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 to young researchers. For this interview, we got the opportunity to talk to Prof. Jacob Moscona.
Jacob Moscona is an Assistant Professor at MIT. His research explores broad questions in economic development, with a focus on the role of innovation, the environment, and political economy.
ETRM: Thanks for sitting with us today. What inspired you to do research in economic development?
It happened during college. I was about halfway through college, and I was really interested in history, understanding historical episodes in different parts of the world. Then almost by accident, just because it sounded interesting, I took a class in the political economy of Africa, taught by James Robinson and Robert Bates. James was coming at it more from an economic perspective and Robert Bates was coming at it more from a political science perspective. It was totally eye-opening to me that you could try to ask and answer these huge questions that I thought were insanely important, but using data analysis, combined with a bunch of examples, and a very rigorous understanding of context and history. The two of them were up there debating with each other about how to think through the methods and how to think through these different episodes, and I just thought that was the coolest thing I’d ever seen.
That’s what got me interested in these topics, and by the end of the semester, I got up enough courage to go and talk to James in one of his office hours and tell him some version of this. He was an incredibly gracious mentor, and during the next semester I started working with him. One thing led to another. That’s kind of how it started, it all clicked in that class and I never looked back after that.
ETRM: This links to our second question. How did you get in contact with different society structures like age-set or segmentary lineage that you study in some of your papers?
The idea of thinking about age-based social organization or linage-based organization first started coming up in this class that James and Bates were teaching. And in the next semester I took Nathan Nunn’s undergraduate development class, which also covered some similar topics about these different features of society. Nathan and James were two amazing mentors and one think I learned from them is the big advantage of talking to people who are working in different fields. As soon as you start talking to someone like Robert Bates, who’s a political scientist who writes some books about the post-colonial period in Africa, things like age sets come up all the time. Same with talking to or reading anthropologists. But then, there’s a lot that can be learned by pairing up all of that knowledge with thinking about how this can matter for questions that economists care about.
In that paper about age sets, we thought, “Well, how could this all matter for some of the most straightforward questions in economics like: How do people share risk? How do they form financial networks? How do they form support systems to help people when they’re in need?” The social organization of society will lead to very different ways of doing all those things. But that just hadn’t been studied before because you need to smash these two different sets of ideas together in order to start making those connections. So, I think partly it’s from interest in other fields and reading a lot, and partly from being lucky and learning about a lot of the stuff from James and Nathan.
ETRM: That’s awesome. Back then, your papers were on historical political economy. Now, you have shifted to environment and innovation research. What led you to explore these topics?
I always was interested in both. The first research paper that I worked on was a final paper in Nathan’s undergraduate class, where I tried to understand the impact of innovation and agricultural productivity growth during the Green Revolution on patterns of structural transformation, first across regions in India and then across countries.
Back then, partly because of my biochemistry-oriented background, I was interested in how innovation shapes productivity, especially in the process of development. Innovation is relevant not just in a handful of rich countries, but around the world. I think people tend to ignore the role of innovation and R&D more generally, assuming that large parts of the world are just adopting technology and don’t play a part in the innovation process. That always seemed a little bit like a blind spot to me.
So I had always been interested in these topics, but then things kind of clicked during the first year of the PhD taking the economic growth class that was being taught by Daron Acemoglu and that class is really focused on the role of the direction of technology and the focus of innovation on shaping who benefits from technology, what problems technology solves, and the problems it doesn’t solve. At the time, those were fascinating ideas, but they existed mostly in theory. There were a lot of models of this, but not much on trying to take those ideas to the data and trying to understand what forces do shape this focus or direction of innovation, how that in turn shapes who benefits from the technology, how that shapes inequality, how that shapes prosperity.
ETRM: Our next question is about your last paper about Embrapa. You found huge effects of public research on productivity. How can research from Embrapa, for example, be exported to other countries?
We already had some of this work showing that most innovation is concentrated in a handful of countries, and as a result, it’s devoted to solving the problems that exist in those countries. In particular, in agriculture, if let’s call it 50% of private R&D investment happens in the U.S., all that investment is going toward designing technologies that work well with the specific ecological conditions in the U.S. and not necessarily elsewhere.
The question is, what is the rest of the world to do about that if you really think that having appropriate technology for your context is essential for productivity growth?
One possibility is to hope that some innovators elsewhere in the world shift the focus of their innovation, or that international institutions get together and provide incentives to develop technologies for developing regions. There are a handful of examples of that working, but they are kind of the exceptions that prove the rule, which shows how challenging those examples were.
Another possibility we think is actual concerted investments in not just technology adoption, but in true R&D investment in different parts of the world. The best example of that that kept coming up, especially in agriculture, is Embrapa. This is a clear example of really being able to do this and solve this inappropriate technology problem or this technology mismatch problem. Brazil goes from, in the 1960s, being a net importer of food and a major recipient of food aid, to the point where now it’s one of the most productive agricultural countries in the world. It had the fastest productivity growth over that period of any other country with a population over a million.
But despite this pattern I think there is a lot of skepticism about: Can true frontier innovation happen outside of a handful of rich countries? Are all of the stories of Embrapa are just extreme anecdotes? Or maybe Embrapa did have a big impact, but the cost was so high that it didn’t justify the benefits. And, if it did work, what are the mechanisms that led this to really happen? That’s what we wanted to answer and how this project fits into that broader set of questions.
To your second question: Could this then benefit other parts of the world? I think the answer is almost certainly yes. To the extent that the kinds of technologies that Embrapa was developing in and for Brazil are more applicable to other parts of the world than technologies getting developed in the U.S. and Europe for those markets, then you can imagine having huge benefits. There are all these initiatives to try to translate technology that Embrapa is developing to parts of Central Africa, which in many ways is ecologically pretty similar to parts of Brazil they developed technology for.
I think that the synergy between R&D and different parts of the world is interesting. Some other work on the rise of China in innovation is focused on high-tech startups and venture capital investment, where China really had a massive takeoff in the 2010s to be the second major hub of VC investment outside the U.S. In one paper we show that that actually had huge benefits for a lot of other developing countries because these apps and technologies developed for the Chinese market were applicable in other developing countries in ways that U.S. apps weren’t. Things like social commerce. Things like online education for primary and secondary school—there is much less demand for that in the U.S. than there is in China. But suddenly in China, this takes off and that’s applicable in India and in Brazil. So, you’re exactly right that it is possible for these kinds of major spillover effects from this innovation happening in different parts of the world.
ETRM: There’s extensive literature on what constrains the adoption of technology and still the adoption of technology in some areas, like Africa for example, is still low. We know the answers. The government knows the answers. What is missing?
Widespread technology adoption is really hard, and I think that not enough credit is given to how much it took in terms of coordinated public investment, in terms of building on institutions and a whole range of things in the U.S., for example, to achieve the level of technology use and productivity that happened. It was massive: building out agricultural experiment stations in different parts of the country, having land grant universities that are building all this technology for different regions of the country, investing tons in R&D, extension programs, teaching farmers how to most productively use technology… During the early parts of the U.S., the Foreign Service everywhere they were in the world were sending seeds back to the U.S. that were being distributed through the postal system, with farmers being told, “Just try this stuff out and see what works.” That’s just one small example. There were tons of programs and experimentation that got to the point where the U.S. is now productive and farmers are using technology really well.
So it is hard. And it is even harder in parts of the world where you don’t have a lot of the complementary infrastructure that exists in the U.S. that might facilitate technology diffusion: access to information, things like credit markets, etc.
I think that a lot of the lack of technology adoption or diffusion in different parts of the world is in part because there just aren’t existing technologies that are that productive for those parts of the world. That’s why Embrapa is, to me at least, an interesting and important example. They tried to do a version of what the U.S. did, they went all in on developing research labs in different parts of the country to develop technology for different parts of the country, and it seemed to work. Of course, then the question is, why doesn’t every country do this? That’s hard, there’s some luck involved. The way that Embrapa was designed was crucial, including the fact that it was able to be so politically durable and exist under all these successive administrations and institutional change, and maintain widespread support and popularity.
We had a conversation with a researcher that has been working in Embrapa since the 70s, and he said that the fact that they maintained public support and political support through all these institutional change was really important and something that they worked on. There were tons of ways in which they consciously tried to maintain support, because for things like R&D, you need long time horizons for things to materialize, to really see the benefits, to really build a base of human capital that can research and get the PhDs and keep investing in this stuff.
I think one problem with a lot of recent research that is really focused on these super micro-interventions of “maybe I just nudged someone this way and that can solve this problem of low adoption” is that it inadvertently might lead one to think that these are easy problems to solve or that they’re not complicated and might not require large-scale, coordinated investments or government institutions. It sort of minimizes it in a way that I think might lead one to miss the fact that this is a really hard thing to do.
ETRM: Thank you so much for your time and insights. We really appreciate you joining us today!
