As part of our interview series, we ask renowned experts in the field about the future of research in development economics, and for their advice for young researchers. For this interview, we got the opportunity to talk to Prof. Fernando Aragon.
Fernando M.Aragon is a Professor of Economics at Simon Fraser University in Canada. He received his Ph.D. from the London School of Economics and joined the department of Economics at SFU in 2010. His research relates to the fields of Development, Environment, and Political Economics. His work is applied and explores the role of natural resources and institutions on local development, economic effects of pollution, and adaptation to climate change, specially of rural households in less developed countries.
ETRM: How would you define development economics on the frontier currently?
Fernando: I think development economics is economics in a context with a lot of market failures. So, you can do labor, education, environmental, but once you have an environment where you have a lot of distortions in the market, then it becomes, I think, development economics.
ETRM: So, it doesn’t necessarily have to be in developing countries, right?
Fernando: No, exactly. At the beginning I thought it was like that: if you do work in Ghana, then it’s development, but if you do work in US is not. But let’s say you’re studying Aboriginal peoples in United States. They have an environment without perfect property rights because it’s a tribal land, and they have to make decisions with small polities. You’re back to using the same models and tools as you will be using in Ghana, because you are in a world with market imperfections. But if you were studying what happened with forests in Silicon Valley, you don’t need those tools. So, I think what is crucial about development economics is that you’re studying economies that have a lot of market distortions. What is on the frontier? I think it varies. Everybody will tell you what they’re working on the frontier, but I think what makes the frontier is when you are asking new questions or answering questions from a new perspective. For example, when you get a new take on something, it could be a new question, a new data, a new approach.
ETRM: Your answer is very informative because when we talk about development economics, even among a lot of us, we think about working in developing countries. But having the perspective from what you just answered, there are a lot of important questions even in developed countries.
Fernando: Yeah, definitely. You will see, if you look at the Journal of Development Economics, you will see that there are papers of Aboriginal people, but also papers in United States that works on a particular population. For instance, you can think about blacks in United States or homeless people. You can use tools of economics, development economics, models and approaches. In those contexts, there are critical constraints, imperfections, distortions that prevent the accumulation of human capital for certain individuals, and inequality of access. That’s a little bit of the trap of development economics: you can start looking at everything from that perspective. So usually what you will see nowadays is that development economics are specializing in different kind of fields. There are people working on environmental and development, or housing and development, or labor and development, because they’re basically applying the tools of that field to that particular context with market imperfections or distortions.
ETRM: Which do you think are the main areas that still need more research and that are also of interest to policymakers?
Fernando: There are two ways to answer this. One is the kind of package answer saying anything that is providing you new insight is important to policymakers. That’s not informative. From my perspective, which is again narrow bias from my particular experience, I think what we don’t really understand is how the context matters for human behavior. Let me be more specific. We have models that explain human behavior, consumer decisions, farm producer decisions, fertility decisions, human capital accumulation, etc. And these models are neoclassical. They make assumptions about the context. Usually, it’s a context which has no market imperfections. But what development economics teach us is that once you have market imperfections, these decisions are no longer separable. A simple example, if you are living autarky, you have no access to input or good markets, if you are a farmer, how much you plant, whether your kids will work on the farm, where you want to farm, depends on what you expect to consume. How many children you will have, will depends on the land you have. So, all the decisions are interlinked, and they are not linking in a monotonical way. That we don’t understand. What it means is that the outcome of interventions varies in contexts with different market failures or distortions. If we do an intervention in a country, we want to say there would be differences. We could learn exactly from our model what is driving these differences, but we don’t know yet what drives them. And that’s really important. I think we’re not ready because that’s basically trying to understand the complexity in our models. That’s something that is missing in the research, and it is super important.
ETRM: In order to understand those contexts for a specific place, it takes a very long time. What do you think would be, one of the better approaches for development economist to try to tackle those issues, focusing on very specific domain, specific country, or trying to understand the context before they design something. It sounds like it goes hand in hand as well.
Fernando: There are two things. First, understanding how these contexts matter, is not a job of one person. It is a job of the whole generation of people producing, and somebody eventually compiling that information to understand that context. So, in a particular place or policy, we know how to do it. Let’s say I’m working for microfinance. I want to know what works and what not. We can do trial and error. We can do the same intervention in a different context and look for common pattern. That could be one agenda. Now that said, our profession doesn’t value replication. We need to start value replication and meta-analysis of our research more. We don’t have good incentives to do it. Because we have so much focus on the originality, sometimes we end up with 20 papers which are not comparable at all.
ETRM: I think your answer connected really good to my next question. Two of your papers (Aragon and Rud, 2013; Aragon and Winkler, 2023) are about mines and their local impact. At first sight their results seem mixed: one finds a positive effect of a gold mine on the economic activities of their area, the other finds no effect of transfers that mainly originate from mining taxes. When you find these different results, how does that motivate/modify your research agenda?
Fernando: First, the two studies are not comparable because they are looking at different ways in which mines interact with communities. In the first one mines are buying services, injecting money directly. The other one is indirectly. The local government decides what to do. What I learned from that is that the way you interact is important. It’s not the same paying wages as giving the money to local governments to distribute. And that has been seen in other context, the way you interact is important. Now, how I modify my research agenda: it makes me more curious. The main skill of researcher is not using analytical tools, or Stata, or R. The most important thing is the creativity, and creativity arises from curiosity in this context. You have to be curious. If you have a paper and you know the answer, then why work on that paper. You have to be surprised. When I get surprised by results, then I feel more motivated to keep working on finding new answers.
ETRM: You have some hypothesis of what might happen, right?
Fernando: Yes, but that’s a discipline. When you approach your research, you need to have a hypothesis, not because you want to prove that hypothesis, but because they give you a discipline on how to approach a problem. You need to be open to the answer. Of course, you need to be sure that is not noise or measurement error or mistakes that you made with coding or the data. But once everything is right, and this is what I find. That’s it. That’s your answer. Don’t try to force the answer.
ETRM: A recent topic in your research is the different productivities across farm sizes and misallocation (Aragon et al, 2022). It was usually believed that the inverse relationship between farm size and productivity was a fact. How did you decide to delve into this topic?
Fernando: I was interested in the determinants of agricultural productivity. I’m not an agricultural economist by training. And I was looking at all these papers that documented the inverse size-productivity relationship. Basically, they look at a measure of average product, plotting against farm size, and they find this negative relationship: small farms are more productive. And I was like, it doesn’t make any sense. So, I worked on that with some co-authors and wrote a paper trying to show what we learn from this relationship. The way I think about papers is that I hear comments, opinions, and then I want to know if they are correct or not. For instance, when I was a kid in Peru, a mining country, there were always people talking that mining brings investment and prosperity. So, we need to incentivize mining. I was like, okay, but is that really true? So that led to my first paper. And that’s been kind of the way I think about papers and start working on research project.
ETRM: It’s a very good chat. Thank you so much for your valuable time!
