An Interview with the Authors of Dark Matter Credit

Imagine a world without banks. Because there are no credit cards, you have to pay cash for everything, and there’s no way to borrow either. How do you buy car or a house, or start a new business? You hide cash under your mattress. Such a world would be desperately poor, or so research in economics teaches us. Yet someone Europe managed to become rich long before banks spread across the continent. How was that possible?

Dark Matter Credit by Philip T. Hoffman, Gilles Postel-Vinay, and Jean-Laurent Rosenthal solves the mystery. Using data on 250,000 loans from France, the authors found that credit abounded in Europe well before banks opened their doors, thanks to a huge shadow credit system whose importance no one has ever measured before. The system let nearly a third of French families borrow way back in 1740, and by 1840 it funded as much mortgage debt as the 1950s US banking system. And when banks finally appeared, it out-competed them, helping people to borrow, save, and even make payments. It thrived right up to World War I, not just in France but Britain, Germany, and the United States, only to be killed off by government intervention after 1918.

According to the authors, their discovery overturns standard arguments about banks and economic growth and reveals a shadow system made up of thousands of loans between individuals, as in modern peer to peer lending.  Dark Matter Credit sheds light on the problems peer to peer lending will face as it spreads and suggests how those problems can be solved.

What led you to uncover a huge and unknown shadow banking system?

We knew that people were borrowing and lending long before banks existed, because thousands of loan contracts survived in the French archives. We wanted to know how that was possible without banks. How did the lenders know that the borrowers would repay? After all, there was no such thing as a credit score or even an easy way to tell if property had been mortgaged, and potential lenders had for centuries been worried about the risk of default. Could lenders only make loans to family members or close friends? Was that how credit markets worked? If so, lending would have been severely limited.  Early investigations suggested, though, that lending was not so small, and not as local as previous scholars had thought. We suspected that informal intermediaries were matching borrowers and lenders and increasing the level of confidence in the market. To get at what had actually happened, we set out to measure all this lending across France and to analyze what made it possible.

How much lending was there?

Well in 1840, outstanding mortgage debt came to 27 percent of GDP. That was almost as much as in the United States during the housing boom in the 1950s, when there were numerous banks, savings and loans, and government backed mortgages, but all the lending in France was done without any bank involvement, and without any of the government support that stimulated housing construction in the United States. Even way back in 1740, the credit system in France allowed a third of all families to borrow and lend. And the system was incredibly persistent: it was only killed off by government intervention after 1918, but even as late as 1931, it was still providing 90 percent of all borrowers with their loans

How did it work?

The loans, it turns out, were arranged by notaries, who had been drawing up legal documents and preserving official copies of records since the Middle Ages. Over time, they began serving as real estate brokers and providing legal and financial advice, and since they knew who had money to lend and who was creditworthy, they were soon matching lenders up with borrowers who had good collateral and were likely repay. And if they couldn’t find a match among their own clients, they referred borrowers and lenders to one another. One notary might send a good borrower off to another notary, or he might receive a lender from yet another notary. That allowed loans to be made when the borrowers and lenders didn’t know one another. The loans didn’t pass through banks at all—they were all loans between individuals, as in modern, web based peer to peer lending, but all without the web obviously.

Did it do anything else?

The notaries also helped people make payments and manage their savings. And their loan business continued to thrive after banks opened their doors. There were in fact more banks in France than anyone imagined (we know—we counted them), but it took them nearly a century to make any serious inroads into mortgage lending. We also discovered that notaries and bankers actually cooperated with one another to devise a new way for peasants to pay their bills at a time when doing so was difficult outside of cities. This sort of innovation is surprising because it runs counter to an influential argument that financial markets should have been stifled by the legal system prevailing in France and many other parts of the world—so called civil law, which was supposedly less favorable to financial development than British and American common law. That argument is also contradicted by the fact that the notaries themselves were thriving loan brokers, because the notaries kept the written records that were at the heart of the civil law.

How did you measure all the lending?

We visited a lot of archives! We had to because we started in a period before there were any government statistics about lending. So we assembled loan information from original contracts and fiscal sources. Of course, reading a quarter of a million loan contracts would have been impossible, but we also knew that summaries of the loans survived in French tax archives from the early eighteenth century up through the 1900s. The tax records plus some ingenious sampling allowed us to gather the data on our quarter of a million loans and to estimate what was happening in the credit market for France as whole across two centuries. With the sample, we could analyze the impact of urbanization, economic growth, financial crises, and enormous institutional changes during the French Revolution and the nineteenth century.   We also investigated the spread of banking in France and the interaction between bankers and notaries, and we compared French banking with banking in Britain. The comparison suggested that Britain probably lacked as strong a peer to peer lending system as in France, although it did have one. Evidence from other countries implies that similar systems operated in Germany, and the United States in 1900. They too had big peer to peer lending systems that have yet to be explored. And one has recently cropped up in China, but it has caused massive losses and triggered protests, because of problems that the French system avoided.

Philip T. Hoffman is the Rea A. and Lela G. Axline Professor of Business Economics and History at the California Institute of Technology. Co-author Gilles Postel-Vinay is professor emeritus at the Paris School of Economics, and co-author Jean-Laurent Rosenthal is the Rea A. and Lela G. Axline Professor of Business Economics and the Ronald and Maxine Linde Leadership Chair in the Division of the Humanities and Social Sciences at the California Institute of Technology.