Increasing the Resilience of African Smallholder Farmers

[Dr. Agnes Kalibata] We're interested in overall ability for Africa to feed itself. And smallholder farmers are a huge part. They are 70% of the people that are growing Africa's food. So Africa will not feed itself unless the smallholder farmer is actually able to produce and produce enough, not just feed himself, but as a business

[JB Wogan] I'm JB Wogan from Mathematica and welcome back to On the Evidence.

[ Music ]

Our guest for this episode is Dr. Agnes Kalibata, who recently wrapped up her 10-year term as president of the Alliance for a Green Revolution in Africa, better known as AGRA. AGRA is an African-led organization, which seeks to create an environment where Africa sustainably feeds itself. It does so by scaling agricultural innovations that help smallholder farmers achieve increased incomes, better livelihoods, and improved food security.

Mathematica partnered with AGRA to evaluate the Alliance's strategy to increase the yields and incomes of 30 million smallholder farming households in 11 African countries, as well as co-developing a monitoring, evaluation, and learning strategy to help those farmers strengthen their crops' resilience to flooding and drought with the intent to ultimately boost income, food security, and job opportunities. Mathematica's president and chief executive officer, Paul Decker, interviewed Dr. Kalibata for On the Evidence during a visit to Nairobi in December of 2024.

Dr. Kalibata has a PhD in entomology, which is the study of insects and their relationship to humans, the environment, and other organisms. She was the minister of agriculture and animal resources in Rwanda for six years before becoming AGRA's president in 2014. In March of 2025, she stepped down as president, having completed her 10-year term. Earlier in the year, we released an episode featuring Paul's interview with Jesse Moore, the chief executive officer of M-KOPA, a company that provides financial services to low-income and unbanked customers in East Africa.

If you're new to Mathematica's On the Evidence podcast, please consider subscribing. More information on how to subscribe on your podcasting app of choice is available at mathematica.org/OnTheEvidence. Once you're there, you'll also find a blog about this episode, which will include a summary, a few key quotes, the full transcript, a video version of the episode, and additional resources, such as Mathematica's evaluation findings about AGRA. I hope you enjoy the conversation.

[Paul Decker] Welcome, Dr. Kalibata, to the On the Evidence podcast. It's great to have you here today.

[Dr. Agnes Kalibata] Thank you for having me.

[Paul Decker] Your early career was spent as a researcher, and as a fellow researcher, I'm curious how that experience influenced your perspective on agricultural systems.

[Dr. Agnes Kalibata] My early research and working research influenced how I think about agricultural food systems, and it comes from three critical places. One, the fact that you learn very early on that farmers are businesses. That's all the business they do. They're entrepreneurs. And if everything we could do would be in service of recognizing that they're entrepreneurs that are looking just like any business out there to be successful, that's extremely important. It's really, really, really critical. So that's number one.

Number two, the fact that farmers understand their environment extremely well. They make decisions based on the biodiversity that exists on their farms. So farmers plant certain varieties because of the value those varieties bring. So maybe this variety has a good taste. Maybe this variety is used for cultural thing X or Y. Maybe this variety is better appreciated, is appreciated extremely well when shared by the community in marriage ceremonies.

So just understanding and recognizing that they have a very important understanding of what the environment is offering in terms of the different varieties within a specific crop. A good example is banana. Within banana, there are so many varieties that farmers look to, and those varieties represent different things. They are drought-tolerant. They are culturally important. Their taste is different. They have less sap. Things that, as a scientist, you would never know until you speak to a farmer.

And there's a third one. The third part of working with farmers that you get to appreciate is probably what I did when I was in the Ministry of Agriculture. The fact that you don't know what you don't know. So being out there and allowing yourself to do the work and then looking at what the data is telling you and being able to do things based on not what you feel or what you think you've seen, but on what the data is saying is a very important part of my growing up as a scientist and a very important part of how we get to do work. But I don't think, from a policymaker's perspective, usually driven by the needs to get stuff done, I don't think we appreciate enough how much building a work on evidence of what is working is important. And again, you don't know until you see the data.

[Paul Decker] Right, right. So it's combining the data with understanding the context in which the farmers are operating.

[Dr. Agnes Kalibata] 100%. Yeah.

[Paul Decker] Yeah. What about for what you're doing now? Were there any key experiences, data, or insights during your research career that shaped the leadership approach that you take today?

[Dr. Agnes Kalibata] I mean, the one I think that is most important in terms of shaping our dependence on data, again, is being able to know that whatever we are doing, it is anchored on the fact that the evidence coming out of the farm is telling you what works and what doesn't work. I mean, that's really purely how we work from a scientific perspective. But given the work we've been doing here at AGRA, when I look at some of the things where evidence and data have been extremely useful, I keep, for example, one of the questions I was asking is, why do we have so much seed and why is nobody taking it up?

In the past, we were saying, okay, if only farmers had seed, but now we have seed and we have ways of delivering seed and farmers don't take it up. And you sit all the time trying to push the same thing. But the moment you start breaking the system down and trying to understand within the system, what part of the system is preventing the uptake, and again, all based on how you generate evidence. So you look at different parts of the system and then you start looking for the weakest link. Again, data, going back to data.

Then you use that information to make decisions as to what is actually preventing this from going forward. Because initially we thought that was technologies. Oh, if they had the right seed, that's the technology. If only farmers had the right seed. If they had enough choice of what to grow, meaning different types of seed, but even when they have them, they don't use them. Then you do, when you break down the system, you do realize, oh, there's actually something that is extremely simple. Once the researchers have made the seed, the journey from the seed to the farm is so long and it's not visible to every man.

It's only visible to scientists and sometimes businesses, but it's such a long journey and it breaks multiple times. And it becomes very difficult for the farmer to access the seed. So until you've done that, you really don't know what is going on. So we've done that very well. We mapped out the ecosystem and we've in fact put in place an institution to help with that journey, to help every government that we work with understand the journey and where the journey breaks and how to invest against those weakest points in that journey. So that's number one.

Number two, I think the other part we've done well that we weren't doing well enough is also recognizing that resources to transform a nation, to transform a people are huge and we can't do it as AGRA. We can only understand different pieces. We can only set the tone. We can only understand some of the, what you're calling the models that can be deployed, just like I just told you, that one. We need to partner and we need to build the right incentives and the right interest with governments to be able to do that work. And I do feel like, again, with an institution that can help them understand what's broken, help them

provide the technical capability when they need it, we've done a good job there as well. So I do feel that we've worked with governments to raise consciousness and appetite enough for them to know that they actually have a job to do here and if they need help, we do have the capability to help them. So those two pieces, I think, have been extremely critical.

[Paul Decker] Your academic experience with a PhD in entomology is distinctive. How is it that you chose that as your field?

[Dr. Agnes Kalibata] So it's very interesting. I was in the agricultural sector, not by accident. When I grew up, I realized, I decided I wanted to do agriculture because it made a lot of sense. It made a lot of sense to be in an area that affects my community, my family, and like I said, for these people that live in rural areas, agriculture is a business. So for me, choosing agriculture was choosing an area where I felt I could impact how my people do their business. So that was number one. Then you do realize that even in agriculture, there are so many things. Entomology presented the three areas that I was particularly interested in.

You could deal with pest problems for crops. So farmers do not optimize their yields because of pest problems. As kids in school, we grew up, you go to school, they serve you beans and the top layer is insects. We are used to, if you go to a boarding school in any of these schools around here, you see that. Those are pests in actual sense. They eat the hell out of the beans. But that's what happens. So how do you manage that? So that was one. Two, genetically, there are solutions in research. Once you do research, right, there are genetic improvements that can be done to make crops less damaged by whatever.

Whether you start with the conventional breeding, we have varieties that are more tolerant to pests and diseases because we're able to do breeding and influence that capability. And you can go all the way to all sorts of commodities, but that capability. So understanding entomology as a basis to understand how you can improve crop availability is important. And the third one is it affects -- it's not just crops, it affects livestock. So again, in my part of the world, farmers struggle with diseases from tsetse flies, from things like that.

So again, if there was an area where you felt like I would be able to impact, if I can impact how farmers access yields better, and entomology sounded like that place that brings all these three areas from pest management to improving varieties to dealing with animal diseases in one place. So it was that interface that seemed, in addition to the fact that it also affects human being. I grew up human beings, I grew up suffering from malaria. So for me, entomology was important.

[Paul Decker] And so your interest in agriculture came first to some extent, and then you recognize the importance of the intersection between agriculture and insects.

[Dr. Agnes Kalibata] I mean, it's very difficult to be interested in agriculture and not appreciate the constraints, the major constraints of the agricultural sector and insects, what we call pests and diseases are one of the biggest constraints to why farmers don't get what they're supposed to get.

[Paul Decker] Yeah.

[Dr. Agnes Kalibata] The post-harvest losses and post-harvest damage to crops is huge, but also in situ, in the farm, damage is huge. And again, it's not just because that's a problem, it's because that problem has a scientific solution. It can be influenced and can be helped through scientific solutions.

[Paul Decker] Right. Given this experience that you're describing, can you highlight a moment when your fundamental understanding of insects had what you might consider a transformative impact on agricultural innovation?

[Dr. Agnes Kalibata] I think probably the one place where I feel like -- I mean, there are several places. And if I look at agriculture from a bigger angle, entomology, first of all, has so many areas of interest. Once you get in at a PhD level, like I was, it goes beyond agriculture. So it becomes really a huge area. Just so you know, when I was doing it as agriculture, I was interested in forensic entomology. It wasn't so much about agriculture entomology. I had already done agriculture entomology at MSC level. And at PhD level, I was now interested in forensic entomology, where you use entomology or insects to solve problems, other problems in the world, criminal cases and stuff like that.

So that's a story for another day. But from where entomology impacts the agricultural sector, I think the two most important areas are the fact that you could build, again, capability of a crop to actually overcome insect damage. I thought that was huge, right? Again, based on what I had grown up seeing. Number two, that you could actually prevent insect attack to crops. So every yield that we get from improved genetics, you almost lose for some commodities, especially grains, in the tropics, you lose in damage to harvest, through insect damage to harvest.

So those are two areas that make it important for us to think about the place and role of entomology. The fact that the genetic gain you're getting on the other side, you're actually using in post-harvest. So either way, you have to manage the insect problem. And that made it so important for me.

[Paul Decker] Switching over to the organization that you lead now, can you provide an overview of AGRA's key objectives and how they address the unique challenges that are faced by African food systems?

[Dr. Agnes Kalibata] AGRA's objectives are huge, but primarily we focus on improving the livelihood of the smallholder farmer. So that's our primary objective. And I can relate to that. Dealing with the challenge of poverty at the farm level is how you improve the livelihood of a smallholder farmer. Dealing with agriculture, ensuring that agriculture is productive is how you address the challenge of poverty at the smallholder farmer. So that's number one. We're interested in overall ability for Africa to feed itself. And smallholder farmers are a huge part.

They are 70% of the people that are growing Africa's food. So Africa will not feed itself unless the smallholder farmer is actually able to produce and produce enough, not just feed himself, but as a business, like I started. So that's a very critical part. They need an ecosystem that works. They are small. Because they are small, they don't own tractors. Because they are small, they can't buy hundreds of tons of fertilizers or hundreds of tons of seeds like you would see a big business do. No, they are small. They probably need one bag or two bags of the right thing. So they have to have an ecosystem that works or a system that works.

A system to deliver good quality seed of their choice. A system to deliver fertilizers of their choice. And a system that delivers mechanization. Those things, thinking about it, remember there are hundreds of them that are making choices at the same time. And to be able to satisfy that, you have to have an ecosystem that works. Or you have to have a system that works. So AGRA focuses on ensuring that that system that can deliver for farmers, irrespective of their choice, is available.

That we can actually create a business ecosystem that allows businesses to build, that allows people to build businesses, micro and small businesses, that can deliver value for the farmer while delivering value to the businesses themselves. So you might be retailing seed. So we have what we call AGRA dealers. These are small businesses in the villages. Sometimes $1,000 type of businesses. But they retail seeds of vegetables. They retail seeds of maize. They retail seeds of beans. They retail pesticides, where farmers need to use pesticides.

That's the service ecosystem I'm talking about. You have other situations where young people have figured out to offer farmers tractors at a fee, a service ecosystem for tractors. One of the biggest mistakes of earlier years of farming in Africa is where people thought that, oh, John Deere is going to drive in here and farmers will buy John Deere tractors. No, it's not going to happen. They need a service. They don't need a tractor. However small, even if they made it to the smallest size possible, these farmers don't need that type of tractor.

But they are willing to pay for a service of a tractor so that their farms can get farmed. So I think it took us forever to figure out that farmers need a service ecosystem. They cannot be treated as farming is done in the rest of the world. So AGRA's job is to really break it down to the farm and community level and bring the service ecosystem for farmers, near farmers, so that they can have choices of seeds, of fertilizers, of mechanization, of irrigation, of even shellers. If I want to shell my produce, which is two bags, I shouldn't be required to go and buy a sheller.

I should just call someone and say, I have two bags to shell. Can you come and shell them for me? Near every farm on the African continent now owns a telephone. So bringing those services alongside that telephone ecosystem is what we work to do. To be able to do that, we need to work with private sector because that's a system that can only be sustained by private sector. SMEs, even very, very small businesses.

We also need to work with governments because SMEs work in an environment where government is providing a suitable environment for them to work, where it recognizes that policies that make it easy for them to work. So really our job is to work through that ecosystem. I feel like my job is about connecting the dots. The farmers need this. We find it, but we also find the people that deliver that as a business. And then we ensure that the ecosystem is working for them. So really just ensuring that the ecosystem is in place for farmers to do farming, smallholder farmers to do farming as a business is what we do.

[Paul Decker] So part of what the organization is doing is managing what are the natural economies of scale that work against the interests of small farmers and to change the market enough to provide more opportunity for the small farmers.

[Dr. Agnes Kalibata] Basically, build value around the smallholder farmer. When I was working as a minister in Rwanda, which is the smallest you can get when it comes to smallholder farmers, scale was extremely important. And the way we went about that, we created a system of aggregation, all the way from the land. Because you would think about aggregation from a crop perspective. No, we even aggregated land because land was too small.

So if we were buying, if we are trying to encourage a business to look at farmers as a source of potatoes because you're processing potatoes, you will not buy one, two, three. No, you want scale. So even aggregating farmers to ensure that they bring scale, farming scale, number one. Number two, they bring business scale to any business. So if you're selling inputs to farmers, you have scale. If you're selling tractor services to farmers, you have scale. So it's that ability to mobilize farmers, like you've said, to build economies of scale in an environment which basically individually has no scale.

[Paul Decker] In working with your partners like Mathematica in generating insights based on the data, are there any examples you can cite that drove changes in AGRA's strategy or its priorities?

[Dr. Agnes Kalibata] There are always, every time we come to an institution like Mathematica, we are looking for what the data is telling us, right? We are looking for what did we miss? We are looking for what is already happening and we actually assumed it wasn't happening. So I think for me, it's not so much the surprise because in some of these areas, there are very little surprises. But it is really, really important to have information that validates some of the gaps that we know exist, right? So for example, when you see all development funding going to one area and you know, okay, so we are all rushing to this area but this area can never move without that area. And it doesn't matter how much you speak.

It doesn't matter how much you say. People have already made their mind that if it is funding ABCD, we'll fund ABCD. But then the real problem is sitting with Y. Until we all start funding Y, we will not unlock that problem. So I do think the opportunity there is being able to say, this area was missing and actually it could have been the weakest link in this whole chain. So that's important. Two, there are things that become apparent only when research is done like what you do with research with Mathematica.

I don't think we are going to deliver a green revolution in Africa the way we thought. I think we are going to have to think about food systems and how food systems deliver. So what are those elements? When we work with Mathematica, what of those things do come out? So here I'm trying to say, we really need to be attentive to what the landscape is telling us. We need to be attentive to how quickly things are changing. And here I'm challenging both of us, AGRA and Mathematica, to be intentional and be attentive.

And again, just think about it. Well, Europe has no challenge with any yields, but when we do averages, Africa disappears into that. When will people understand what is going on in Africa? That every two seasons, farmers lose everything. So I'm not sure that I have answered your question, but I just felt like it's important to bring out the fact that when we are following up on these types of data, there are things we learn and we appreciate that.

So for example, most recently, I found out that one of the assumptions we had made around whether farmers use, what is it, climate smart farming, they're actually doing more than we thought. So we are going to have to recalibrate that when we're doing our baselines with Mathematica. We found that farmers are doing it and found that farmers had already exceeded what we were expecting. So we are going, those types of things are important.

[Paul Decker] Are there any other ways in which AGRA, over the time that you've been leading it, has adapted its approach because of those climate-related challenges? So kind of looking historically, what changes between 2014 and now?

[Dr. Agnes Kalibata] We've made a lot of adjustments. One that stands out to me is move from fertilizers as how we do business to soil health management. Because then you find that you're going to have to find a balance between fertilizers and actual other things in the ground that are critical. Even how we think about carbon. We've most recently embraced the whole idea of fertilizers are helped by how much carbon is in the soil. And this is not how we were thinking in the past. In the past, we were just like, the more you have in good seed, the more you have in fertilizers, the better.

So we are evolving based on what we are learning. We have to have the right balance between how we are trying to build food security and food for people and how we use the environment. Right now, most of the production that happens on African continent happens at the expense of the environment. We cut more to produce food. We need to use more inputs. We need to use better seed that give us more, that have better drought tolerance. But we also need to use fewer fertilizers, but use fertilizers, which is very different from some places where people say, oh no, you can't use fertilizer.

No, we will use fertilizers because our soils are depleted. What we'll use, we'll be very intentional and very careful about how we use fertilizers. So I think those conversations are all new. We had the Food Systems Summit that allowed us to really go deeper into conversations around, so what is appropriate now? What type of changing environment are we sitting in? And again, rains are reducing. As they reduce, how do you capture more? More carbon in the soil helps a lot in terms of retaining moisture. These are conversations we're not having before. Everybody's now thinking about how do you improve that? How do you support that?

In drier parts of Kenya where we work, we think about how do we retain every drop that falls? Ensure that it gives more value. Again, you can do that by increasing the amount of carbon in the soils through regenerative agricultural practices, which we've seen deliver value. They are expensive. Farmers are finding it challenging to get them started, but when they embrace them, when they manage to do it, they actually have a major yield differential that shows that this can work. When we use fertilizers and we use them the right way, with the right balance, without using too much or too little, we get the right yields.

So I mean, here, the conversation has shifted from putting everything at all costs to why do we need to do this? And what differential does it add to what I'm doing? And I think we are going to see more of that. We are going to see more. We are going to move towards being a little bit more careful towards how we use the environment and being able to determine good tipping points for farmers. We talk about tipping points.

If a farmer can get to that place where they are producing enough to have the quality of life they need, then we probably need to manage it around there. But what does that mean? What does that mean? What is that tipping point for the farmer? So establishing those and understanding what they mean and how we can sustain them would be a good thing to ensure that we are producing food, not at the expense of the environment, but with balance with the environment.

[Paul Decker] Reflecting back on your tenure with AGRA over the last 10 years, is there one achievement that particularly stands out that helps highlight the impact of your organization on the region's agriculture?

[Dr. Agnes Kalibata] We built a very strong system for seeds access in Africa. We have put in place an institution that will help governments that need to build seed systems. And I look at it because it all starts with seed. You have a good seed, a seed that can make a difference between one ton per hectare and 10 metric tons per hectare. But of course you need other things, but it starts with that gene. Having the right genes packaged in that seed is extremely important. And I feel like across the 11 countries we work with, that ecosystem is working and beginning to work very well. The technologies, if seed is a technology, are available, which is really good.

And the business ecosystem is functioning in ways that can deliver value. The public ecosystem as well, to ensure that the neighboring environment is working, that is also beginning to work in many places. We have one or two struggling countries, but nothing that can't be fixed because the model exists. So I feel that that one I'm extremely proud of because I feel like for any country in Africa, being able to have the right material, the right genes, the right technology to ensure that farmers can produce what they need, that is available. And the system of delivery up to the farmer is available. Yeah.

[Paul Decker] I like the story about the seeds being the beginning and the focus on that. So that's about AGRA and AGRA's impact. In further reflection, more broadly maybe, what lessons from AGRA's work would you highlight as most crucial for achieving locally-led development more generally?

[Dr. Agnes Kalibata] To achieve locally-led development will depend a lot on how well we build on efforts of existing leadership in countries. Countries do have the ability to understand their problems and they do this analysis to understand their problems. I think a lot of time they lack the ability, the investments that they require. And when they have the investments, maybe they don't deliver value in real time. So ensuring that investments are delivering value in real time. But I think it's wrong to think that countries or communities don't understand their needs. They do, they do.

They may not contextualize them enough. So institutions like AGRA, what we do really is to help package. We help package and ensure that once they are packaged, and they are packaged in ways that people understand them, right? Understand that this is working. If it is something that's a local solution that is working, it has to be packaged in ways that shows that there's evidence actually that it is working. If it is something that is popular, it has to be packaged that way. So being able to build on local efforts, being able to help pull it out.

Sometimes the information is available. If it is not available, talking to people and being able to pull it out is important. Because if it has worked, it will probably give the highest results. So I give you a very simple example. In Rwanda, we have a program called One Cow per Poor Family. That program builds on the fact that if there's one thing that a Rwandan household would want, loves, is a cow, right? It's used as a dowry, it is used in everything. Most importantly, they look at it as not just a source of milk, but a source of manure. So for them, it's a factory.

We have small land holdings and you almost like need a manure factory and that cow is your manure factory. So it's really like a sacred thing. If you want to build wealth, there's no Rwandan that will not treat a cow like this is my source of wealth. So we use that as an instrument of building a capital base for communities that were struggling. And it works very well. And it again, helps them deal with the challenge of malnutrition, helps them deal with the challenge of crop production, helps them deal even with the challenge of coming together as a community, which we needed badly.

So it's really important that even when local context is not clear, that we take our time to bring it out. And it's not very difficult. You need to sit with the community and have conversations with them and you will understand to them what is really important. And build on that, you will almost always be successful if you're building a local context. If you came here to Kenya and you built on the fact that Kenyans eat maize, you'll be successful. But in other places, it's like, why do people eat maize? But that's what we eat.

So you build on that, you'll be successful. There's a market, there's a community that understands how to grow maize. And if you just help them grow it right, you'll be successful. So I think again, building on what people know goes a very long way. And we've a lot of times taken advantage of that by ensuring that always when we go to countries, our starting point is engaging governments to understand where they're prioritizing because they already understand what their biggest competitive advantage is or comparative advantage. We're also engaging communities to understand where they're at.

[Paul Decker] Dr. Kalibata, I appreciate your insights and I also appreciate AGRA's mission and the impact that it's had in your world. So thank you for participating in the podcast today.

[Dr. Agnes Kalibata] Thank you and we look forward to building on your work on evidence as a policymaker. Evidence is always the place to start, and it really helps governments when they have evidence as a place to start designing their policies and that's what Mathematica does. So we appreciate that.

[Paul Decker] Thank you.

[Dr. Agnes Kalibata] Thank you.

[JB Wogan] Thanks to our guest, Dr. Agnes Kalibata and thanks to Paul Decker for stepping in as the guest host for this episode of On the Evidence, the Mathematica podcast.. If you’re interested in innovative Africa-based leaders using data to improve lives across the region, you should check out our interview with Jesse Moore, CEO of M-KOPA, a company that provides financial services to low-income and unbanked entrepreneurs in East Africa. And if you're a fan of the show, please consider leaving us a rating and review wherever you listen to podcasts. To catch future episodes, subscribe at mathematica.org/OnTheEvidence.

[Music]