It's a wonderful bank

The fundamental promise of banking is that everyone can have their money back whenever they want it — as long as they don't all want it at the same time.

We learned that from George Bailey.

"You're thinking of this place all wrong, as if I had the money back in a safe,” he told the customers making a run on the Bailey Brothers Building & Loan. “The money's not here. Your money's in Joe's house, that's right next to yours. And in the Kennedy house, and Mrs. Macklin's house, and a hundred others.”

“You’re lending them the money to build and then they’re going to pay it back to you as best they can," he explained.

The crowd of worried customers was not immediately reassured. George had to offer $2,000 of his own money to fend off the run. Even then, the bank wasn’t truly saved until the very end of the movie, when friends and customers donated enough money to cover an $8,000 hole in the bank’s balance sheet.

Austrian economist Murray Rothbard would have told them to let Bailey Bros. fail. “Fractional-reserve banking is a shell game, a Ponzi scheme, a fraud,” he once wrote.

He believed banks were only able to perpetrate that fraud because bankers like George Bailey misrepresent how they were able to make so many loans.

“Because everybody's been inured to thinking that the banks simply borrow our money and re-lend it,” Rothbard explained in a lecture, “it's very difficult to make the mind shift and realize that the banks are really engaging in a form of legalized counterfeiting.”

In other words, if people understood how fractional-reserve banking really worked — that the banks create money “out of thin air” — everyone would ask for their money back at the same time. Even the best banks would fail.

This bleak view of banks seemed to receive academic support when economists Douglas Diamond and Philip Dybvig wrote a canonical study on the precariousness of fractional-reserve banking: Bank Runs, Deposit Insurance, and Liquidity.

The study formalized Rothbard’s intuition that banks that fund long-term loans with on-demand deposits are at risk of a debilitating run, even if the assets are sound. 

Fears of a bank failure can therefore be self-fulfilling: "During a bank run, depositors rush to withdraw their deposits because they expect the bank to fail,” the authors explain. “In fact, the sudden withdrawals can force the bank to liquidate many of its assets at a loss and to fail."

“It need not be anything fundamental about the bank's condition,” they added. Instead, “anything that causes [depositors] to anticipate a run will lead to a run.”

“Even 'healthy' banks can fail.”

Diamond and Dybvig reached that worrying conclusion primarily with a theoretical model grounded in math and game theory.

A new study suggests the model does not reflect reality.

In Bank Runs With and Without Bank Failure, economists Sergio Correia and Stephan Luck of the Federal Reserve, and Emil Verner of MIT, assembled a database of 3,984 bank runs that occurred in the United States between 1863 and 1934.

Each incident — scraped from newspaper accounts by a large language model — is documented on a website that details why the run started and how it was resolved.

The surprising finding is that most of the runs ran out of steam before threatening the bank: “There are more runs without bank failure than runs with failure,” the authors found.

This is not what you’d expect from the self-fulfilling Diamond-Dybvig model. 

Even among banks with “very weak fundamentals,” only 59% failed after experiencing a run.

Rothbard, I think, would have expected that number to be 100%.

Banks with the strongest fundamentals, meanwhile “seldom fail,” even when subject to a run.

The authors’ takeaway? “This pattern casts doubt on a strong form of the view that liquidity problems alone can trigger severe financial distress.”

That, I think, is their polite way of saying that Diamond, Dybvig, Rothbard, gold bugs, and Bitcoiners are wrong about fractional-reserve banking.

Stack ‘em high

Diamond and Dybvig were right about at least one thing: "A bank run in our model is caused by a shift in expectations,” they noted, “which could depend on almost anything.”

My random sampling of the Bank Runs database turned up some prime examples. 

In 1910, a run on the Merchants National Bank in Los Angeles, CA, began when a visit from the boxer Jim Jeffries drew a crowd of boxing fans to the bank. “Scores of depositors, thinking something was wrong, began cashing in,” a newspaper reported, “and not until the fighter retired did the frightened patrons become reassured.”

Jeffries, it turned out, was at the bank to open an account and deposit some of his championship winnings.

In 1924, a run on the Metals Bank & Trust Company of Butte, MT, got underway after someone overheard a joking bet that the bank would not open the next day. The bank stayed open for four hours after the regular closing time to satisfy withdrawals, a newspaper reported, “and only ceased payments to depositors when darkness made it unsafe.”

The joke was that the bank would indeed be closed the next day — for Lincoln’s birthday.

In 1929, a run on the Bay Ridge Savings Bank in Brooklyn, NY began amid rumors that its president had died. Fortunately, the bank had been “warned in advance of the false rumors,” a newspaper reported, giving the bank time to have $14 million of cash on hand to meet withdrawals.

The truth was that the bank president had gone to Connecticut to have a carbuncle removed from his neck. (He survived the procedure.)

Again, this was just a random sampling of the database.

But the peaceful resolution of each of these runs seems to refute the strongest interpretation of the Diamond-Dybvig theory: Bank runs, it turns out, are rarely self-fulfilling.

It did sometimes happen, though.

In 1930, a fistfight two doors down started a run on the Independence State Bank of Chicago. “Answering the call to the restaurant,” a newspaper reported, “the police patrol parked in front of the bank building and gave rise to a rumor that a run on the institution was in progress.” Somehow, more than $1.6 million of the bank’s $5.6 million of deposits was withdrawn — which must have exhausted the bank’s liquid assets because state officials felt compelled to close the bank.

The Bank Runs website doesn’t say whether the balance sheet of Independence Bank was fundamentally sound. But the authors’ study suggests that if it was, the bank almost certainly would have survived.

In many cases, all it took to weather a run was a public display of ready cash.

A run in 1907 was stopped “by a big display of bills and money on the counter in sight of the depositors,” for example. 

An “unprofitable run” on an Alabama bank in 1857 was cut short when depositors saw “a Malakoff of gold and a Redan of silver” piled high on a teller’s desk. (Malakoff and Redan being famous Russian fortresses.) 

And in 1924, the manager of a Brooklyn bank halted a run by piling bills in denominations as high as $1,000 in the bank’s front window — “heaped in careless profusion” — for all to see.

Did the bank’s customers understand that, however high the cash was piled, it could not have been enough to pay everyone if everyone wanted their money all at once?

I suspect they did.

— Byron Gilliam

_{Brought to you by:}

Avalanche Summit NYC returns September 16–17, bringing together the institutions, enterprises, investors, and builders turning blockchain technology into real business outcomes.

From tokenized markets and institutional finance to payments and consumer applications, the Summit will explore how production-ready infrastructure is enabling faster settlement, lower costs, and entirely new products and revenue streams.

Use promo code BLOCKWORKS15 for 15% off!