Jeremy Majerovitz is an Assistant Professor in the Department of Economicsics and affiliate faculty at the Building Inclusive Growth (BIG) Lab.
How did you get interested in development economics and research?
Jeremy Majerovitz: I was interested in economics and research for a long time. Bob Lucas, who was a macroeconomist and Nobel Laureate, famously said that once you start thinking about cross-country differences in output per capita and living standards, it is hard to think about anything else. And that is, I think, one of the most important economic issues. And once I started thinking about it, it was very hard to stop thinking about anything else. What's interesting as well is it's actually something where we really still don't understand very much about. Why is Kenya so much poorer than the United States? And that's why I decided to do research.
In your paper Scaling Up Agricultural Insurance, you look at risk and agricultural insurance. Could you tell us how you identified that Ghana was the place you wanted to examine that in?
Ghana in particular was the place that we decided to look at because that is where some interesting microeconomic experiments have been done. This is a topic that's important in a lot of different settings. There's a very nice paper by Dean Karlan and co-authors where they experimentally gave agricultural insurance, like rainfall insurance, to some farmers and not to others, and they looked at the effect that that had on a bunch of different outcomes, including on agricultural investment. That paper was influential because it showed that when you give farmers agricultural insurance against drought, they invest more and they switch their crops towards crops that seem to be higher return, but also higher risk. Because once they're insured, now they can make these investments that are high return, but if they didn't have the insurance, they would be a risky bet.
What we're doing is we're saying, “Okay, we have these microeconomic experiments. What does that tell us about whether this is a good policy to scale up nationally?” And that's where a lot of research in macroeconomics and development has been going, and where some of my research has been going. So, we're building a model of the Ghanaian economy on top of that experiment and saying, “Let's take these microeconomic experiments that people have run, let's build on top of that an economic model, bring in some additional data from other sources, and try to see what are the effects going to be when we scale this up?”
How was it working on this project and where were some of the challenges you ran into?
The big difficulty with these sorts of projects is figuring out “How am I going to make a tight connection between the data that I actually have and the question that I want to answer?” Something that came out of this project that was nice is that once we wrote down the model, we were able to understand what actually matters in our model and more generally in different models of agricultural risk. Part of what came out of it was understanding that what matters for the decision of whether the government should subsidize something is whether there are benefits from the thing being subsidized, beyond what the recipients will internalize. In this context, the households benefit from insurance in all sorts of ways. That doesn't necessarily mean that the government should subsidize things. In the same way that if I get a cheeseburger, I benefit (maybe there are some costs as well from eating that cheeseburger). But that doesn't mean that there should be a government subsidy, because I'm already internalizing those benefits when I make the decision of how many cheeseburgers to buy.
Here, what is unique is that when you're in a world where insurance is incomplete, there are some benefits that come through changes in prices. What we discovered in the process of building the model is that one of the key forces is that if insurance is subsidized, that's going to encourage people to make more of the risky crop. They're going to grow more maize - in this setting in Ghana, maize was the risky crop. That is going to change the relative price of maize compared to a less risky crop. Then those changes in relative prices, it turns out, provide some insurance benefit. If maize is the risky crop and mangoes are the low-risk crop (those are the two crops in this setting in Ghana) then it's good for households that maize becomes less valuable relative to mangoes, because it helps them by reallocating some of their income from the good state of the world where it rains, to the bad state of the world where there’s no rain and the maize doesn't grow.
Once we understood that, that became a more general force in all sorts of different versions of the model that you can write down. What also is nice is that it clarified, “What are the things that we want from the experiment?” On the one hand, we want to know how insurance demand responds to changes in prices. That's something that the experiment gave us directly. Another thing that we would want to know is how offering insurance affects the relative price of different agricultural goods. That's not something that's quite directly provided by the experiment, but the experiment does give us information about how crop growing patterns respond to offering households insurance. So, on the one hand, it's always difficult to know, do I have everything from a micro-experiment to really extrapolate to the effects of a policy? But I think that in this case, we were able to figure out that the experiment was giving us a lot of the right pieces of information that we needed.
More broadly, how do you see your work affecting people’s lives and policy?
Good question! Because we still understand so little about why poor countries are poor, there's a lot of basic science to be done. That basic science is going to frame what policies we should be thinking about. Of course, my hope is to also do some more directly policy relevant work. But for example, one theory of lack of development is that resources are misallocated. So, a lot of my recent work has focused on whether capital is allocated efficiently across different firms, both in developed and developing countries. If I think that we are in a world where capital is efficiently allocated across firms, then that tells me that there's very little scope for things like microfinance, subsidized credit, or other financial interventions to improve things. If I think that there is misallocation of capital and/or of credit, then that's a world where I want to explore those types of government policies. I have a paper looking at microenterprises and finding that there is a lot of misallocation of capital across microenterprises. That suggests there's at least some scope for financial interventions to improve things, but we still need to do a lot of work on different dimensions of figuring out where there's scope for interventions to matter and where there isn't.
What new areas or new projects are you looking forward to?
Something I've been working on a lot is whether capital is allocated efficiently across firms. I've been working on a paper in the US looking at whether the cost of capital is the same across firms. What we find there, using credit registry data in the US, is that the cost of capital is pretty similar for different firms that are borrowing from commercial banks. That suggests that there's relatively little misallocation of capital, at least among that segment of US firms that makes it to the point where they would borrow and get a bank loan. We want to extend that same methodology to other countries; we're in the very early stages of figuring out what countries that's feasible for. The idea would be to then say, we think of the US as this efficient benchmark. If I go to other countries, is there a lot of misallocation of capital coming through imperfections in the credit system? If so, then that is something for policymakers and others to work on, and if not, well, we shut down one channel to worry about.
Has it been interesting having done a domestic project versus now you're trying to extrapolate that internationally? That seems like quite a challenge.
Certainly, for me, it was a funny place to start, because most of my research is focused on developing countries rather than in the US, but I think that in a way it makes a lot of sense, because you have to make that comparison. Sometimes we'll notice something in developing countries and say, “this seems very inefficient,” or “this seems very unusual”. Then you always want to try to do a comparison. For example, there's a big difference in developing countries between the price that you pay for food when you buy it at the market and the price that the farmer gets. It turns out there's also the same big difference in the United States as well. So we want to do some comparisons to try to determine what is different about developing countries, as opposed to what is just a fact of life wherever you go.
This interview has been edited for length and clarity. It was conducted as part of a series of faculty profiles on BIG Lab Affiliated Faculty coming out this spring. Stay up to date with the BIG Lab by following us on Twitter, LinkedIn, or Bluesky.