In Conversation: Lauren Falcao Bergquist & Amit Khandelwal on trade & development

Development economics and the study of international trade, while traditionally distinct, are increasingly overlapping. Trade economists now pay closer attention to how market frictions like credit constraints or labor market issues affect global trade, and development economists are beginning to utilize tools and models from trade economics to better understand the growth and development process in low- and middle-income countries.

Yale is at the frontier of research on international and domestic trade and the functioning of markets, melding techniques from traditional macroeconomic and microeconomic approaches. Amit Khandelwal, the Dong-Soo Hahn Professor of Global Affairs and Economics, has analyzed the dynamics of the US-China trade war, and Lauren Falcao Bergquist, Assistant Professor of Economics and Global Affairs, runs randomized experiments in African crop markets. Both are EGC affiliates and members of the Markets and Development initiative. In a recent interview, they discussed how insights from their respective subfields are increasingly influencing each other – and the implications for research and policy.

Both of your research focuses on the role of market frictions in trade and development. Why are markets such an important lens?

Bergquist: Development economists think a lot about markets when studying why poor countries are poor or thinking about what’s happening within countries. Markets exist to connect producers and consumers, and when things go wrong, there can be real welfare implications. Agricultural markets, for example, provide income to farmers and food security to consumers – so their proper functioning is critical for livelihoods, consumption, and welfare.

But market frictions and trade costs can create wedges between farmers and consumers. Credit constraints, for instance, can lead to large price fluctuations: if farmers lack access to bank loans, they’re often forced to sell their crops right after harvest, when prices are low, to pay urgent bills – which means that consumers buying food during the lean season will face very high prices, so much so that in many contexts this is called the “hunger season.” Better access to financial markets can lead to less price volatility.

Khandelwal: Markets are a really useful concept in trade, too. When you open up to trade, you can shift resources to sectors with comparative advantage – but in developing countries, that process is often fraught with local market distortions and frictions. If the US imposes new tariffs on China, for example, standard trade models would say that firms in countries like India should expand and export more to the United States. But once an Indian factory exceeds 100 workers, it’s subject to more onerous labor laws. Credit constraints can prevent firms from making investments. Land titling issues can pose major challenges.

In some sense, domestic policy is international policy: while global trade is a powerful force for improving welfare, whether that happens depends on how domestic markets actually function.

How are trade and development economics overlapping, in terms of the questions they ask and methodologies they use?

Bergquist: Historically there was a kind of split, with trade economists often taking an international, cross-country perspective and development economists emphasizing domestic institutions, context, and market failures. This was especially true over the past 20 years, as development economists focused more on randomized controlled trials – or RCTs – and other forms of micro-measurement that aren’t as amenable to questions of international trade. 

But the lines are blurring, leading to helpful cross-pollination. You see more trade economists doing domestic work and more micro-development folks thinking about how their work interacts with international forces.

Khandelwal: One reason you’re seeing them merge is that development economists now have a lot of evidence on how local policies work, so they’ve started thinking about scaling them up. But to think about how local policies might affect prices, labor, and other factors at the national level, you need a different apparatus. As it happens, trade economists have been thinking about the national implications of single policies for a long time – because the key policy we tend to study is tariffs, which are national by definition.

Bergquist: Tools from trade economics, like the focus on general equilibrium (GE) effects – or how policy changes ripple through an economy – can be really useful. A recent study in Kenya, for example, uses experimental evidence to estimate the effects of cash transfers on broader outcomes like prices and GDP.

Similarly, in a recent paper we focus on scaling up fertilizer subsidies – one of the most basic policy interventions in agriculture. Many small-scale RCTs have shown that fertilizer subsidies, on average, help farmers increase crop yields. But what happens when you scale the program nationally? Output prices may fall, or complementary input prices may rise – and these GE responses will change the returns to the program. The overall GE effects will depend on what farmers are growing across the country, what people are consuming across the country, and the trade costs and frictions in the markets that connect them. Trade economics offers tools to model these questions. How can we develop a machinery for doing that that's tailored to small-scale RCTs?

Using experimental evidence and administrative data from Uganda, paired with GE modeling, we found that overall gains at scale are lower than what the RCT results would suggest. By increasing output, the subsidy reduces crop prices while increasing labor costs for farmers. But there are some interesting distributional effects: while the local policy is regressive, since large land-owning households benefit more from the subsidy, once GE price effects are factored in, poorer households benefit more. Domestic trade costs play a big role: in remote and disconnected areas, the price effects are much larger compared to regions that are better integrated with global markets.

Has the influence gone the other way, from development economics to trade?

Khandelwal: A similar intersection is happening with tools, insights, and approaches being borrowed from development and labor economics to try and be a little more micro and make fewer assumptions. With tariffs, a classic assumption is “complete pass-through”: when a tariff increases by, say, 20%, trade economists traditionally assumed that it gets entirely passed through to consumers. In reality, we know that’s not always the case and that it varies greatly across markets – but to make the problem solvable, we tend to put a lot of assumptions into our theories and models. More and more, trade economists are saying: "Let's get some direct evidence and see, in that example, by how much prices actually rose when the tariffs went into effect. If we don’t see complete pass-through, then maybe we need to revisit our assumptions.”

Merging micro and macro approaches would seem to add new levels of complexity and uncertainty. How do you manage those trade-offs?

Khandelwal: All researchers face trade-offs between specificity, or knowing everything about a context, and generalizability. The most effective research integrates general methods with insights from specific studies, but we’re also starting to see attempts to blend the two.

In Chile, for example, comprehensive data linked to tax IDs allows researchers to observe all transactions between companies and individuals. In that context, you don't necessarily need models to understand policy impacts: you can just use before-and-after comparisons to see how different groups are affected. This is very recent work, but those kinds of data sets could enable research to be simultaneously very broad and very rich.

Bergquist: What you’re really trying to learn is something general, to understand how things work and make better-educated guesses. Like Amit said: "In general, how do firms share tariffs with consumers?" What does that tell you about the market structure and demand, and what predictions can we make about future policy changes? Economics is a little bit like basic science: we’re not working in a lab, by any means, but we are trying to understand some of the more fundamental forces at play.

As trade and development economics evolve, how can researchers seek to better inform policy?

Khandelwal: Many academics want to work on topics that will inform policy, but academics themselves are probably not best-placed to design policies. They don't know the setting or the political strengths and weaknesses – which, of course, policymakers within countries know best. But there is a layer of economists at the World Bank, International Monetary Fund, World Trade Organization, and regional development banks who sit in between academics and policymakers. They can digest the latest economic research, but they also know what policymakers are interested in. And they can manage that timescale, where policymakers need to quickly learn everything about a given topic. This layer plays a really important role in disseminating knowledge.

Bergquist: Research networks are important, too – like the International Growth Centre, which has economists who actually sit with policymakers – and places like Yale’s Economic Growth Center, which helps to communicate academic ideas broadly. One additional thing I’ve found effective, in terms of policy impact, is deeply engaging policymakers, NGOs, or even private companies from the beginning. Researchers can get more buy-in, tailor questions to what policymakers actually care about, and share real-time updates rather than waiting years for the academic paper. 

There's also a lot of tacit knowledge to share. For example, I'm working with Rwanda’s Ministry of Trade and Industry on an RCT of an export promotion program, and it’s been a very productive relationship. Our initial analysis helped them decide to ask for renewed funding, and we’re constantly learning a lot, too – about new things coming up and things we just haven't thought about yet.