change campaigns have been a big part of the nutrition work that’s happening. But these are not easy campaigns. They can’t be done at scale and they tend to be expensive. 

Now, there are few models. One is the Got Milk? campaign in the U.S. I think that had a big impact on U.S. consumer behavior relative to some of the other campaigns around meat consumption and reducing meat consumption. 

So these broad messages at scale through media and through advertising can have an impact, but I don’t think the nutrition community has really used these tools to create behavioral change in developing countries. That’s something I think is possible. 

A: And what kinds of agricultural research would you really like to see more of?

P: I think a lot more attention needs to be given to commodities that have higher nutrition value: millets, sorghum, pulses, lentils, et cetera. These are crops that are so essential for nutrition, but the productivity levels are still very low. And many of them are grown in high-temperature environments, so they could also be very useful in terms of climate resilience. 

Livestock is another area we should be looking at more. There’s a strong negative reaction to livestock research in the West, but we have to be very careful not to throw out all livestock research because there are significant climate impacts. We need to ask how we make livestock-production systems more climate-friendly and more climate resilient rather than scrap the idea altogether. 

Finally, I certainly don’t imply through my book that we should stop working on rice and wheat and maize. There still needs to be a lot of work in those areas to make them more efficient and more adapted to climate change.