��ɫ����

Sitting six levels underground, Chenzi Xu started to question her life choices. Here, in a secluded room beneath the vaults of the Bank of England and beyond the reach of cellphone reception, she felt like an intruder. She was monitored by cameras and forbidden from taking pictures of her surroundings because the bank didn’t want anyone to glean information about its security protocols.

But Xu, then a graduate student in economics at Harvard, wasn’t interested in the central bank’s gold. She was combing through its archives to find a different kind of asset: data.

Stitching together tens of thousands of 19th-century loan and shipping documents, Xu — now an assistant professor of finance at ��ɫ���� Graduate School of Business and a faculty fellow at the ��ɫ���� Institute for Economic Policy Research (SIEPR) — pieced together the story of how the London banking crisis of 1866 rippled across the world, slowing development and disrupting patterns of international trade for decades. Her reconstruction of this major panic, detailed in this , could help scholars understand how more recent disasters like the 2008 financial crash can play out for years.

It’s often tricky to zero in on the impact of bank failures because banks rarely fail in a vacuum. Usually other parts of the economy aren’t doing well either, Xu explains, so it may be hard to say if failing banks were a symptom or a cause of the larger financial crisis.

Even if researchers could detangle the various economic threads, they still can’t conclude what the long-term effects of more recent crises will be. “You don’t get to see those outcomes for many, many, many years,” Xu says.

She realized she could avoid those problems by examining an older example, and the London banking crisis of 1866 was the perfect test case. The financial similarities to today are striking. By the mid-19th century there was a huge network of international trade that is considered the beginning of the modern financial era. And the crisis wasn’t precipitated by an underlying weakness in the economy. Instead, it began when a poorly managed interbank lender unexpectedly failed, triggering a run on other banks as worried customers desperately tried to get their money out.

Panic in the suites of London

The panic caused 17% of the multinational banks headquartered in London to fail — a devastating shock to a globalized economy that was largely operated on British currency. “Corporate bonds, sovereign bonds, bank lending — the vast majority of financing around the world was in pounds, much like the way it is in dollars today,” Xu says. Additionally, the financial instruments used to finance trade in the 19th century are still used today. Then as now, exporters took out loans so they could keep operating while awaiting payment as their goods traveled to faraway destinations. “That financial instrument hasn’t changed at all,” she says.

There was just one big problem with studying this episode. There is a ton of data from this period, but Xu had to collect it first. That’s how she found herself in the subterranean archives of the Bank of England for weeks at a time. “I think it’s one of these projects where had I known when I started it how long it would have taken, I probably wouldn’t have started it,” she says.

Bankers recorded transactions at branches around the world in enormous ledgers, so big and heavy it takes two people to carry them. Those ledgers are also nearly illegible. “They had really bad handwriting because this wasn’t meant for the public to see,” Xu says. She returned to the archives over several years, photographing entries for more than 11,000 loans made by 128 banks between 1865 and 1866. Transcribing each entry, she then digitized the data herself.

She also combed through archival copies of Lloyd’s List, a newspaper that reported on the progress of ships, noting when they left and reached port. The paper was like a tracking system for investors who anxiously followed their cargoes and wanted to know when ships were delayed, damaged, or sunk.

These sources helped Xu construct what British lending and international trade was like before and after the crash.

The immediate impacts of the 1866 crisis were huge. One year after the crisis, countries that were more heavily dependent on failed banks saw their exports decrease by over 8%. The long-term repercussions were also significant. The crisis stunted trade growth for many economies. Even four decades after the crisis, importers bought, on average, 21% fewer goods from companies in countries that had depended on the failed banks. “This is a period when the levels were growing, everybody was trading more,” Xu says. “But when you had more bank failures, you grew relatively less. This means that in the long term, you missed out on a lot of the massive expansion in trade that happened during that period.”

New lessons from old papers

Xu says she’s often envious of colleagues who study more recent events and whose data arrives digitized and ready to use. Yet she says there’s lots of value in digging into historical moments that can shed light on why resources are distributed in certain ways today and how these imbalances and inequalities unfolded over time.

Trade is a way to indirectly measure a country’s economic output, but it also directly impacts people’s well-being, Xu explains. “A reduction in trade usually means people consume fewer types of goods and at higher prices,” she says, which means the 1866 crash harmed not just bankers and merchants but people around the world.As more archives are digitized, Xu thinks they will provide economists with a valuable source of big data. “Institutions like central banks and census bureaus are also taking an interest in at least scanning these sources and making them available online for researchers,” saving them from the task of sifting through old ledgers by hand.

It’s also an area Xu will continue to explore. She’s trying to incorporate historical textual data into her current work. “Newspapers and journals provide daily data on the actual information investors had and their views, which is actually a lot better than most quantitative financial data, which is usually monthly,” Xu says. “Systematically processing and analyzing that text can provide us with many new insights on our understanding of the international economy during this period.” She hopes to create a set of open-source tools that will lower the barriers to entry for other researchers who want to work with these documents.

Ultimately, Xu believes this historical lens helps scholars build stronger theories that explain the underlying relationships inherent in how economies function. “We’d like to be able to say that we’re explaining something that’s kind of fundamental, right? That we’re getting at something more like a natural law,” she says. “You don’t want your theory to only work in the post-World War II world.” Given the slow economic recovery from the 2008 crisis, Xu says, historical examples like the crisis of 1866 are more important than ever if economists want to predict when and how countries will bounce back.

A version of this article was  by ��ɫ���� Graduate School of Business.