Bank runs historically not a 'death sentence' for banks: Fed study

• Key insight: Federal Reserve Bank of New York researchers found that banks with healthy fundamentals have historically survived bank runs in most cases.
• Supporting data: In a study of failures from 1863 and 1934, banks failed in only 38% of cases where a run occurred.
• Expert quote: "A run is not a death sentence," — New York Fed paper

A new study by New York Fed economists found that most bank runs do not result in the bank's failure; by contrast, the bank's underlying financial health, rather than depositor panic alone, is the strongest predictor of whether a run sinks a bank or spreads to other banks.

The report's findings complicate the long-held view that depositor panic alone can push otherwise healthy banks into failure, suggesting instead that, while runs can occur at both weak and strong institutions, they rarely prove fatal for well-capitalized banks.

"Poor fundamentals are necessary for runs to result in bank failures," said the paper, authored by Sergio Correia, Stephan Luck and Emil Verner. "Runs seem to only be associated with major economic distress when they take place in the context — and typically as a consequence — of already weak fundamentals."

The research comes as regulators continue to assess the lessons from bank failures in recent years and the role depositor runs play in financial crises. While only two banks have failed this year so far, the Federal Deposit Insurance Corp. is reworking its rules governing how the agency prepares for distress at a supervised firm, including by overhauling the agency's resolution planning rule for large banks. The findings suggest that while preventing liquidity runs are important, making sure banks remain fundamentally financially sound is ultimately the best defense against failure. 

The study draws on a newly compiled database of over 3,000 bank runs at U.S. banks between 1863 and 1934, as chronicled in newspapers at the time. The period studied predates  regulatory backstops like federal deposit insurance and the Federal Reserve's modern role as lender-of-last-resort. Using large language models, the researchers matched those publicly reported events with bank balance-sheet data and economic statistics to analyze how a bank's financial condition impacted its chances of survival.

While bank runs in popular imagination are often viewed as catastrophic events that must be contained lest they take down otherwise healthy institutions, the researchers found that banks failed in only about 38% of run episodes, meaning most survived.

"This is a high conditional probability of failure, given that the annual unconditional probability of bank failure is 0.85%," the researchers stated. "At the same time, it also implies that a run is not a death sentence."

What differentiated the survivors of runs from those that failed was their financial condition before the run began. Banks with high levels of debt and lower capital, weaker profitability and less stable funding were significantly more likely to experience runs in the first place and to fail in the face of one. Among banks in the weakest decile of the researchers' measure of financial health, roughly 63% failed after a run. Banks in the strongest decile, by contrast, hardly ever failed. 

"Weak bank fundamentals are necessary for a run to be associated with bank failure," they wrote. "Banks with strong fundamentals typically do not fail when subject to a run."

Researchers found that financially 'healthy' banks experienced depositor panics as well, though much less frequently. Unlike weaker peers, however, they were generally able to withstand the pressure without collapsing.

In examining accounts describing how banks responded during runs, they found banks that survived often continued meeting withdrawal requests while borrowing from other banks, raising additional capital or publicly demonstrating they had plenty of cash on hand. Clearing houses and bank supervisors also sometimes stepped in to provide support or assess a bank's condition. 

"To conserve liquidity and cool the run, banks would often partially suspend convertibility, invoking 30/60-day notice rules for savings deposits," they wrote. "In extreme runs, banks fully suspend convertibility."

FDIC

FDIC proposes slim resolution plans, lower insurance fees

June 25, 2026 3:50 PM

By Ebrima Santos Sanneh

The broader economic fallout of a bank run also varied substantially depending on whether the bank in question ultimately failed. Runs that ended in bank failures were associated with substantially larger declines in local deposits, lending and manufacturing activity over the following 18 months. Runs at banks that survived, however, had little measurable effect on local economic conditions.

Although the historical period studied predates modern safeguards, the researchers argue the evidence remains relevant for current policy debates. Maintaining strong bank fundamentals, they argue, is the most effective way to limit the damage bank runs can inflict.

"Our findings suggest that poor bank fundamentals are necessary for bank runs to translate into failure and, more broadly, for bank distress to generate severe economic consequences," they wrote. "Failures, in turn, are a key proxy for economic distress resulting from turmoil in the banking sector. Events that are primarily liquidity driven are less commonly associated with severe economic outcomes and should be less of a concern than solvency-driven crises."