In Conversation: Andreas Ferrara & Melanie Xue on understanding race and gender through economic history

Economic history offers a powerful lens on how today’s outcomes are shaped by deep historical forces. Combining historical data with modern empirical methods, economic historians can generate new insights on how economies grow and develop – from markets, trade, and inequality to wider social and political change, often casting fresh light on today’s world.

As part of its effort to foster research in this exciting field, EGC’s Program in Economic History regularly hosts visiting scholars. In a recent conversation, Melanie Xue of the London School of Economics and Andreas Ferrara of the University of Pittsburgh discuss how they use historical data to study gender, race, and other questions at the heart of contemporary policy debates.

How did you become interested in economic history?

Melanie Xue: Growing up in China, my dad’s generation came of age just before the Cultural Revolution. He became an electrical engineer, but he’s very well-rounded – especially in history. That gave me an early interest in the humanities and social sciences.

I studied economics in college, which felt natural coming from a financial center like Shanghai. But while most of my classmates went into consulting or banking, I was drawn to academic research, which led me to the United States for graduate study. I started in political economy and public choice, but came to see those topics as best studied in historical context. Much of institutional economics, like Douglass North’s work, is also economic history.

Andreas Ferrara: My dad was also an electrical engineer, but I stumbled into economic history by accident. After doing mandatory military service in Germany, I went abroad to Scotland for college, where an early mentor – Sascha Becker – introduced me to economic research. During my graduate study in Italy and England, I eventually focused on migration and discrimination using older data and natural experiments from history.

How would you explain your research to someone unfamiliar with economic history?

Xue: Traditional economic history simply asks economic questions about the past while more recent developments in the field focus on using historical context to address economic questions. My own research aims to do both – or at least to strike a balance between the two approaches. my research uses historical context to study economic questions more broadly. I focus on cultural values and long-term development, across contexts and time periods. Seeing very different cultural values after growing up in China and then studying in the United States, I thought about those questions constantly.

Ferrara: Like Melanie, I focus on unearthing new historical data to answer causal questions that modern data can’t. The topics I study are often politically controversial – migration, discrimination, underrepresented groups, and labor – but I approach them as economic questions. Data collection in economic history is harder and often takes longer, but that’s part of the appeal: we ask innovative questions while also bringing innovative data to the table.

Much of your work focuses on race and gender in US history. What are some of your main findings and how are they relevant to today?

Ferrara: I focus mostly on the United States from the 1850s onward. For example, one project studies how the Civil War impacted women’s work and political participation. Analyzing Union Army records and US Census data, we find that high shares of wounded or disabled veterans increased women’s labor force participation and women’s political activity in the temperance movement, ultimately leading to higher support for Prohibition.

A related project looks at who left the South in the decades following the Civil War. We focus on white Southerners linked to the Confederacy – former plantation owners, slaveholders, and officers’ families – who migrated between the 1860s and the early 20th century. Using linked Census records and new archival data, we find these migrants brought significant grievances to frontier areas, reestablishing Southern communities without slavery but with a clear goal of keeping Black people and other minorities out. These towns were more likely to have streets or schools named after Confederate leaders, children named after Confederate figures, higher rates of Ku Klux Klan activity and racial violence, and a greater likelihood of so-called “sundown towns” and expulsion events – when Black, Jewish, Mexican, and Chinese communities were forced out of many US towns in the late 19th and early 20th centuries. They were also more likely to become so-called “sundown towns,” which enforced racial segregation through signs warning non-whites to leave by sunset or face violence, harassment, or discriminatory laws. While many towns are actively working to address this historical legacy, it persists through informal exclusion, subtle hostility, and persistent segregation that keep some communities overwhelmingly white.The paper helps explain why much of the US interior has lower shares of Black people and other minorities, compared to coastal areas and the Northeast.

Another paper studies what we call the “Other Great Migration,” parallel to the Great Migration of Black Southerners to northern industrial cities in the 20th Century. We asked: what about white Southerners? And we found that over the same period, an even larger number of white Southerners also left the South. Interestingly, they tended to go to rural and suburban areas in the western United States, rather than northern cities.

These migrants tended to hold more racially and religiously conservative views – even in modern survey data, white Southerners still have more conservative attitudes and beliefs than white non-Southerners. How did this Other Great Migration influence local politics in the places where they resettled? We show that the inflows shifted political outcomes toward more conservative voting, with effects that persist today: places that received more of these migrants still show more conservative voting outcomes, including higher support for Donald Trump in 2016.

Melanie, what does your work reveal about social norms, cultural values, and gender in Chinese history?

Xue: My job market paper looks at gender norms in the context of China’s “Cotton Revolution,” roughly from the 1300s to the 1800s. During this period, new technologies and regional specialization in China’s cotton weaving industry allowed women to become major income earners, often matching or even exceeding men’s earnings. Unlike the need for physical strength in agriculture, women’s smaller hands and greater dexterity gave them a distinct advantage in weaving. They often became the main income earners and made household decisions, even after a husband’s death.

Using a difference-in-differences approach, I show that these large productivity shocks increased women’s earnings and reshaped gender norms. In regions where women earned substantial income from weaving, women were described more positively in classical poetry, and there were more cases of husbands joining their wife’s family. These shifts persist in modern data: women from these regions exhibit higher educational attainment, marry later, and are more likely to hold leadership positions. In regions where women only did lower-value work like cotton cultivation, these effects don’t appear.

Many people think cultural values come from the sky and don’t change; my research suggests that they have material roots, and that they can change as women’s economic circumstances change. Notably, gender norms shift not just when women work, but when their earnings reshape attitudes about women’s capabilities and status.

What new trends are shaping economic history today, and how is your work part of that shift?

Xue: One important trend in economic history is a growing effort to measure concepts that were once treated as hard to quantify, such as values, beliefs, and informal norms, using historical data. Economic historians are increasingly pairing modern empirical tools with new kinds of historical sources, including text and other unstructured material, to build measures of these forces and link them to economic change over time.

Some of my work is part of that shift by developing ways to measure cultural values from unconventional historical sources. One example is a paper on folklore that emerged after Stelios Michalopoulos and I came across a unique catalog of oral traditions spanning nearly 1,000 societies. We were among the first economists to show that oral traditions contain recoverable signals about values, beliefs, attitudes, and decision-making. Beyond making that conceptual point, the project also developed a concrete methodology: we broke the corpus apart, analyzed it in detail, and designed a way to extract usable and comparable measures.

I apply a related idea in more recent work on British proverbs used around the time of the Industrial Revolution. As in the folklore paper, the focus is on measuring cultural values embedded in oral traditions, but the historical question is different. The project brings new data and methods to bear on one of the central topics in economic history, while also showing how these kinds of sources can be used to study cultural change in a systematic, empirical way.

How has technology reshaped economic history and its methods over time?

Ferrara: Computing power has greatly expanded what we can do with historical data. In my research on the Civil War, for example, I use roster data from the entire Union Army’s 2.2 million soldiers – compared to Bob Fogel’s famous sample from the 1970s, which included just 40,000. What makes this moment exciting is the rise of large language models and AI, which have revolutionized what we can do with different types of data – from analyzing massive amounts of unstructured text to uncovering patterns we didn’t know existed. Recent work by Clément Gorin, Stefan Heblich, and Yanos Zylberberg, for example, extracted sentiment data from old paintings. 

The challenge is understanding what’s happening under the hood. These models are complex and we don’t fully understand why they work; for economists, it’s important to understand how the data are generated and what the machine is doing. That’s a key methodological challenge for economic history.

What stood out from your semester at Yale?

Ferrara: Being exposed to the campus environment was really special. Everybody at Yale is a leader in their field, the resources are remarkable, and the university attracts top speakers from around the world. EGC supports a really vibrant intellectual atmosphere, including a lot of engagement with economists across fields; I found the weekly seminars on economic history particularly stimulating.

Why does economic history matter for policy today?

Xue: For researchers who study social norms and cultural values, including around race and gender, a key concern is that policies are often made without considering these factors. Historically, we’ve seen how disruptive, ineffective, and even dangerous it can be when externally imposed policies ignore local environments. Even today, policies and research on development and growth can and should take these considerations more seriously. Economic history is a great place to start.